DETAILED ACTION
Notice of Pre-AIA or AIA Status
The present application, filed on or after March 16, 2013, is being examined under the first inventor to file provisions of the AIA .
Claim Rejections - 35 USC § 101
35 U.S.C. 101 reads as follows:
Whoever invents or discovers any new and useful process, machine, manufacture, or composition of matter, or any new and useful improvement thereof, may obtain a patent therefor, subject to the conditions and requirements of this title.
Claims 1-27 are rejected under 35 U.S.C. 101 because the claimed invention is directed to an abstract idea (mental process or math concept) without significantly more.
Claim 1:
Regarding claim 1, in step 1 of the 101-analysis set forth in MPEP 2106, the claim recites “a method for building an application outage predictor, the method comprising: training a utilization model to output, from historical application-level network utilization data for a computer network, …”, and a method is one of the four statutory categories of invention.
In step 2A prong 1 of the 101-analysis set forth in the MPEP 2106, the examiner has determined that the following limitations recite a process that, under the broadest reasonable interpretation, covers a mental process but for recitation of generic computer components:
… combine the application health score … and the utilization prediction … into a combined health score for the application, (mental process, a person can mentally evaluate and combine two outcomes into a combined score using pen and paper, see MPEP 2106.04(a)(2)(III)),
… predict, from multimodal application health metric data, an application health score for the application, (mental process, a person can mentally evaluate and predict from multimodal or multiple sources of data, an application health score, see MPEP 2106.04(a)(2)(III)),
If claim limitations, under their broadest reasonable interpretation, covers performance of the limitations as a mental process but for the recitation of generic computer components, then it falls within the mental process grouping of abstract ideas. Accordingly, the claim “recites” an abstract idea.
In step 2A prong 2 of the 101-analysis set forth in MPEP 2106, the examiner has determined that the following additional elements do not integrate this judicial exception into a practical application:
A method for building an application outage predictor, the method comprising: training a utilization model to output, from historical application-level network utilization data for a computer network, a utilization prediction of future application-level network utilization for an application executing on the computer network; (Mere instructions to apply an exception using generic computer – see MPEP 2106.05(f)),
training an application health model …; (Mere instructions to apply an exception using generic computer – see MPEP 2106.05(f)),
wherein the multimodal application health metric data comprises a plurality of independent datasets each representing a status of the application within the computer network; (Mere instructions to apply an exception using generic computer – see MPEP 2106.05(f)),
wherein training of the utilization model is independent of training of the application health model; (Mere instructions to apply an exception using generic computer – see MPEP 2106.05(f)),
and providing a combiner adapted … from the application health model and … from the utilization model…(Mere instructions to apply an exception using generic computer – see MPEP 2106.05(f)),
Since the claim as a whole, looking at the additional elements individually and in combination, does not contain any other additional elements that are indicative of integration into a practical application, the claim is “directed” to an abstract idea.
In step 2B of the 101-analysis set forth in the 2019 PEG, the examiner has determined that the claim does not include additional elements that are sufficient to amount to significantly more than the judicial exception.
As discussed above, additional elements iii, iv, v, vi, and vii recite mere instructions to apply the judicial exception using generic computer components, which are not indicative of significantly more.
Considering the additional elements individually and in combination, and the claim as a whole, the additional elements do not provide significantly more than the abstract idea. Therefore, the claim is not patent eligible.
Claim 2:
Regarding claim 2, it is dependent upon claim 1, and thereby incorporates the limitations of, and corresponding analysis applied to claim 1. Further, claim 2 recites the following additional element:
The method of claim 1, further comprising providing for conformation of at least one of the application health score and the utilization prediction so that the application health score and the utilization prediction share a common format and are combinable with one another by the combiner, (In step 2A, prong 2, this is considered mere instructions to apply an exception using generic computer – see MPEP 2106.05(f)), (In step 2B, this is also considered mere instructions to apply an exception using generic computer – see MPEP 2106.05(f)),
Since the claim does not recite additional elements that either integrate the judicial exception into a practical application, nor provide significantly more than the judicial exception, the claim is not patent eligible.
Claim 3:
Regarding claim 3, it is dependent upon claim 1, and thereby incorporates the limitations of, and corresponding analysis applied to claim 1. Further, claim 3 recites the following additional element:
The method of claim 1, further comprising providing an evaluator adapted to apply a threshold test to the combined health score, wherein the evaluator is configured to initiate remedial action in response to the combined health score failing the threshold test, (In step 2A, prong 2, this is considered mere instructions to implement an exception using generic computer components – see MPEP 2106.05(f)). (In step 2B, this is also considered mere instructions to implement an exception using generic computer components – see MPEP 2106.05(f)).
Since the claim does not recite additional elements that either integrate the judicial exception into a practical application, nor provide significantly more than the judicial exception, the claim is not patent eligible.
Claim 4:
Regarding claim 4, it is dependent upon claim 3, and thereby incorporates the limitations of, and corresponding analysis applied to claim 3. Further, claim 4 recites the following additional element:
The method of claim 3, wherein the remedial action is at least one of an alarm or an automated correction procedure, (In step 2A, prong 2, this is considered mere instructions to apply an exception using generic computer – see MPEP 2106.05(f)), (In step 2B, this is also considered mere instructions to apply an exception using generic computer – see MPEP 2106.05(f)),
Since the claim does not recite additional elements that either integrate the judicial exception into a practical application, nor provide significantly more than the judicial exception, the claim is not patent eligible.
Claim 5:
Regarding claim 5, it is dependent upon claim 1, and thereby incorporates the limitations of, and corresponding analysis applied to claim 1. Further, claim 5 recites the following additional element:
The method of claim 1, wherein the multimodal application health metric data comprises at least Information Technology Service Management (ITSM) data, infrastructure metrics, and outage information for the computer network, (In step 2A, prong 2, this recites an indication to a field of use or technological environment – see MPEP 2106.05(h)), (In step 2B, this also recites a field of use or technological environment – see MPEP 2106.05(h)),
Since the claim does not recite additional elements that either integrate the judicial exception into a practical application, nor provide significantly more than the judicial exception, the claim is not patent eligible.
Claim 6:
Regarding claim 6, it is dependent upon claim 5, and thereby incorporates the limitations of, and corresponding analysis applied to claim 5. Further, claim 6 recites the following additional element:
The method of claim 5, wherein the outage information comprises volumetric problem report data from at least one external public Internet platform that is outside of the computer network and that is nonspecific to the application, (In step 2A, prong 2, this recites an indication to a field of use or technological environment – see MPEP 2106.05(h)), (In step 2B, this also recites a field of use or technological environment – see MPEP 2106.05(h)),
Since the claim does not recite additional elements that either integrate the judicial exception into a practical application, nor provide significantly more than the judicial exception, the claim is not patent eligible.
Claim 7:
Regarding claim 7, it is dependent upon claim 1, and thereby incorporates the limitations of, and corresponding analysis applied to claim 1. Further, claim 7 recites the following additional element:
The method of claim 1, wherein the utilization model is a Long Short Term Memory (LSTM) neural network model, (In step 2A, prong 2, this is considered mere instructions to apply an exception using generic computer – see MPEP 2106.05(f)), (In step 2B, this is also considered mere instructions to apply an exception using generic computer – see MPEP 2106.05(f)),
Since the claim does not recite additional elements that either integrate the judicial exception into a practical application, nor provide significantly more than the judicial exception, the claim is not patent eligible.
Claim 8:
Regarding claim 8, it is dependent upon claim 1, and thereby incorporates the limitations of, and corresponding analysis applied to claim 1. Further, claim 8 recites the following abstract idea:
i. The method of claim 1, wherein the application health score indicates a probability of failure of the application, (This is considered a mathematical relationship, mathematical formula or equation, or mathematical calculation, see … in paragraph [0036] from the specification state “The combined health score 318 preferably indicates a probability of failure of the application. For example, the combined health score may be a percentage probability of failure, or the inverse, e.g. probability of failure = 100% minus combined health score (represented as a percentage). Alternatively, a scaled score could be used, such as a scale of 1 to 5, or 1 to 10.”, see MPEP 2106.04(a)(2), subsection I),
If claim limitations, under their broadest reasonable interpretation, covers performance of the limitations as a mathematical concept but for the recitation of generic computer components, then it falls within the mathematical concept grouping of abstract ideas. Accordingly, the claim “recites” an abstract idea.
Since the claim does not recite additional elements that either integrate the judicial exception into a practical application, nor provide significantly more than the judicial exception, the claim is not patent eligible.
Claim 9:
Regarding claim 9, it is dependent upon claim 1, and thereby incorporates the limitations of, and corresponding analysis applied to claim 1. Further, claim 9 recites the following additional element:
A computer program product comprising at least one tangible, non-transitory computer-readable medium embodying instructions which, when executed by at least one processor, cause the at least one processor to implement a method according to claim 1, (In step 2A, prong 2, this recites a generic computer component being used as a tool, and mere instructions to implement an abstract idea using generic computer – see MPEP 2106.05(f)). (In step 2B, this also recites a generic computer component being used as a tool, and mere instructions to implement an abstract idea using generic computer – see MPEP 2106.05(f)).
Since the claim does not recite additional elements that either integrate the judicial exception into a practical application, nor provide significantly more than the judicial exception, the claim is not patent eligible.
Claim 10:
Regarding claim 10, it is dependent upon claim 1, and thereby incorporates the limitations of, and corresponding analysis applied to claim 1. Further, claim 10 recites the following additional element:
A data processing system comprising at least one processor and memory coupled to the at least one processor wherein the memory stores instructions which, when executed by the at least one processor, cause the data processing system to implement a method according to claim 1. (In step 2A, prong 2, this recites a generic computer component being used as a tool, and mere instructions to implement an abstract idea using generic computer – see MPEP 2106.05(f)). (In step 2B, this also recites a generic computer component being used as a tool, and mere instructions to implement an abstract idea using generic computer – see MPEP 2106.05(f)).
Since the claim does not recite additional elements that either integrate the judicial exception into a practical application, nor provide significantly more than the judicial exception, the claim is not patent eligible.
Claim 11:
Regarding claim 11, in step 1 of the 101-analysis set forth in MPEP 2106, the claim recites “a method for predicting an application outage on a computer network, the method comprising: receiving, from a trained utilization model, a utilization prediction of future application-level network utilization for an application executing on the computer network; receiving, from a trained application health model trained independently of the neural network model, an application health score for the application; combining the application health score from the application health model and the utilization prediction from the neural network model into a combined health score for the application”, and a method is one of the four statutory categories of invention.
In step 2A prong 1 of the 101-analysis set forth in the MPEP 2106, the examiner has determined that the following limitations recite a process that, under the broadest reasonable interpretation, covers a mental process but for recitation of generic computer components:
combining the application health score… and the utilization prediction into a combined health score for the application (mental process, a person can mentally evaluate and combine two outcomes into a combined score, see MPEP 2106.04(a)(2)(III))
If claim limitations, under their broadest reasonable interpretation, covers performance of the limitations as a mental process but for the recitation of generic computer components, then it falls within the mental process grouping of abstract ideas. Accordingly, the claim “recites” an abstract idea.
In step 2A prong 2 of the 101-analysis set forth in MPEP 2106, the examiner has determined that the following additional elements do not integrate this judicial exception into a practical application:
… from the application health model and … from the neural network model, (Mere instructions to apply an exception using generic computer – see MPEP 2106.05(f)),
A method for predicting an application outage on a computer network, the method comprising: receiving, from a trained utilization model, a utilization prediction of future application-level network utilization for an application executing on the computer network; (In step 2A, prong 2, this recites mere data gathering, which is considered an insignificant extra-solution activity – see MPEP 2106.05(g)),
receiving, from a trained application health model trained independently of the neural network model, an application health score for the application; (In step 2A, prong 2, this recites mere data gathering, which is considered an insignificant extra-solution activity – see MPEP 2106.05(g)),
Since the claim as a whole, looking at the additional elements individually and in combination, does not contain any other additional elements that are indicative of integration into a practical application, the claim is “directed” to an abstract idea.
In step 2B of the 101-analysis set forth in the 2019 PEG, the examiner has determined that the claim does not include additional elements that are sufficient to amount to significantly more than the judicial exception.
As discussed above, additional element ii recites mere instructions to apply the judicial exception using generic computer components, which is not indicative of significantly more. The additional elements iii and iv recite mere data gathering, and are considered insignificant extra-solution activities. In step 2B, these insignificant extra-solution activities are well understood routine and conventional activities, which include receiving or transmitting data over a network from court case Symantec, 838 F.3d at 1321, 120 USPQ2d at 1362 (utilizing an intermediary computer to forward information); TLI Communications LLC v. AV Auto. LLC, 823 F.3d 607, 610, 118 USPQ2d 1744, 1745 (Fed. Cir. 2016), – see MPEP 2106.05(d) (II)(i)),
Considering the additional elements individually and in combination, and the claim as a whole, the additional elements do not provide significantly more than the abstract idea. Therefore, the claim is not patent eligible.
Claim 12:
Regarding claim 12, it is dependent upon claim 11, and thereby incorporates the limitations of, and corresponding analysis applied to claim 11. Further, claim 12 recites the following additional element:
The method of claim 11, further comprising, before combining the application health score and the utilization prediction, conforming at least one of the application health score and the utilization prediction so that the application health score and the utilization prediction share a common format and are combinable with one another by the combiner, (In step 2A, prong 2, this is considered mere instructions to implement an abstract idea using generic computer – see MPEP 2106.05(f)). (In step 2B, this is also considered mere instructions to implement an abstract idea using generic computer – see MPEP 2106.05(f)).
Since the claim does not recite additional elements that either integrate the judicial exception into a practical application, nor provide significantly more than the judicial exception, the claim is not patent eligible.
Claim 13:
Regarding claim 13, it is dependent upon claim 11, and thereby incorporates the limitations of, and corresponding analysis applied to claim 11. Further, claim 13 recites the following additional element:
The method of claim 11, wherein the application health model and the utilization model are configured so that the application health score and the utilization prediction share a common format, (In step 2A, prong 2, this is considered mere instructions to implement an abstract idea using generic computer – see MPEP 2106.05(f)). (In step 2B, this is also considered mere instructions to implement an abstract idea using generic computer – see MPEP 2106.05(f)).
Since the claim does not recite additional elements that either integrate the judicial exception into a practical application, nor provide significantly more than the judicial exception, the claim is not patent eligible.
Claim 16:
Regarding claim 16, it is dependent upon claim 11, and thereby incorporates the limitations of, and corresponding analysis applied to claim 11. Further, claim 16 recites the following abstract idea:
predicts from historical application-level network utilization data for the computer network, (This recites a mental process, a person can mentally evaluate and predict from historical application-level network utilization data, see MPEP 2106.04(a)(2)(III)),
Further, claim 16 recites the following additional element:
The method of claim 11, wherein the utilization model …, (In step 2A, prong 2, this is considered mere instructions to implement an abstract idea using generic computer – see MPEP 2106.05(f)). (In step 2B, this is also considered mere instructions to implement an abstract idea using generic computer – see MPEP 2106.05(f)).
Since the claim does not recite additional elements that either integrate the judicial exception into a practical application, nor provide significantly more than the judicial exception, the claim is not patent eligible.
Claim 17:
Regarding claim 17, it is dependent upon claim 11, and thereby incorporates the limitations of, and corresponding analysis applied to claim 11. Further, claim 17 recites the following abstract idea:
…predicts from multimodal application health metric data, (This recites a mental process, a person can mentally evaluate and predict from multimodal application health metric data, see MPEP 2106.04(a)(2)(III)),
Further, claim 17 recites the following additional element:
The method of claim 11, wherein the application health model …, (In step 2A, prong 2, this is considered mere instructions to implement an abstract idea using generic computer – see MPEP 2106.05(f)). (In step 2B, this is also considered mere instructions to implement an abstract idea using generic computer – see MPEP 2106.05(f)).
Since the claim does not recite additional elements that either integrate the judicial exception into a practical application, nor provide significantly more than the judicial exception, the claim is not patent eligible.
Claims 14-15, 18, 19-22:
Regarding claims 14-15, 18, 19-22, all of claim 11’s dependent claims follow the deficiencies of their parent claim. Since claims 14-15, 18, 19-22 recite similar limitations as corresponding claims 3-4, 5, 7-10, respectively listed above, and are rejected for similar reasons under 35 U.S.C. 101.
Claim 23:
Regarding claim 23, in step 1 of the 101-analysis set forth in MPEP 2106, the claim recites “a method for predicting an application outage on a computer network, the method comprising:
combining a utilization prediction of future application-level network utilization for an application executing on the computer network with an application health score for the application derived from multiple sources into a combined health score for the application,” and a method is one of the four statutory categories of invention.
In step 2A prong 1 of the 101-analysis set forth in the MPEP 2106, the examiner has determined that the following limitations recite a process that, under the broadest reasonable interpretation, covers a mental process but for recitation of generic computer components:
A method for predicting an application outage on a computer network, the method comprising: combining a utilization prediction of future application-level network utilization … with an application health score for the application derived from multiple sources into a combined health score for the application, (mental process, a person can mentally evaluate and combine a utilization prediction and an application health score from multiple sources into ta combined health score , see MPEP 2106.04(a)(2)(III)),
If claim limitations, under their broadest reasonable interpretation, covers performance of the limitations as a mental process but for the recitation of generic computer components, then it falls within the mental process grouping of abstract ideas. Accordingly, the claim “recites” an abstract idea.
In step 2A prong 2 of the 101-analysis set forth in MPEP 2106, the examiner has determined that the following additional elements do not integrate this judicial exception into a practical application:
…for an application executing on the computer network…(Mere instructions to apply an exception using generic computer – see MPEP 2106.05(f)),
Since the claim as a whole, looking at the additional elements individually and in combination, does not contain any other additional elements that are indicative of integration into a practical application, the claim is “directed” to an abstract idea.
In step 2B of the 101-analysis set forth in the 2019 PEG, the examiner has determined that the claim does not include additional elements that are sufficient to amount to significantly more than the judicial exception.
As discussed above, additional element ii recites mere instructions to apply the judicial exception using generic computer components, which is not indicative of significantly more.
Considering the additional elements individually and in combination, and the claim as a whole, the additional elements do not provide significantly more than the abstract idea. Therefore, the claim is not patent eligible.
Claims 24-25, 26-27:
Regarding claims 24-25, 26-27, all of claim 23’s dependent claims follow the deficiencies of their parent claim. Since claims 24-25, 26-27 recite similar limitations as corresponding claims 3-4, 9-10, respectively listed above, and are rejected for similar reasons under 35 U.S.C. 101.
Claim Rejections - 35 USC § 103
In the event the determination of the status of the application as subject to AIA 35 U.S.C. 102 and 103 (or as subject to pre-AIA 35 U.S.C. 102 and 103) is incorrect, any correction of the statutory basis (i.e., changing from AIA to pre-AIA ) for the rejection will not be considered a new ground of rejection if the prior art relied upon, and the rationale supporting the rejection, would be the same under either status.
The following is a quotation of 35 U.S.C. 103 which forms the basis for all obviousness rejections set forth in this Office action:
A patent for a claimed invention may not be obtained, notwithstanding that the claimed invention is not identically disclosed as set forth in section 102, if the differences between the claimed invention and the prior art are such that the claimed invention as a whole would have been obvious before the effective filing date of the claimed invention to a person having ordinary skill in the art to which the claimed invention pertains. Patentability shall not be negated by the manner in which the invention was made.
The factual inquiries for establishing a background for determining obviousness under 35 U.S.C. 103 are summarized as follows:
1. Determining the scope and contents of the prior art.
2. Ascertaining the differences between the prior art and the claims at issue.
3. Resolving the level of ordinary skill in the pertinent art.
4. Considering objective evidence present in the application indicating obviousness or nonobviousness.
Claims 1, 3, 4, 5, 8, 9, 10, 11, 14, 15, 16, 17, 18, 20, 21, and 22 are rejected under 35 U.S.C. 103 over Vijayakumar, M. et al., (US PG Pub. US20190044825A), published on February 7, 2019, (hereafter, VIJAYAKUMAR) in view of White, G. et al., in “Short-term QoS forecasting at the edge for reliable service applications,” published on February 24, 2020, available at https://ieeexplore.ieee.org/stamp/stamp.jsp?tp=&arnumber=9007504, (hereafter, WHITE), further in view of Yin, Y. et al., in “Collaborative service selection via ensemble learning in mixed mobile network environments,” published on July 20, 2017, available at https://www.mdpi.com/1099-4300/19/7/358 , (hereafter, YIN).
Claim 1:
Regarding claim 1, VIJAYAKUMAR teaches “A method for building an application outage predictor, the method comprising: training a utilization model to output, from historical application-level network utilization data for a computer network, a utilization prediction of future application-level network utilization for an application executing on the computer network;”
See VIJAYAKUMAR in [0042] describe “The training module 407 is configured to train a prediction model based on the historical data using a machine learning engine. The historical data is used as a training dataset, which is provided as an input to the training model. In some embodiments, the data classifier module 406 may be configured to categorize the data as historical if it has been used in a training dataset. Further, the prediction module 408 is configured to feed the current data to the prediction model for predicting status score, utilization score, and performance score of each of the plurality of nodes.” Here, VIJAYAKUMAR mentions training a model that can predict performance, utilization or status using historical data.
Further, see VIJAYAKUMAR in [0029] mention “A method and a system for determining and preventing outages in a network is disclosed. The method and system allows for predicting one or more of status, utilization, or performance of IT resources. The present subject matter includes extraction and classification of metrics for one or more parameters associated with a plurality of nodes. A set of historical metrics and real-time metrics are used for predicting status score, utilization score, and performance score of IT infrastructure resources.” See VIJAYAKUMAR also in [0036] describe “According to an embodiment, a network architecture of the IT environment 300 comprising a system 301 for determining application health is illustrated in FIG. 3. The architecture may include the system 301, one or more local networks 303-1-303-N comprising one or more servers 304, one or more end user devices 305, all of which may be connected over the network 302. The one or more servers 304 render essential services required in IT environments. In some examples, the servers may include web servers for delivering content or services to end users through the network; application servers to facilitate installation, operation, hosting of applications; database servers to run database applications”. The examiner construes application to mean any program or service that requires a computer network (like the internet or a local area network) to operate. Here, VIJAYAKUMAR mentions using data (both historical and real-time) that relates to network, use of application servers or host applications, which relate to application operation status.
Further, see VIJAYAKUMAR in [0047] describe “The above subject matter and its embodiments provide method and system to determine and prevent potential outages. The present subject matter predicts utilization and performance of infrastructure resources, which helps in enhancing the overall performance of the network. The predictions also enable the operations team to optimize processes, anticipate suspicious trends before loss occurs, gain insights into the causes and relationships of downtime with performance. Further, the invention helps to anticipate performance issues in future and enhance overall productivity, revenue, and security of enterprises.” Here in [0047], VIJAYAKUMAR mentions that this invention including models or algorithms that accompany the invention helps to also predict future performance usage including performance of applications implemented on network servers mentioned in [0036] (i.e. relates to prediction of future application-level network).
Further, VIJAYAKUMAR mentions “training an application health model to predict, from multimodal application health metric data, an application health score for the application;”
See VIJAYAKUMAR in [0042] describe “The training module 407 is configured to train a prediction model based on the historical data using a machine learning engine. The historical data is used as a training dataset, which is provided as an input to the training model... Further, the prediction module 408 is configured to feed the current data to the prediction model for predicting status score, utilization score, and performance score of each of the plurality of nodes.” Here, VIJAYAKUMAR mentions the process of training a model that can predict performance, utilization or status score.
Also, see VIJAYAKUMAR in [0028] describe “ ‘Nodes’ refer to a device or system in the network that can receive, create, store or send data along distributed network routes. Nodes may include web servers, application servers, database servers, laptops, computers, mobile devices, smart devices, etc.” Here, nodes relate to application servers, which are part of applications within a network. From [0042], VIJAYAKUMAR mentions that the prediction of status, utilization and performance scores relate to each of the nodes, which also includes applications run in servers defined in [0028].
Further, see VIJAYAKUMAR in [0036] describe “According to an embodiment, a network architecture of the IT environment 300 comprising a system 301 for determining application health is illustrated in FIG. 3. The architecture may include the system 301, one or more local networks 303-1-303-N comprising one or more servers 304, one or more end user devices 305, all of which may be connected over the network 302. The one or more servers 304 render essential services required in IT environments. In some examples, the servers may include web servers for delivering content or services to end users through the network; application servers to facilitate installation, operation, hosting of applications; database servers to run database applications” Here, VIJAYAKUMAR mentions using data that relates to a network, use of application servers or host applications, which relate to application operation status or application health.
See VIJAYAKUMAR in [0008] describe “The method includes extracting, from one or more data sources, data for one or more parameters associated with a plurality of nodes in the IT environment. The data comprising at least utilization metrics, performance metrics, and a time identifier for each of the performance and utilization metric. The extracted data is classified as historical data or a current data based on the time identifier. A status score, utilization score, and performance score, of the plurality of nodes, or a combination thereof for the plurality of nodes from the classified data are predicted.” VIJAYAKUMAR mentions taking one or more data sources that has performance and utilization metrics (i.e. multimodal application health metric data), to predict a score that relates to an application health score (from [0036]). From [0042], VIJAYAKUMAR mentions that the prediction of status, utilization and performance scores relate to each of the nodes, which also includes applications run in servers defined in [0028].
Further, VIJAYAKUMAR mentions “wherein the multimodal application health metric data comprises a plurality of independent datasets each representing a status of the application within the computer network;”
See VIJAYAKUMAR in [0008] describe “The method includes extracting, from one or more data sources, data for one or more parameters associated with a plurality of nodes in the IT environment. The data comprising at least utilization metrics, performance metrics, and a time identifier for each of the performance and utilization metric. The extracted data is classified as historical data or a current data based on the time identifier. A status score, utilization score, and performance score, of the plurality of nodes, or a combination thereof for the plurality of nodes from the classified data are predicted.” VIJAYAKUMAR mentions taking one or more data sources (which relates to a plurality of independent datasets), and each of the datasets has performance metrics, utilization metrics and a time identifier correspond to a status of the computer application.
Also, see VIJAYAKUMAR in [0028] describe “ ‘Nodes’ refer to a device or system in the network that can receive, create, store or send data along distributed network routes. Nodes may include web servers, application servers, database servers, laptops, computers, mobile devices, smart devices, etc.” Here, nodes relate to application servers, which are part of applications within a network. From [0008], VIJAYAKUMAR mentions that the prediction of status, utilization and performance scores relate to each of the nodes, which also includes applications that are run in servers defined in [0028].
However, VIJAYAKUMAR did not teach “wherein training of the utilization model is independent of training of the application health model; and providing a combiner adapted to combine the application health score from the application health model and the utilization prediction from the utilization model into a combined health score for the application.”
In an analogous art, WHITE teaches “wherein training of the utilization model is independent of training of the application health model;”
See WHITE in page 1095, second full paragraph, section 5.2 Metrics, mention “For each of the models, we record the training and prediction time on each of the datasets. The training time is the average time taken to train the model on the training dataset and the prediction time is the average time to generate a forecast given the test data.” Further, see WHITE in page 1089, Introduction section in paragraph 2, describe “These equivalent services can be used during a service adaptation that is responding to forecast QoS changes (e.g., unacceptable response time). To achieve this, knowledge about QoS values of the services is required to make timely and accurate adaptation decisions such as: when to trigger an adaptation action; which executing service(s) to replace and what candidate service(s) to choose.” Later, see WHITE in page 1089 abstract mention “Accurate short-term forecasts allow dynamic systems to adapt their behaviour when degradation is forecast e.g., transportation forecasting allows for alternative routing of traffic before gridlock… can be applied to service-oriented computing when creating and managing service applications. Recent approaches to improve reliability in service applications … to make QoS predictions for similar users.” Here, WHITE talks about using two models and train them separately. Then, WHITE describes using these models to forecast QoS or quality of service changes to service applications. See WHITE in page 1091, paragraph 3, section 3 for details.
It would have been obvious for one of ordinary skill in the art before the effective filing date of the claimed invention to combine the base reference of VIJAYAKUMAR and incorporate into the teachings of WHITE because both references teach the process of training models from historical or multimodal data sources to predict network application outages.
One of ordinary skill in the art would be motivated to do so because “The results show the improved forecasting accuracy using the noisy echo forecasting approach for 80 percent of the test datasets. The two datasets that get better predictions from traditional time series-based,” (see WHITE in page 1096, section 6.1, Prediction accuracy).
However, VIJAYAKUMAR in view of WHITE did not teach “and providing a combiner adapted to combine the application health score from the application health model and the utilization prediction from the utilization model into a combined health score for the application.”
In an analogous method, YIN teaches “and providing a combiner adapted to combine the application health score from the application health model and the utilization prediction from the utilization model into a combined health score for the application.”
See YIN in page 9, section 3.3, Proposed Prediction methods, paragraph 3, describe “To further improve prediction accuracy, we propose a hybrid prediction method that combines the prediction results of UCF-E and SCF-E, aiming to fully take advantage of the whole QoS data. We name this method Hybrid CF with Ensemble learning (HCF-E), given as follows:
PNG
media_image1.png
76
1149
media_image1.png
Greyscale
where the parameter θ is used to control the proportions of the two individual models. UCF-Eu,j and SCF-Eu,j are the prediction values of the two individual models respectively.” Here, YIN shows using a method to combine two models UCF-E and SCF-E together to get one combined model. The HCF is considered a combined prediction value (i.e. combined health score) for the application. Also, see first few paragraphs on page 9, section 3.3, where YIN notes “Since the ensemble learning model takes an important role in our framework, we name the proposed method User-based CF with Ensemble learning (UCF-E), and the corresponded prediction is given by,
PNG
media_image2.png
1
753
media_image2.png
Greyscale
where wv,j is the probability of qv,j belonging to the label and qv,j is the real value that user v received after invoking service j.
In a similar way, we propose a new service-based CF method using the probability belonging to the labels of the missing QoS values, also replacing the traditional similarity. This proposed method is named Service-based CF with Ensemble learning (SCF-E), and the prediction is given as,
PNG
media_image2.png
1
753
media_image2.png
Greyscale
where wu,h is the probability of qu,h belonging to the label. qu,h is the real value after user u invoked service h.”
See YIN in page 13, conclusion, describe “we also plan to construct a real-world service selection system for mixed mobile networks.” Here, YIN refers the system to be a combiner to combine the results for two models. The prediction value is measured with accuracy, see Tables 4 and 5 in YIN for details.
Further, see YIN in page 11, section 4.4. Sensitivity Analysis of θ mention “The parameter θ is used to control the weight of the two individual models (UCF-E and SCF-E) in the combination model.” YIN mentions combining the models also includes combining their weights. See YIN in abstract in page 1 for more details.
Further, see YIN in page 1, Introduction mention “developing mobile services has become an increasingly important way for various enterprises to deliver their marketing applications that only customs’ functional demands and satisfy their expected non-functional requirements.” YIN relates the method to user applications such as marketing applications.
See also page 2 in YIN from Introduction section mention “There has been much research into QoS prediction [10,11,12]. For example, collaborative filtering (CF) uses historical QoS records to predict unknown QoS values.” Since quality of service (QoS) predictions and network application health scores achieve the same goal of achieving optimal user experience for network application performance, both the two models UCF-E and SCF-E relate to QoS predicted values and thus network application health. While QoS prediction anticipates network performance, application health scores quantify how well the application performs under those conditions, these two scores ultimately serve a similar purpose. Since both are analogous topics, YIN page 9 mentions that the UCF-E model, being user-based, measures the predictions on users with similar usage patterns (i.e. relates to prediction from utilization model) while the SCF-E model measures QoS predictions (i.e. relates to application health model) that are found on similar services with one of the predicted values is qu,h , which is the real value after user u invoked service h. This shows combine the application health score from the application health model and the utilization prediction from the utilization model into a combined health score, where the examiner construes score to be any predicted result, outcome, or calculated value from a model.
It would have been obvious to one of ordinary skill in the art before the effective filing date of the claimed invention to combine the reference of VIJAYAKUMAR along with the reference of WHITE with the teachings of YIN by using the teachings of VIJAYAKUMAR and WHITE of using historical and multimodal data to train utilization model and application health model respectively, with YIN’s teaching of combining the two models.
One of ordinary skill in the art would be motivated to do so because by integrating YIN’s framework into the methods of VIJAYAKUMAR and WHITE, one with ordinary skill in the art would achieve the goal of providing an “approach … oriented to a mixed mobile network environment. Our approach has the capability to improve similarity between two users/services. Moreover, it can also handle abnormal QoS data” (see YIN in page 2).
Claim 3:
Regarding claim 3, VIJAYAKUMAR in view of WHITE, and further in view of YIN, teach the limitations in claim 1.
Further, VIJAYAKUMAR teaches “The method of claim 1, further comprising providing an evaluator adapted to apply a threshold test to the combined health score, wherein the evaluator is configured to initiate remedial action in response to the combined health score failing the threshold test.”
See VIJAYAKUMAR in [0010] describe “Based on the potential outages the alert module sends alerts to one or more devices if the predicted scores exceed a threshold limit or potential outage is identified. The summary generation module is configured to generate a summary comprising at least trends and statistics associated with the predicted status, utilization, and performance, of the plurality of nodes. The data cleansing module is configured to detect inaccurate data in the data store and perform a corrective action on the data.” Here, VIJAYAKUMAR mentions if a predicted score exceeds a threshold limit, there is a corrective action or alert sent to the devices.
See VIJAYAKUMAR in [0032] mention “A status score, utilization score, performance score, or a combination thereof are predicted based on the historical data and the current data, at block 103 …The status may indicate a failure status of the nodes, supplemented with information including, but not limited to, root cause analysis, time, reason, severity, probability of potential failure of the node, etc. A potential outage is determined in the IT environment from the predicted scores at block 104. The determination may include comparison of the predicted score and a predetermined threshold limit set by an administrator.” VIJAYAKUMAR shows if the predicted score exceeds the threshold limit on the score (that score is a combination of status, utilization and performance), then there is a potential outage. See VIJAYAKUMAR in [0043] for more details.
Claim 4:
Regarding claim 4, VIJAYAKUMAR in view of WHITE, and further in view of YIN, teach the limitations in claim 3.
Further, VIJAYAKUMAR teaches “The method of claim 3, wherein the remedial action is at least one of an alarm or an automated correction procedure.”
See VIJAYAKUMAR in [0043] describe “The alert module 410 is configured to send alerts to one or more devices if the predicted scores or potential outage exceed a threshold limit. The predicted scores may be compared with a predetermined threshold limit to identify potential outage in the IT environment.” Further, see VIJAYAKUMAR in [0044] mention “Further, automated workflows may be triggered to kill processes, which are consuming more utilization, based on the utilization anomaly predicted for a CPU, memory, etc. In scenarios where one or more servers may be on the verge of shutting down due to high demand during peak time, the system may automatically identify the processes that may be insignificant or may require substantial resources from the servers, ” Here, VIJAYAKUMAR describes sending alerts to devices if an outage is detected, and an automated workflow triggered to stop processes is initiated as another corrective action (i.e. remedial action is an alarm).
Claim 5:
Regarding claim 5, VIJAYAKUMAR in view of WHITE, and further in view of YIN, teach the limitations in claim 1.
Further, VIJAYAKUMAR teaches “The method of claim 1, wherein the multimodal application health metric data comprises at least Information Technology Service Management (ITSM) data, infrastructure metrics, and outage information for the computer network.”
See VIJAYAKUMAR in [0030] describe “In some embodiments, a method 100 for predicting status, utilization, performance, or a combination thereof for network nodes in an IT environment is provided as illustrated in FIG. 1. The method 100 includes extracting data or metrics related to one or more parameters associated with network nodes from one or more data sources… The data sources may include service desk tools or service monitoring tools deployed in each of the nodes in the network. The nodes may primarily include servers, such as web server, application server, database server, and user devices.” Here, VIJAYAKUMAR shows the information technology service management (ITSM) data sources include information from service monitoring or desk tools.
Further, see VIJAYAKUMAR in [0023] describe “ ‘Outage’ refers to a period when one or more systems in a network fail to perform their primary functions and operations. Outages may occur due to several reasons, such as unplanned events, exceptional events, network errors, anomalies, routine maintenance, etc. The term “outage” may be collectively referred to as downtime or network outage.” VIJAYAKUMAR shows instances when outages can occur or outage information for the computer network.
Further, see VIJAYAKUMAR in [0025] describe “Performance is usually estimated based on the metrics of the infrastructure resources, such as CPU, network, etc.” VIJAYAKUMAR shows performance is evaluated from infrastructure information such as from CPU.
Claim 8:
Regarding claim 8, VIJAYAKUMAR in view of WHITE, and further in view of YIN, teach the limitations in claim 1.
Further, VIJAYAKUMAR teaches “The method of claim 1, wherein the application health score indicates a probability of failure of the application.”
See VIJAYAKUMAR in [0027] describe “ ‘Status’ refers to the condition of a node in the network. Status may indicate the probability of failure at the node, degradation of the functions of the node, etc.” Later, VIJAYAKUMAR in [0039] mention “The network interface enables the system to communicate with one or more nodes in the IT environment. The one or more nodes may be end user devices 305, one or more servers 306, one or more servers in the local networks.” Here, VIJAYAKUMAR shows nodes can relate to devices or servers in a network.
See VIJAYAKUMAR in [0042] mention “the prediction module 408 is configured to feed the current data to the prediction model for predicting status score, utilization score, and performance score of each of the plurality of nodes.” Here, the status score also represents an application health score since the status shows probability of failure at a node or device.
Also, see VIJAYAKUMAR in [0028] describe “ ‘Nodes’ refer to a device or system in the network that can receive, create, store or send data along distributed network routes. Nodes may include web servers, application servers, database servers, laptops, computers, mobile devices, smart devices, etc.” Here, nodes relate to application servers, which are part of applications that run within a network. From [0042], VIJAYAKUMAR notes that the prediction of status, utilization and performance scores relate to the nodes, which also includes applications run in servers defined in [0028].
Further, see VIJAYAKUMAR in [0036] describe “According to an embodiment, a network architecture of the IT environment 300 comprising a system 301 for determining application health is illustrated in FIG. 3. The architecture may include the system 301, one or more local networks 303-1-303-N comprising one or more servers 304, one or more end user devices 305, all of which may be connected over the network 302. The one or more servers 304 render essential services required in IT environments. In some examples, the servers may include web servers for delivering content or services to end users through the network; application servers to facilitate installation, operation, hosting of applications; database servers to run database applications”. The examiner construes application to mean any program or service that requires a computer network (like the internet or a local area network) to operate. Here, VIJAYAKUMAR mentions using data that relates to a network, use of application servers or host applications, which relate to application operation status or application health.
See VIJAYAKUMAR in [0032] mention “A status score, utilization score, performance score, or a combination thereof are predicted based on the historical data and the current data, at block 103 …The status may indicate a failure status of the nodes, supplemented with information including, but not limited to, root cause analysis, time, reason, severity, probability of potential failure of the node, etc. A potential outage is determined in the IT environment from the predicted scores at block 104. The determination may include comparison of the predicted score and a predetermined threshold limit set by an administrator.” VIJAYAKUMAR shows that the status indicates a failure status (i.e. application health score indicates a probability of failure of the application). VIJAYAKUMAR also shows if the predicted score exceeds the threshold limit on the score (that score is a combination of status, utilization and performance), then there is a potential outage. See VIJAYAKUMAR in [0043] for more details.
Claim 9:
Regarding claim 9, VIJAYAKUMAR in view of WHITE, further in view of YIN, teach the limitations of claim 1.
Further, VIJAYAKUMAR teaches “A computer program product comprising at least one tangible, non-transitory computer-readable medium embodying instructions which, when executed by at least one processor, cause the at least one processor to implement a method according to claim 1.”
See VIJAYAKUMAR in [0011] describe “According to yet another embodiment, the present subject matter relates to a computer program product having non-volatile memory carrying computer executable instructions stored therein for determining potential outages in an information technology (IT) environment. The instructions comprising extracting, from one or more data sources, data for one or more parameters associated with a plurality of nodes in the IT environment.” Here, VIJAYAKUMAR teaches a computer program product that has non-volatile memory similar to a non-transitory computer-readable medium that contain instructions.
Further, see VIJAYAKUMAR in [0009] describe “According to an embodiment of the present subject matter, a system for determining potential outages in the IT environment is disclosed. The system comprises a user interface, one or more processing units, a memory unit coupled to the one or more processing units. The memory unit comprises at least a data extraction module, a data classifier module, a prediction module, and a display module.” Here, VIJAYAKUMAR teaches processing units and a memory that work with the processors.
Claim 10:
Regarding claim 10, VIJAYAKUMAR in view of WHITE, and further in view of YIN, teach the limitations in claim 1.
Further, VIJAYAKUMAR teaches “A data processing system comprising at least one processor and memory coupled to the at least one processor wherein the memory stores instructions which, when executed by the at least one processor, cause the data processing system to implement a method according to claim 1.”
See VIJAYAKUMAR in [0009] describe “a system for determining potential outages in the IT environment is disclosed. The system comprises a user interface, one or more processing units, a memory unit coupled to the one or more processing units. The memory unit comprises at least a data extraction module, a data classifier module, a prediction module, and a display module.” Here, VIJAYAKUMAR teaches a processing unit, memory, and a data extraction module that are part of a data processing system.
Claim 11:
Regarding claim 11, VIJAYAKUMAR teaches “A method for predicting an application outage on a computer network, the method comprising: receiving, from a trained utilization model, a utilization prediction of future application-level network utilization for an application executing on the computer network;”
See VIJAYAKUMAR in [0046] mention “The training module 407 receives the cleansed data to train the prediction model before the prediction module 408 and the alert module 410 perform their respective functions. In some embodiments, a summary generation module 609 may generate a summary of the predicted status, utilization, and performance of the nodes. The summary may provide statistical and graphical representations of the status of a node or the performance and utilization of the nodes. Further, it may provide additional information, such as one or more opportunities of potential outages, the expected time, root cause analysis, reason, severity, probability of potential failure of the node, recommended operators to mitigate the failure, etc.” Here, VIJAYAKUMAR shows a system that receives cleaned data to train a prediction model, and the system later generates outputs in form of predicted status, utilization or performance of devices of the network (i.e. a utilization prediction of future application-level network utilization).
Also, see VIJAYAKUMAR in [0028] describe “ ‘Nodes’ refer to a device or system in the network that can receive, create, store or send data along distributed network routes. Nodes may include web servers, application servers, database servers, laptops, computers, mobile devices, smart devices, etc.” Here, nodes also include application servers, which are part of applications within a network. From [0046], VIJAYAKUMAR mentions that the prediction of status, utilization and performance scores relate to each of the nodes, which also includes applications run in servers defined in [0028].
See VIJAYAKUMAR also in [0036] describe “According to an embodiment, a network architecture of the IT environment 300 comprising a system 301 for determining application health is illustrated in FIG. 3. The architecture may include the system 301, one or more local networks 303-1-303-N comprising one or more servers 304, one or more end user devices 305, all of which may be connected over the network 302. The one or more servers 304 render essential services required in IT environments. In some examples, the servers may include web servers for delivering content or services to end users through the network; application servers to facilitate installation, operation, hosting of applications; database servers to run database applications”. The examiner construes application to mean any program or service that requires a computer network (like the internet or a local area network) to operate. Here, VIJAYAKUMAR mentions using data that relates to network usage, use of application servers or host applications, which relate to application level operation and application health status. Further, see VIJAYAKUMAR in [0030] for more details.
However, VIJAYAKUMAR did not teach “receiving, from a trained application health model trained independently of the neural network model, an application health score for the application; combining the application health score from the application health model and the utilization prediction from the neural network model into a combined health score for the application”
In an analogous art, WHITE teaches “receiving, from a trained application health model trained independently of the neural network model, an application health score for the application;”
See WHITE in page 1089, abstract, describe “Accurate short-term forecasts allow dynamic systems to adapt their behaviour when degradation is forecast …This rationale can be applied to service-oriented computing when creating and managing service applications. Recent approaches to improve reliability in service applications have focused on reducing the time to recovery of application using collaborative filtering-based approaches to make QoS predictions for similar users. In this article, we focus on reducing the time to detection of a failure by forecasting when a service is about to degrade in quality.” WHITE mentions using approaches to improve reliability in service applications by forecasting service quality change corresponds to detecting application health.
Further, see WHITE in page 1090, second paragraph, mention “ For example, in SmartSantander the QoS values from temperature sensors can be sent to the forecasting devices to learn a model that can be used to forecast future QoS values for the service. These forecast values can be used to identify when a service is about to fail… To address the limitations of LSTM, which can take a couple of hours to train on the IoT dataset, this work investigates echo state networks, which are a recurrent neural network with a sparsely connected hidden layer (typically 1 percent connectivity). ” Here, WHITE describes QoS values to forecast values that identifies when a service is not working shows application health scores. WHITE also shows using a recurrent neural network called echo state networks as one of the neural network models to predict QoS values. The examiner relates quality of service (QoS) in networking refers to a set of technologies and mechanisms that manage network traffic to ensure the performance of critical applications, especially under limited network capacity. This allows organizations to prioritize specific applications, ensuring that high-priority traffic receives the necessary bandwidth and low-latency parameters, thereby improving overall network performance. Here, QoS is synonymous for tracking the health status for an application in a network.
See WHITE in page 1095, second full paragraph, section 5.2 Metrics, mention “For each of the models, we record the training and prediction time on each of the datasets. The training time is the average time taken to train the model on the training dataset and the prediction time is the average time to generate a forecast given the test data.” Further, see WHITE in page 1089, Introduction section in paragraph 2, describe “These equivalent services can be used during a service adaptation that is responding to forecast QoS changes (e.g., unacceptable response time). To achieve this, knowledge about QoS values of the services is required to make timely and accurate adaptation decisions such as: when to trigger an adaptation action; which executing service(s) to replace and what candidate service(s) to choose.” Here, WHITE talks about using two models and train them separately. Later, see WHITE in page 1090 in section 2. Related work describes “Classical QoS forecasting methods have included traditional AR-based methods/variations, such as SETAR, ARIMA and GARCH models as well as baseline approaches such as linear moving average [15] and persistence models. Two time-series methods ARIMA and GARCH have been combined to produce more accurate QoS forecasting results than traditional ARIMA but requiring extra processing time .” WHITE mentions the use of a model called ARIMA, which means an AutoRegressive Integrated Moving Average that predicts QoS results, and not a neural network. Since QoS values are synonymous for tracking the health status or performance (measured as a prediction or score) for an application in a network, WHITE teaches receiving an application health score for the application from a trained application health model trained independently of the neural network model mentioned in page 1090, second paragraph. See section 5.3 in page 1095 in WHITE for details. Further, see WHITE in page 1091, paragraph three, describe “Both of these approaches can be included in a middleware to manage service-based applications from a number of different providers. Fig. 1 shows how the two models can be incorporated into a middleware, in this case a prediction engine that receives QoS values from a monitoring engine.” Here, WHITE talks about using two models and let each predict QoS prediction results. Then, WHITE describes using these models to forecast QoS or quality of service changes to service applications. See WHITE in page 1089 abstract for details.
See WHITE mention in page 1090, section 2. Related work “time series forecasting techniques can be categorised into two groups: classical methods based on statistical/mathematical concepts and modern heuristic methods based on artificial intelligence algorithms [14]. The former includes exponential smoothing models, regression models, ARIMA models, threshold models and GARCH models. The latter includes artificial neural networks, which we extend to include RNN and ESN-based approaches.” Here, WHITE describes the use of a neural network model as one of the models.
Further, WHITE teaches “A method for predicting an application outage on a computer network, the method comprising: combining the application health score from the application health model and the utilization prediction from the neural network model into a combined health score for the application.”
See WHITE mention in page 1090, section 2. Related work “time series forecasting techniques can be categorised into two groups: classical methods based on statistical/mathematical concepts and modern heuristic methods based on artificial intelligence algorithms [14]. The former includes exponential smoothing models, regression models, ARIMA models, threshold models and GARCH models. The latter includes artificial neural networks, which we extend to include RNN and ESN-based approaches…” Here, WHITE describes the use of a neural network model as one of the models to detect QoS forecasting of network applications. See WHITE in page 1089, abstract, where WHITE mentions using approaches to improve reliability in service applications by forecasting service quality change corresponds to detecting application health for more details.
Further, see WHITE in page 1090, second paragraph, mention “ For example, in SmartSantander the QoS values from temperature sensors can be sent to the forecasting devices to learn a model that can be used to forecast future QoS values for the service. These forecast values can be used to identify when a service is about to fail… this work investigates echo state networks, which are a recurrent neural network with a sparsely connected hidden layer (typically 1 percent connectivity). ” Here, WHITE describes QoS values to forecast values that identifies when a service is not working shows application health scores. WHITE also shows using a recurrent neural network called echo state networks as one of the models to predict QoS values.
It would have been obvious for one of ordinary skill in the art before the effective filing date of the claimed invention to combine the base reference of VIJAYAKUMAR and incorporate into the teachings of WHITE because both references teach the process of training models from historical or multimodal data sources to predict network application outages.
One of ordinary skill in the art would be motivated to do so because “The results show the improved forecasting accuracy using the noisy echo forecasting approach for 80 percent of the test datasets. The two datasets that get better predictions from traditional time series-based,” (see WHITE in page 1096, section 6.1, Prediction accuracy).
However, VIJAYAKUMAR in view of WHITE did not teach “combining the application health score from the application health model and the utilization prediction from the neural network model into a combined health score for the application.”
In an analogous method, YIN teaches “combining the application health score from the application health model and the utilization prediction from the neural network model into a combined health score for the application.”
See YIN in page 9, section 3.3, Proposed Prediction methods, paragraph 3, describe “To further improve prediction accuracy, we propose a hybrid prediction method that combines the prediction results of UCF-E and SCF-E, aiming to fully take advantage of the whole QoS data. We name this method Hybrid CF with Ensemble learning (HCF-E), given as follows:
PNG
media_image1.png
76
1149
media_image1.png
Greyscale
where the parameter θ is used to control the proportions of the two individual models. UCF-Eu,j and SCF-Eu,j are the prediction values of the two individual models respectively.” Here, YIN shows using a method to combine two models UCF-E and SCF-E together to get one combined model. The HCF is considered a combined prediction value (i.e. combined health score) for the application. Also, see first few paragraphs on page 9, section 3.3, where YIN notes “Since the ensemble learning model takes an important role in our framework, we name the proposed method User-based CF with Ensemble learning (UCF-E), and the corresponded prediction is given by,
PNG
media_image2.png
1
753
media_image2.png
Greyscale
where wv,j is the probability of qv,j belonging to the label and qv,j is the real value that user v received after invoking service j.
In a similar way, we propose a new service-based CF method using the probability belonging to the labels of the missing QoS values, also replacing the traditional similarity. This proposed method is named Service-based CF with Ensemble learning (SCF-E), and the prediction is given as,
PNG
media_image2.png
1
753
media_image2.png
Greyscale
where wu,h is the probability of qu,h belonging to the label. qu,h is the real value after user u invoked service h.”
See YIN in page 13, conclusion, describe “we also plan to construct a real-world service selection system for mixed mobile networks.” Here, YIN refers the system to be a combiner to combine the results for two models. The prediction value is measured with accuracy, see Tables 4 and 5 in YIN for details.
Further, see YIN in page 11, section 4.4. Sensitivity Analysis of θ mention “The parameter θ is used to control the weight of the two individual models (UCF-E and SCF-E) in the combination model.” YIN mentions combining the models also includes combining their weights. See YIN in abstract in page 1 for more details.
Further, see YIN in page 1, Introduction mention “developing mobile services has become an increasingly important way for various enterprises to deliver their marketing applications that only customs’ functional demands and satisfy their expected non-functional requirements.” YIN relates the method to user applications such as marketing applications.
See also page 2 in YIN from Introduction section mention “There has been much research into QoS prediction [10,11,12]. For example, collaborative filtering (CF) uses historical QoS records to predict unknown QoS values.” Since quality of service (QoS) predictions and network application health scores achieve the same goal of achieving optimal user experience for network application performance, both the two models UCF-E and SCF-E relate to QoS predicted values and thus network application health. While QoS prediction anticipates network performance, application health scores quantify how well the application performs under those conditions, these two scores ultimately serve a similar purpose. Since both are analogous topics, YIN page 9 mentions that the UCF-E model, being user-based, measures the predictions on users with similar usage patterns (i.e. relates to prediction from utilization model) while the SCF-E model measures QoS predictions (i.e. relates to application health model) that are found on similar services with one of the predicted values is qu,h , which is the real value after user u invoked service h. This shows combine the application health score from the application health model and the utilization prediction from the utilization model into a combined health score, where the examiner construes score to be any predicted result, outcome, or calculated value from a model.
It would have been obvious to one of ordinary skill in the art before the effective filing date of the claimed invention to combine the reference of VIJAYAKUMAR along with the reference of WHITE with the teachings of YIN by using the teachings of VIJAYAKUMAR and WHITE of using historical and multimodal data to train utilization model and application health model respectively, with YIN’s teaching of combining the two models.
One of ordinary skill in the art would be motivated to do so because by integrating YIN’s framework into the methods of VIJAYAKUMAR and WHITE, one with ordinary skill in the art would achieve the goal of providing an “approach … oriented to a mixed mobile network environment. Our approach has the capability to improve similarity between two users/services. Moreover, it can also handle abnormal QoS data” (see YIN in page 2).
Claim 14:
Regarding claim 14, VIJAYAKUMAR in view of WHITE, further in view of YIN, teach the limitations of claim 11. Regarding claim 14, the claim recites similar limitations as corresponding claim 3 and is rejected for similar reasons as claim 3 using similar teachings and rationale.
Claim 15:
Regarding claim 15, VIJAYAKUMAR in view of WHITE, further in view of YIN, teach the limitations of claim 14. Regarding claim 15, the claim recites similar limitations as corresponding claim 4 and is rejected for similar reasons as claim 4 using similar teachings and rationale.
Claim 16:
Regarding claim 16, VIJAYAKUMAR in view of WHITE, further in view of YIN, teach the limitations of claim 11.
Further, VIJAYAKUMAR teaches “The method of claim 11, wherein the utilization model predicts from historical application-level network utilization data for the computer network,”
See VIJAYAKUMAR in [0042] describe “The training module 407 is configured to train a prediction model based on the historical data using a machine learning engine. The historical data is used as a training dataset, which is provided as an input to the training model. In some embodiments, the data classifier module 406 may be configured to categorize the data as historical if it has been used in a training dataset. Further, the prediction module 408 is configured to feed the current data to the prediction model for predicting status score, utilization score, and performance score of each of the plurality of nodes.” Here, VIJAYAKUMAR mentions training a model that can predict performance, utilization or status using historical data.
See VIJAYAKUMAR in [0028] describe “ ‘Nodes’ refer to a device or system in the network that can receive, create, store or send data along distributed network routes. Nodes may include web servers, application servers, database servers, laptops, computers, mobile devices, smart devices, etc.” Here, VIJAYAKUMAR shows that nodes include computers and other devices that are part of a network that can receive or send data along distributed network routes, and relate to a network system that includes computers or other devices (i.e. part of computer network).
Claim 17:
Regarding claim 17, VIJAYAKUMAR in view of WHITE, further in view of YIN, teach the limitations of claim 11.
Further, VIJAYAKUMAR teaches “The method of claim 11, wherein the application health model predicts from multimodal application health metric data.”
See VIJAYAKUMAR in [0008] describe “The method includes extracting, from one or more data sources, data for one or more parameters associated with a plurality of nodes in the IT environment. The data comprising at least utilization metrics, performance metrics, and a time identifier for each of the performance and utilization metric. The extracted data is classified as historical data or a current data based on the time identifier. A status score, utilization score, and performance score, of the plurality of nodes, or a combination thereof for the plurality of nodes from the classified data are predicted.” VIJAYAKUMAR mentions taking one or more data sources that has performance and utilization metrics (i.e. multimodal application health metric data), to predict a score that relates to an application health score (from [0036]).
Claim 18:
Regarding claim 18, VIJAYAKUMAR in view of WHITE, further in view of YIN, teach the limitations of claim 17. Regarding claim 18, the claim recites similar limitations as corresponding claim 5 and is rejected for similar reasons as claim 5 using similar teachings and rationale.
Claim 20:
Regarding claim 20, VIJAYAKUMAR in view of WHITE, further in view of YIN, teach the limitations of claim 11. Regarding claim 20, the claim recites similar limitations as corresponding claim 8 and is rejected for similar reasons as claim 8 using similar teachings and rationale.
Claim 21:
Regarding claim 21, VIJAYAKUMAR in view of WHITE, further in view of YIN, teach the limitations of claim 11. Regarding claim 21, the claim recites similar limitations as corresponding claim 9 and is rejected for similar reasons as claim 9 using similar teachings and rationale.
Claim 22:
Regarding claim 22, VIJAYAKUMAR in view of WHITE, further in view of YIN, teach the limitations of claim 11. Regarding claim 22, the claim recites similar limitations as corresponding claim 10 and is rejected for similar reasons as claim 10 using similar teachings and rationale.
Claims 2, 12, and 13 are rejected under 35 U.S.C. 103 over VIJAYAKUMAR in view of WHITE, further in view of YIN, and further in view of Khan, S. et al., in “Towards a Cloud-based Machine Learning for Health Monitoring and Fault Diagnosis”, published on July 14, 2017, available at https://www.papers.phmsociety.org/index.php/phmap/article/view/1843 , (hereafter, KHAN).
Claim 2:
Regarding claim 2, VIJAYAKUMAR in view of WHITE, further in view of YIN, teach the limitations of claim 1.
Further, YIN teaches “the method of claim 1, further comprising providing for conformation of at least one of the application health score and the utilization prediction so that the application health score and the utilization prediction share a common format and are combinable with one another by the combiner”
See YIN in page 9, section 3.3, Proposed Prediction methods, paragraph 3, describe “To further improve prediction accuracy, we propose a hybrid prediction method that combines the prediction results of UCF-E and SCF-E, aiming to fully take advantage of the whole QoS data. We name this method Hybrid CF with Ensemble learning (HCF-E), given as follows:
PNG
media_image1.png
76
1149
media_image1.png
Greyscale
where the parameter θ is used to control the proportions of the two individual models. UCF-Eu,j and SCF-Eu,j are the prediction values of the two individual models respectively.” Here, YIN shows using a method to combine two models UCF-E and SCF-E together to get one combined model. The HCF is considered a combined prediction value (i.e. combined health score) for the application. Also, see first few paragraphs on page 9, section 3.3, where YIN notes “Since the ensemble learning model takes an important role in our framework, we name the proposed method User-based CF with Ensemble learning (UCF-E), and the corresponded prediction is given by,
PNG
media_image2.png
1
753
media_image2.png
Greyscale
where wv,j is the probability of qv,j belonging to the label and qv,j is the real value that user v received after invoking service j.
In a similar way, we propose a new service-based CF method using the probability belonging to the labels of the missing QoS values, also replacing the traditional similarity. This proposed method is named Service-based CF with Ensemble learning (SCF-E), and the prediction is given as,
PNG
media_image2.png
1
753
media_image2.png
Greyscale
where wu,h is the probability of qu,h belonging to the label. qu,h is the real value after user u invoked service h.”
See YIN in page 13, conclusion, describe “we also plan to construct a real-world service selection system for mixed mobile networks.” Here, YIN refers the system to be a combiner to combine the results for two models. The prediction value is measured with accuracy, see Tables 4 and 5 in YIN for details.
Further, see YIN in page 11, section 4.4. Sensitivity Analysis of θ mention “The parameter θ is used to control the weight of the two individual models (UCF-E and SCF-E) in the combination model.” YIN mentions combining the models also includes combining their weights. See YIN in abstract in page 1 for more details.
Further, see YIN in page 1, Introduction mention “developing mobile services has become an increasingly important way for various enterprises to deliver their marketing applications that only customs’ functional demands and satisfy their expected non-functional requirements.” YIN relates the method to user applications such as marketing applications.
See also page 2 in YIN from Introduction section mention “There has been much research into QoS prediction [10,11,12]. For example, collaborative filtering (CF) uses historical QoS records to predict unknown QoS values.” Since quality of service (QoS) predictions and network application health scores achieve the same goal of achieving optimal user experience for network application performance, both the two models UCF-E and SCF-E relate to QoS predicted values and thus network application health. While QoS prediction anticipates network performance, application health scores quantify how well the application performs under those conditions, these two scores ultimately serve a similar purpose. Since both are analogous topics, YIN page 9 mentions that the UCF-E model, being user-based, measures the predictions on users with similar usage patterns (i.e. relates to prediction from utilization model) while the SCF-E model measures QoS predictions (i.e. relates to application health model) that are found on similar services with one of the predicted values is qu,h , which is the real value after user u invoked service h. This shows combine the application health score from the application health model and the utilization prediction from the utilization model into a combined health score, where the examiner construes score to be any predicted result, outcome, or calculated value from a model.
However, VIJAYAKUMAR in view of WHITE, further in view of YIN, did not teach “the method of claim 1, further comprising providing for conformation of at least one of the application health score and the utilization prediction so that the application health score and the utilization prediction share a common format and are combinable with one another by the combiner,”
In an analogous art, KHAN teaches “the method of claim 1, further comprising providing for conformation of at least one of the application health score and the utilization prediction so that the application health score and the utilization prediction share a common format and are combinable with one another by the combiner,”
See KHAN in page 505, section 4, Cloud-based decision support systems describe “when data is collected, it needs to be conditioned to be compatible with the common format. It is converted before it can be stored or be used by for any ‘training’ or application control to be available for any authorized user who wishes to access the processed information. This includes details of any alarms, maintenance schedules, fault analysis results, prognostic results, costings, and more, hence enabling users to work more efficiently.” Here, KHAN shows that the data and its results need to be compatible with a common format. Further, see KHAN in page 502, section 2.2, second to last paragraph in section note “For health management and fault analysis, a cloud service can overcome limitations of handling large data sets, often located in various repositories. However, this introduced challenges of utilization, network bandwidth, resource provisioning, improving application-network interaction and performance characteristics. Furthermore, such technology can be used to produce models of the decision-making process.” Here, KHAN describes building models for network applications to detect results of fault analysis, which is a method of identifying, evaluating, and predicting defects or abnormal conditions within a system, which relates to the analogous art of network applications.
Further, see KHAN in page 503, section 2.2 cloud-based machine learning describe “These techniques can be recognizing patterns, help cluster and classify data to extract features to perform regression (or reinforcement learning) for anomaly detection problems. In sudden fluctuating conditions, such as the ones considered in pervasive computing, systems are expected to adapt their behavioral models according to current conditions.” This shows that the methods use models to detect any anomaly or abnormal behavior of devices.
It would have been obvious to one of ordinary skill in the art before the effective filing date of the claimed invention to combine the references of VIJAYAKUMAR, WHITE, and YIN and incorporate with the teachings of KHAN by using the teachings of VIJAYAKUMAR, WHITE, and YIN for a method of training models for an application outage predictor, with KHAN’s teaching of models sharing a common format.
One of ordinary skill in the art would be motivated to do so because by integrating KHAN’s framework into the methods of VIJAYAKUMAR, WHITE, and YIN, one with ordinary skill in the art would achieve “now with big data applications, more data patterns can help manage and efficiently collate data behavior,” (see KHAN in page 502, first half paragraph).
Claim 12:
Regarding claim 12, VIJAYAKUMAR in view of WHITE, further in view of YIN, teach the limitations of claim 11. Regarding claim 12, the claim recites similar limitations as corresponding claim 2 and is rejected for similar reasons as claim 2 using similar teachings and rationale.
Claim 13:
Regarding claim 13, VIJAYAKUMAR in view of WHITE, further in view of YIN, teach the limitations of claim 11. Further, YIN teaches “13. The method of claim 11, wherein the application health model and the utilization model are configured so that the application health score and the utilization prediction share a common format,”
See YIN in page 9, section 3.3, Proposed Prediction methods, paragraph 3, describe “To further improve prediction accuracy, we propose a hybrid prediction method that combines the prediction results of UCF-E and SCF-E, aiming to fully take advantage of the whole QoS data. We name this method Hybrid CF with Ensemble learning (HCF-E), given as follows:
PNG
media_image1.png
76
1149
media_image1.png
Greyscale
where the parameter θ is used to control the proportions of the two individual models. UCF-Eu,j and SCF-Eu,j are the prediction values of the two individual models respectively.” Here, YIN shows using a method to combine two models UCF-E and SCF-E together to get one combined model. The HCF is considered a combined prediction value (i.e. combined health score) for the application.
See YIN in page 13, conclusion, describe “we also plan to construct a real-world service selection system for mixed mobile networks.” Here, YIN refers the system to be a combiner to combine the results for two models. The prediction value is measured with accuracy, see Tables 4 and 5 in YIN for details.
Further, see YIN in page 11, section 4.4. Sensitivity Analysis of θ mention “The parameter θ is used to control the weight of the two individual models (UCF-E and SCF-E) in the combination model.” YIN mentions combining the models also includes combining their weights. See YIN in abstract in page 1 for more details.
See YIN in first paragraph of page 7 note “Such filtering can improve the selection accuracy of similar neighbors. Figure 4 shows the framework of the AdaBoost classifier. In this paper, we adopt an ensemble learning method (i.e., AdaBoost algorithm) and use the decision tree as the weak classifier to further filtering similar neighbors.” The examiner construes format to mean any framework, structure, or version type in which a model (and sometimes its associated data) is stored so the model can be saved, shared, loaded, and used by different systems. Here, YIN shows the models used share a framework or a format, of an ensemble learning method.
Further, see YIN in page 1, Introduction mention “developing mobile services has become an increasingly important way for various enterprises to deliver their marketing applications that only customs’ functional demands and satisfy their expected non-functional requirements.” YIN relates the method to user applications such as marketing applications.
See also page 2 in YIN from Introduction section mention “There has been much research into QoS prediction [10,11,12]. For example, collaborative filtering (CF) uses historical QoS records to predict unknown QoS values.” Since quality of service (QoS) predictions and network application health scores achieve the same goal of achieving optimal user experience for network application performance, both the two models UCF-E and SCF-E relate to QoS predicted values and thus network application health. While QoS prediction anticipates network performance, application health scores quantify how well the application performs under those conditions, these two scores ultimately serve a similar purpose. Since both are analogous topics, YIN page 9 mentions that the UCF-E model, being user-based, measures the predictions on users with similar usage patterns (i.e. relates to prediction from utilization model) while the SCF-E model measures QoS predictions (i.e. relates to application health model) that are found on similar services with one of the predicted values is qu,h , which is the real value after user u invoked service h. This shows combine the application health score from the application health model and the utilization prediction from the utilization model into a combined health score, where the examiner construes score to be any predicted result, outcome, or calculated value from a model.
However, VIJAYAKUMAR in view of WHITE, further in view of YIN, did not teach “13. The method of claim 11, wherein the application health model and the utilization model are configured so that the application health score and the utilization prediction share a common format,”
In an analogous art, KHAN teaches “13. The method of claim 11, wherein the application health model and the utilization model are configured so that the application health score and the utilization prediction share a common format,”
See KHAN in page 505, section 4, Cloud-based decision support systems describe “when data is collected, it needs to be conditioned to be compatible with the common format. It is converted before it can be stored or be used by for any ‘training’ or application control to be available for any authorized user who wishes to access the processed information. This includes details of any alarms, maintenance schedules, fault analysis results, prognostic results, costings, and more, hence enabling users to work more efficiently.” Here, KHAN shows that the data and its results need to be compatible with a common format. Further, see KHAN in page 502, section 2.2, second to last paragraph in section note “For health management and fault analysis, a cloud service can overcome limitations of handling large data sets, often located in various repositories. However, this introduced challenges of utilization, network bandwidth, resource provisioning, improving application-network interaction and performance characteristics. Furthermore, such technology can be used to produce models of the decision-making process.” Here, KHAN describes building models for network applications to detect results of fault analysis, which is a method of identifying, evaluating, and predicting defects or abnormal conditions within a system, which relates to the analogous art of identifying outages in network applications. See KHAN in page 503, section 2.2 for further details.
It would have been obvious to one of ordinary skill in the art before the effective filing date of the claimed invention to combine the references of VIJAYAKUMAR, WHITE, and YIN and incorporate with the teachings of KHAN by using the teachings of VIJAYAKUMAR, WHITE, and YIN for a method of training models for an application outage predictor, with KHAN’s teaching of models sharing a common format.
One of ordinary skill in the art would be motivated to do so because by integrating KHAN’s framework into the methods of VIJAYAKUMAR, WHITE, and YIN, one with ordinary skill in the art would achieve “now with big data applications, more data patterns can help manage and efficiently collate data behavior,” (see KHAN in page 502, first half paragraph).
Claims 6, 7, and 19 are rejected under 35 U.S.C. 103 over VIJAYAKUMAR in view of WHITE, further in view of YIN, and further in view of Côté, D. et al., US PG Pub. US20190280942-A1, published on September 12, 2019, (hereafter, CÔTÉ).
Claim 6:
Regarding claim 6, VIJAYAKUMAR in view of WHITE, further in view of YIN, teach the limitations of claim 5.
However, VIJAYAKUMAR in view of WHITE, further in view of YIN, did not teach “The method of claim 5, wherein the outage information comprises volumetric problem report data from at least one external public Internet platform that is outside of the computer network and that is nonspecific to the application.”
In an analogous art, CÔTÉ teaches “The method of claim 5, wherein the outage information comprises volumetric problem report data from at least one external public Internet platform that is outside of the computer network and that is nonspecific to the application.”
See CÔTÉ in [0038] describe “For instance, layer-3 network performance is characterized by bandwidth, throughput, latency, jitter and error rate. End-users', environmental, or business data typically come from third-party databases.” Here, CÔTÉ describes third-party databases which relate to external public Internet platforms, which is outside of the computer network, and CÔTÉ in [0040] mention “When networks contain a large number of devices and services, with high-frequency data-collection and/or long storage periods, the result is large data volumes. The combined Variety, Velocity and Volume is often referred to as “Big Data.” Here, CÔTÉ shows that data is in volumetric problem reports.
It would have been obvious to one of ordinary skill in the art before the effective filing date of the claimed invention to combine the references of VIJAYAKUMAR, WHITE, and YIN and incorporate with the teachings of CÔTÉ by using the teachings of VIJAYAKUMAR, WHITE, and YIN for a method of training models for an application outage predictor, with CÔTÉ’s teaching of volumetric problem report data from at least one external public Internet platform.
One of ordinary skill in the art would be motivated to do so because by integrating CÔTÉ’s framework into the methods of VIJAYAKUMAR, WHITE, and YIN, one with ordinary skill in the art would achieve the goal of providing “ML applications 22 have the potential reduce the cost of network operations amid an unprecedented time of increased complexity. They can also improve end-user experience and create new revenue opportunities for network service providers,” (see CÔTÉ in [0078]).
Claim 7:
Regarding claim 7, VIJAYAKUMAR in view of WHITE, further in view of YIN, teach the limitations of claim 1.
However, VIJAYAKUMAR in view of WHITE, further in view of YIN, did not teach “The method of claim 1, wherein the utilization model is a Long Short Term Memory (LSTM) neural network model.”
In an analogous field, CÔTÉ teaches “The method of claim 1, wherein the utilization model is a Long Short Term Memory (LSTM) neural network model.”
See CÔTÉ describe in [0167] “The converting cam utilize a 1st or 2nd order polynomial for scenarios in which performance of the component, device, or link is degrading continuously, a piece-wise combination of the 1st or 2nd order polynomials for scenarios in which the performance is first stable, then starts degrading, and a Long Short-Term Memory (LSTM) neural network or Autoregressive Integrated Moving Average (ARIMA) model for scenarios in which the performance varies with seasonal effects.” See CÔTÉ for more details in [0068, 0082].
See CÔTÉ describe in [0073] “The objective of the machine learning system 10 and machine learning process is to identify problems before outages, service disruption, etc. Thus, the remedial action is anything to further those objectives.” Here, CÔTÉ describes using a LSTM neural network model to detect problems that occur before outages or track service disruption of a system.
It would have been obvious to one of ordinary skill in the art before the effective filing date of the claimed invention to combine the references of VIJAYAKUMAR, WHITE, and YIN and incorporate with the teachings of CÔTÉ by using the teachings of VIJAYAKUMAR, WHITE, and YIN for a method of training models for an application outage predictor, with CÔTÉ’s teaching of using a Long Short Term Memory (LSTM) neural network model.
One of ordinary skill in the art would be motivated to do so because by integrating CÔTÉ’s framework into the methods of VIJAYAKUMAR, WHITE, and YIN, one with ordinary skill in the art would achieve the goal of providing “ML applications 22 have the potential reduce the cost of network operations amid an unprecedented time of increased complexity. They can also improve end-user experience and create new revenue opportunities for network service providers,” (see CÔTÉ in [0078]).
Claim 19:
Regarding claim 19, VIJAYAKUMAR in view of WHITE, further in view of YIN, teach the limitations of claim 11. Regarding claim 19, the claim recites similar limitations as corresponding claim 7 and is rejected for similar reasons as claim 7 using similar teachings and rationale.
Claims 23, 24, 25, 26, and 27 are rejected under 35 U.S.C. 103 over VIJAYAKUMAR in view of YIN.
Claim 23:
Regarding claim 23, VIJAYAKUMAR teaches “A method for predicting an application outage on a computer network, the method comprising: combining a utilization prediction of future application-level network utilization for an application executing on the computer network with an application health score for the application derived from multiple sources into a combined health score for the application.”
See VIJAYAKUMAR in [0008] describe “the method includes extracting, from one or more data sources, data for one or more parameters associated with a plurality of nodes in the IT environment. The data comprising at least utilization metrics, performance metrics, and a time identifier for each of the performance and utilization metric. The extracted data is classified as historical data or a current data based on the time identifier. A status score, utilization score, and performance score, of the plurality of nodes, or a combination thereof for the plurality of nodes from the classified data are predicted.” VIJAYAKUMAR mentions a method taking one or more data sources that has performance and utilization metrics (i.e. derived from multiple sources), to predict a score that relates to an application health score (from [0036]).
See VIJAYAKUMAR also in [0036] describe “According to an embodiment, a network architecture of the IT environment 300 comprising a system 301 for determining application health is illustrated in FIG. 3. The architecture may include the system 301, one or more local networks 303-1-303-N comprising one or more servers 304, one or more end user devices 305, all of which may be connected over the network 302. The one or more servers 304 render essential services required in IT environments. In some examples, the servers may include web servers for delivering content or services to end users through the network; application servers to facilitate installation, operation, hosting of applications; database servers to run database applications”. The examiner construes application to mean any program or service that requires a computer network (like the internet or a local area network) to operate. Here, VIJAYAKUMAR mentions using data (both historical and real-time) that relates to network, use of application servers or host applications, which relate to application operation status.
Further, see VIJAYAKUMAR in [0047] describe “The above subject matter and its embodiments provide method and system to determine and prevent potential outages. The present subject matter predicts utilization and performance of infrastructure resources, which helps in enhancing the overall performance of the network. The predictions also enable the operations team to optimize processes, anticipate suspicious trends before loss occurs, gain insights into the causes and relationships of downtime with performance. Further, the invention helps to anticipate performance issues in future and enhance overall productivity, revenue, and security of enterprises.” Here in [0047], VIJAYAKUMAR mentions that this invention including models or algorithms that accompany the invention helps to ‘anticipate performance issues’ or also predict future performance usage including performance of applications implemented on network servers mentioned in [0036] (i.e. relates to prediction of future application-level network).
However, VIJAYAKUMAR did not teach “combining a utilization prediction of future application-level network utilization for an application executing on the computer network with an application health score for the application derived from multiple sources into a combined health score for the application,”
In an analogous art, YIN teaches “combining a utilization prediction of future application-level network utilization for an application executing on the computer network with an application health score for the application derived from multiple sources into a combined health score for the application,”
See YIN in page 9, section 3.3, Proposed Prediction methods, paragraph 3, describe “To further improve prediction accuracy, we propose a hybrid prediction method that combines the prediction results of UCF-E and SCF-E, aiming to fully take advantage of the whole QoS data. We name this method Hybrid CF with Ensemble learning (HCF-E), given as follows:
PNG
media_image1.png
76
1149
media_image1.png
Greyscale
where the parameter θ is used to control the proportions of the two individual models. UCF-Eu,j and SCF-Eu,j are the prediction values of the two individual models respectively.” Here, YIN shows using a method to combine two models UCF-E and SCF-E together to get one combined model. The HCF is considered a combined prediction value (i.e. combined health score) for the application. Also, see first few paragraphs on page 9, section 3.3, where YIN notes “… the proposed method User-based CF with Ensemble learning (UCF-E), and the corresponded prediction is given by,
PNG
media_image2.png
1
753
media_image2.png
Greyscale
where wv,j is the probability of qv,j belonging to the label and qv,j is the real value that user v received after invoking service j. In a similar way, we propose a new service-based CF method using the probability belonging to the labels of the missing QoS values, also replacing the traditional similarity. This proposed method is named Service-based CF with Ensemble learning (SCF-E), and the prediction is given as,
PNG
media_image2.png
1
753
media_image2.png
Greyscale
where wu,h is the probability of qu,h belonging to the label. qu,h is the real value after user u invoked service h.”
See YIN in page 13, conclusion, describe “we also plan to construct a real-world service selection system for mixed mobile networks.” Here, YIN refers the system to be a combiner to combine the results for two models. The prediction value is measured with accuracy, see Tables 4 and 5 in YIN for details.
Further, see YIN in page 11, section 4.4. Sensitivity Analysis of θ mention “The parameter θ is used to control the weight of the two individual models (UCF-E and SCF-E) in the combination model.” YIN mentions combining the models also includes combining their weights. See YIN in abstract in page 1 for more details.
Further, see YIN in page 1, Introduction mention “developing mobile services has become an increasingly important way for various enterprises to deliver their marketing applications that only customs’ functional demands and satisfy their expected non-functional requirements.” YIN relates the method to user applications for networks such as marketing applications.
See also page 2 in YIN from Introduction section mention “There has been much research into QoS prediction [10,11,12]. For example, collaborative filtering (CF) uses historical QoS records to predict unknown QoS values.” Since quality of service (QoS) predictions and network application health scores achieve the same goal of achieving optimal user experience for network application performance, both the two models UCF-E and SCF-E relate to QoS predicted values and thus network application health. While QoS prediction anticipates network performance, application health scores quantify how well the application performs under those conditions, these two scores ultimately serve a similar purpose. Since both are analogous topics, YIN page 9 mentions that the UCF-E model, being user-based, measures the predictions on users with similar usage patterns (i.e. relates to prediction from utilization model) while the SCF-E model measures QoS predictions (i.e. relates to application health model) that are found on similar services with one of the predicted values is qu,h , which is the real value after user u invoked service h. This shows combine the application health score from the application health model and the utilization prediction from the utilization model into a combined health score, where the examiner construes score to be any predicted result, outcome, or calculated value from a model.
It would have been obvious to one of ordinary skill in the art before the effective filing date of the claimed invention to combine the reference of VIJAYAKUMAR, and incorporate with the teachings of YIN by using the teachings of VIJAYAKUMAR of using multiple sources for predicting network application use, with YIN’s teaching of combining the two models.
One of ordinary skill in the art would be motivated to do so because by integrating YIN’s framework into the methods of VIJAYAKUMAR, one with ordinary skill in the art would achieve the goal of providing an “approach … oriented to a mixed mobile network environment. Our approach has the capability to improve similarity between two users/services. Moreover, it can also handle abnormal QoS data” (see YIN in page 2).
Claim 24:
Regarding claim 24, VIJAYAKUMAR in view of YIN, teach the limitations of claim 23.
Further, VIJAYAKUMAR teaches “The method of claim 23, further comprising:
applying a threshold test to the combined health score; and initiating remedial action in response to the combined health score failing the threshold test.”
See VIJAYAKUMAR in [0010] describe “Based on the potential outages the alert module sends alerts to one or more devices if the predicted scores exceed a threshold limit or potential outage is identified. The summary generation module is configured to generate a summary comprising at least trends and statistics associated with the predicted status, utilization, and performance, of the plurality of nodes. The data cleansing module is configured to detect inaccurate data in the data store and perform a corrective action on the data.” Here, VIJAYAKUMAR mentions if a predicted score exceeds a threshold limit, there is a corrective action or alert sent to the devices.
See VIJAYAKUMAR in [0032] mention “A status score, utilization score, performance score, or a combination thereof are predicted based on the historical data and the current data, at block 103 …The status may indicate a failure status of the nodes, supplemented with information including, but not limited to, root cause analysis, time, reason, severity, probability of potential failure of the node, etc. A potential outage is determined in the IT environment from the predicted scores at block 104. The determination may include comparison of the predicted score and a predetermined threshold limit set by an administrator.” VIJAYAKUMAR shows if the predicted score exceeds the threshold limit on the score (that score is a combination of status, utilization and performance), then there is a potential outage. See VIJAYAKUMAR in [0043] for more details.
Claim 25:
Regarding claim 25, VIJAYAKUMAR in view of YIN, teach the limitations of claim 24.
Further, VIJAYAKUMAR teaches “The method of claim 24, wherein the remedial action is at least one of an alarm or an automated correction procedure,”
See VIJAYAKUMAR in [0043] describe “The alert module 410 is configured to send alerts to one or more devices if the predicted scores or potential outage exceed a threshold limit. The predicted scores may be compared with a predetermined threshold limit to identify potential outage in the IT environment.” Further, see VIJAYAKUMAR in [0044] mention “Further, automated workflows may be triggered to kill processes, which are consuming more utilization, based on the utilization anomaly predicted for a CPU, memory, etc. In scenarios where one or more servers may be on the verge of shutting down due to high demand during peak time, the system may automatically identify the processes that may be insignificant or may require substantial resources from the servers, ” Here, VIJAYAKUMAR describes sending alerts to devices if an outage is detected, and an automated workflow triggered to stop processes is initiated as another corrective action (i.e. remedial action is an alarm).
Claim 26:
Regarding claim 26, VIJAYAKUMAR in view of YIN, teach the limitations of claim 23. Further, VIJAYAKUMAR teaches “A computer program product comprising at least one tangible, non-transitory computer-readable medium embodying instructions which, when executed by at least one processor, cause the at least one processor to implement a method according to claim 23.”
See VIJAYAKUMAR in [0011] describe “According to yet another embodiment, the present subject matter relates to a computer program product having non-volatile memory carrying computer executable instructions stored therein for determining potential outages in an information technology (IT) environment. The instructions comprising extracting, from one or more data sources, data for one or more parameters associated with a plurality of nodes in the IT environment.” Here, VIJAYAKUMAR teaches a computer program product that has non-volatile memory similar to a non-transitory computer-readable medium that contain instructions.
Further, see VIJAYAKUMAR in [0009] describe “ According to an embodiment of the present subject matter, a system for determining potential outages in the IT environment is disclosed. The system comprises a user interface, one or more processing units, a memory unit coupled to the one or more processing units. The memory unit comprises at least a data extraction module, a data classifier module, a prediction module, and a display module.” Here, VIJAYAKUMAR teaches processing units and a memory that work with the processors.
Claim 27:
Regarding claim 29, VIJAYAKUMAR in view of YIN, teach the limitations of claim 23.
Further, VIJAYAKUMAR teaches “A data processing system comprising at least one processor and memory coupled to the at least one processor wherein the memory stores instructions which, when executed by the at least one processor, cause the data processing system to implement a method according to claim 23.”
See VIJAYAKUMAR in [0009] describe “a system for determining potential outages in the IT environment is disclosed. The system comprises a user interface, one or more processing units, a memory unit coupled to the one or more processing units. The memory unit comprises at least a data extraction module, a data classifier module, a prediction module, and a display module.” Here, VIJAYAKUMAR teaches a processing unit, memory, and a data extraction module that are part of a data processing system.
Conclusion
Any inquiry concerning this communication or earlier communications from the examiner should be directed to WENWEI ZENG whose telephone number is (571)272-7111. The examiner can normally be reached Monday-Friday, 8am-5pm.
Examiner interviews are available via telephone, in-person, and video conferencing using a USPTO supplied web-based collaboration tool. To schedule an interview, applicant is encouraged to use the USPTO Automated Interview Request (AIR) at http://www.uspto.gov/interviewpractice.
If attempts to reach the examiner by telephone are unsuccessful, the examiner’s supervisor, Usmaan Saeed can be reached at (571) 272-4046. The fax phone number for the organization where this application or proceeding is assigned is 571-273-8300.
Information regarding the status of published or unpublished applications may be obtained from Patent Center. Unpublished application information in Patent Center is available to registered users. To file and manage patent submissions in Patent Center, visit: https://patentcenter.uspto.gov. Visit https://www.uspto.gov/patents/apply/patent-center for more information about Patent Center and https://www.uspto.gov/patents/docx for information about filing in DOCX format. For additional questions, contact the Electronic Business Center (EBC) at 866-217-9197 (toll-free). If you would like assistance from a USPTO Customer Service Representative, call 800-786-9199 (IN USA OR CANADA) or 571-272-1000.
/WenWei Zeng/Examiner, Art Unit 2146
/USMAAN SAEED/Supervisory Patent Examiner, Art Unit 2146
Prosecution Insights