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 .
Response to Amendment and Arguments
Applicant’s amendment filed on July 13, 2026 has been entered and made of record. Claims 1-27 are pending and are being examined in this application.
In light of Applicant’s remarks, the 112(d) rejection is replaced with a 112(b) rejection per applicant’s clarification that the substantially similar subject matter of dependent claims 2, 10, and 21 is intentional in order to positively recite the training of the set of predictive models.
Applicant’s arguments with respect to the 102 and 103 rejections have been fully considered, but are unpersuasive for at least the following reasons:
Applicant argues that Tabuchi does not disclose the claim limitations “assign the data stream to a data stream group” and “select, from a set of predictive models, a predictive model corresponding to the data stream group assigned to the data stream.” In particular, applicant argues “in Tabuchi, the set of interpreted data is grouped solely to facilitate storage and retrieval of the data on an AI data storage system—the grouping is not used to select an AI logic unit to process the set of interpreted data” and “in Tabuchi, the AI logic unit used to process the interpreted data is selected based on the profiles of the available AI logic units rather than the grouping that was used to store the interpreted data.” Remarks, pgs. 3-5.
However, Tabuchi is directed to “determining an artificial intelligence (AI) logic unit to perform a processing operation on a set of data collected from a network communication environment” [par. 66]. The set of (raw) data is preprocessed into a set of interpreted data before the processing operation is performed [par. 67]. The network communication environment includes a group of information sources having a plurality of information sources [par. 45]. The data “may differ based on the information source from which it was collected, and as such, it may be desirable to perform different processing operations on the data based on the nature and attributes of the information source” [par. 67]. “The set of interpreted data may be stored in the AI-based data storage system based on the set of attributes. Generally, storing can include saving, recording, collecting, aggregating, caching, or otherwise maintaining the set of interpreted data in the AI-based data storage system” [par. 77]. Bolded emphases added by examiner.
Per the above, Tabuchi discloses determining an AI unit (i.e., selecting a predictive model) to perform a processing operation on a set of data based on a set of attributes (i.e., a set of feature values). Also, the set of data is stored or otherwise maintained (i.e., grouped) based on the set of attributes. Thus, Tabuchi clearly discloses the claim limitation “assign the data stream to a data stream group.” Applicant’s assertion that “in Tabuchi, the AI logic unit used to process the interpreted data is selected based on the profiles of the available AI logic units rather than the grouping that was used to store the interpreted data” fails to mention that both the determining of the AI logic unit and the storing or otherwise maintaining of the set of data are based on the set of attributes.
Per the above, Tabuchi further discloses that different processing operations are performed by different AI logic units (i.e., a set of predictive models) based on different sets of attributes. As mentioned earlier, the AI logic unit to perform the processing operation on the set of data is determined (i.e., selected from the set of predictive models) based on the set of attributes, and the set of data is stored or otherwise maintained (i.e., grouped) based the set of attributes. Thus, Tabuchi clearly discloses “select, from a set of predictive models, a predictive model corresponding to the data stream group assigned to the data stream.” This is contrary to applicant’s assertion that “the set of interpreted data is grouped solely to facilitate storage and retrieval of the data on an AI data storage system—the grouping is not used to select an AI logic unit to process the set of interpreted data.”
Applicant also argues that Tabuchi does not disclose the claim limitations “wherein the predictive models are trained to predict a target variable for different data stream groups in the set of data stream groups based on different subsets of training data streams in a training dataset, wherein the training data streams comprise feature values corresponding to the same feature set as the data stream, and wherein each predictive model is trained to predict the target variable for a corresponding data stream group based on a corresponding subset of the training data streams.” In particular, applicant argues that Tabuchi “fails to disclose training AI logic units for corresponding groups of data using group-specific subsets of training data with a common feature set.” Remarks, pg. 5.
However, Tabuchi discloses “[T]he set of raw data may include an information source identification element for a first information source of the set of information sources… the information source identification element may include a tag, label, name, code, series of characters, or other attribute that uniquely identifies a particular information source among the set of information sources…consider a set of information sources including several hundred sensors for measuring different properties of a environment (e.g., humidity, temperature, pressure, luminescence). A unique identification element in the form of a 12-character string may be assigned to each individual sensor of the set of information sources” [par. 73]. The set of attributes is analyzed to “select an appropriate AI logic unit to process a set of interpreted data based on the attributes of the data” [par. 78]. “[A] first subset of the set of interpreted data may be identified for machine learning at block 521. The first subset of the set of interpreted data may be identified by comparing the set of attributes of the set of interpreted data with a set of machine learning usability criteria…the subset of interpreted data may include a portion of the set of interpreted data that is determined to be relevant, beneficial, helpful, effective, or otherwise useful for training a machine learning engine…the set of machine learning usability criteria may include factors such as whether data having a particular attribute has already been processed by the machine learning engine (e.g., data having new attributes may be relevant for training the machine learning technique), the type of information source from which the data was received (e.g., a photodetector, an accelerometer, an MES), contextual factors of the ingested data (e.g., the reason why the data is being collected), or the like” [par. 84].
Per the above, Tabuchi discloses using a “first subset” (i.e., training data stream) of the set of interpreted data as training data for a machine learning engine (i.e., a particular AI logic unit) and that the first subset is selected from other potential subsets (i.e., different subsets of training data streams) in the set of interpreted data (i.e., the training dataset) based on machine learning usability criteria such as particular attributes or types of information sources (i.e., feature values). The machine learning usability criteria are used to determine relevance or otherwise usefulness of a potential subset to the machine learning engine. The different subsets all share at least one attribute (i.e., feature value) in common, which is the information source identifier, because they are all part of the set of interpreted data and thus associated with the same information source. Thus, Tabuchi discloses “training AI logic units for corresponding groups of data using group-specific subsets of training data with a common feature set.”
Claim Rejections - 35 USC § 112
The following is a quotation of 35 U.S.C. 112(b):
(b) CONCLUSION.—The specification shall conclude with one or more claims particularly pointing out and distinctly claiming the subject matter which the inventor or a joint inventor regards as the invention.
The following is a quotation of 35 U.S.C. 112 (pre-AIA ), second paragraph:
The specification shall conclude with one or more claims particularly pointing out and distinctly claiming the subject matter which the applicant regards as his invention.
Claims 2, 10, and 21 are rejected under 35 U.S.C. 112(b) or 35 U.S.C. 112 (pre-AIA ), second paragraph, as being indefinite for failing to particularly point out and distinctly claim the subject matter which the inventor or a joint inventor (or for applications subject to pre-AIA 35 U.S.C. 112, the applicant), regards as the invention.
The claims have antecedent basis errors rendering the scope indefinite. For example, the claims recite the limitation “a set of training data streams,” which does not properly refer back to “the different subsets of training data streams,” the “corresponding subset of the training data streams,” or other element. There are also antecedent basis issues for the limitation “set of data stream groups” not properly referring back to “different data stream groups,” the “corresponding data stream group,” or other element. Please carefully review the language of these claims with the language of the independent claims from which they depend. Appropriate correction is required.
Claim Rejections - 35 USC § 102
The following is a quotation of the appropriate paragraphs of 35 U.S.C. 102 that form the basis for the rejections under this section made in this Office action:
A person shall be entitled to a patent unless –
(a)(1) the claimed invention was patented, described in a printed publication, or in public use, on sale, or otherwise available to the public before the effective filing date of the claimed invention.
(a)(2) the claimed invention was described in a patent issued under section 151, or in an application for patent published or deemed published under section 122(b), in which the patent or application, as the case may be, names another inventor and was effectively filed before the effective filing date of the claimed invention.
Claims 1, 2, 9, 10, 20, 21, and 23 are rejected under 35 U.S.C. 102(a)(1) and (a)(2) as being anticipated by Tabuchi et al. (US Pub. 20190244129).
Referring to claim 1, Tabuchi discloses A computing device, comprising: interface circuitry; and processing circuitry [fig. 3, computer system 300, network interface 318, processor 302] to:
receive, via the interface circuitry [fig. 3; network interface 318 connects to network 330], a data stream associated with a machine captured at least partially by one or more sensors [pars. 66, 70, and 71; a set of raw data is collected from an information source in a network communication environment; the network communication environment may include a system of interconnected sensors that transmit active data (i.e., are configured to continuously transmit data)], wherein the data stream comprises a set of feature values corresponding to an unlabeled instance [par. 85; note instance-based algorithms, cluster analysis (i.e., an unsupervised machine learning technique used with unlabeled data), and combinations thereof] of a feature set, wherein at least some of the feature values are captured by one or more sensors associated with the machine [pars. 75 and 76; the raw data is analyzed using a data interpretation dictionary to determine a set of attributes (i.e., a feature set), map the set of attributes to corresponding values (i.e., feature values) in the raw data (e.g., sensor measurements), and generate a set of interpreted data];
assign the data stream to a data stream group wherein the data stream group is selected from a set of data stream groups based on the set of feature values in the data stream [fig. 1; pars. 45, 67, 73, and 77; the network communication environment includes a group of information sources (i.e., set of data stream groups); the set of interpreted data is stored or otherwise maintained (i.e., grouped) based on the set of attributes];
select, from a set of predictive models, a predictive model corresponding to the data stream group assigned to the data stream [pars. 78, 79, and 82; an AI logic unit (e.g., a natural language processing technique, image analysis technique, predictive analytics, statistical analysis, prescriptive analytics, market modeling, web analytics, security analytics, risk analytics, software analytics, and the like) is selected from available AI logic units based on different sets of attributes], wherein the predictive models are trained to predict a target variable for different data stream groups in the set of data stream groups based on different subsets of training data streams in a training dataset [pars. 67, 73, 78-80, and 84-86; the AI logic units are trained to perform different processing operations (e.g., to extract a conclusion or inference) on sets of data associated with different sets of attributes (i.e., different data stream groups); subsets (i.e., subsets of training data streams) of the set of interpreted data (i.e., a training dataset) are used as training data for the AI logic units], wherein the training data streams comprise feature values corresponding to the same feature set as the data stream [pars. 73, 78, 79, and 84-86; note that the subsets are all subsets of the set of interpreted data and thus have at least one attribute in common (e.g., an information source identifier)], and wherein each predictive model is trained to predict the target variable for a corresponding data stream group based on a corresponding subset of the training data streams [pars. 73, 78-80, and 84-86; each AI logic unit is trained to perform a processing operation on a set of data received from an information source and associated with a set of attributes based on a subset of the set of interpreted data identified using machine learning usability criteria associated with the AI logic unit (i.e., corresponding subset of the training data streams)]; and
predict the target variable for the data stream using the predictive model, wherein the predictive model infers the target variable based on the set of feature values in the data stream [par. 80; the selected AI logic unit processes the set of interpreted data (e.g., to extract a conclusion or inference from the set of interpreted data) based on the set of attributes].
Referring to claim 2, see the rejection for claim 1.
Referring to claim 9, see at least the rejection for claim 1. Tabuchi further discloses At least one non-transitory computer-readable storage medium having instructions stored thereon, wherein the instructions, when executed on processing circuitry, cause the processing circuitry to perform the claimed steps [fig. 3, computer system 300, memory 304, data orchestration platform management 350, processor 302].
Referring to claim 10, see the rejection for claim 9.
Referring to claim 20, see the rejection for claim 1, which incorporates the claimed method.
Referring to claim 21, see the rejection for claim 20.
Referring to claim 23, see at least the rejection for claim 1. Tabuchi further discloses A system, comprising: one or more sensors [par. 66; interconnected sensors] interface circuitry; and processing circuitry to perform the claimed steps [fig. 3, computer system 300, network interface 318, processor 302].
Claim Rejections - 35 USC § 103
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.
Claims 3-7, 11-17, 22, and 26 are rejected under 35 U.S.C. 103 as being unpatentable over Tabuchi in view of DeLand et al. (US Pub. 20170124178).
Referring to claim 3, Tabuchi discloses The computing device of Claim 1, wherein the processing circuitry to assign the data stream to the data stream group is further to: select the data stream group to assign to the data stream using a grouping model [par. 77; the grouping according to attributes is performed including a machine learning model].
Tabuchi does not appear to explicitly disclose wherein the grouping model selects the data stream group from the set of data stream groups based on a comparison of the data stream to a grouping dataset, wherein the grouping dataset comprises a set of representative data streams for each data stream group in the set of data stream groups.
However, DeLand discloses wherein the grouping model selects the data stream group from the set of data stream groups based on a comparison of the data stream to a grouping dataset, wherein the grouping dataset comprises a set of representative data streams for each data stream group in the set of data stream groups [fig. 2, steps 205-215; pars. 21, 22, and 33-35; a clustering algorithm (i.e., grouping model) assigns new streaming data to a cluster (i.e., data stream group) of a model core group based on its feature data by comparing the feature data to data representing each cluster].
It would have been obvious to one of ordinary skill in the art before the effective filing date of the claimed invention to modify the grouping according to attributes taught by Tabuchi so that the grouping is performed by the clustering algorithm taught by DeLand, with a reasonable expectation of success. The motivation for doing so would have been to provide improved clustering for multi-dimensional dynamic data [DeLand, pars. 22 and 23].
Referring to claim 4, DeLand discloses The computing device of Claim 3, wherein the processing circuitry to select the data stream group to assign to the data stream using the grouping model is further to: compute, based on a distance calculation, a distance of the data stream to each data stream group in the set of data stream groups, wherein the distance to each data stream group is computed based on the set of representative data streams for each data stream group; and select, from the set of data stream groups, the data stream group having a closest distance to the data stream [par. 21; the clustering algorithm is a k-means algorithm that assigns objects to a cluster determined to be the nearest to the object based on comparing the Euclidean distances along one or more data dimensions between the data representing the object and the data representing the cluster].
Referring to claim 5, DeLand discloses The computing device of Claim 3, wherein the grouping model comprises a clustering model [par. 21; note the clustering algorithm].
Referring to claim 6, Tabuchi and DeLand disclose The computing device of Claim 3, wherein the processing circuitry is further to: select the set of representative data streams for each data stream group from the training dataset, wherein the training dataset comprises a set of training data streams; and generate the grouping dataset for the grouping model, wherein the grouping dataset comprises the set of representative data streams selected for each data stream group [Tabuchi: pars. 73, 78, 79, and 84-86; note the subsets of interpreted data used to respectively train the AI logic units corresponding to different attributes / DeLand: fig. 2, steps 205-215; pars. 21, 22, and 33-35; note the assigning of the new streaming data to a cluster (i.e., data stream group) of the model core group based on its feature data by comparing the feature data to data representing each cluster].
Referring to claim 7, Tabuchi does not appear to explicitly disclose The computing device of Claim 1, wherein the processing circuitry is further to: detect a change in the set of feature values in the data stream; determine, based on the change in the set of feature values, that a grouping of the data stream is to be updated, wherein the data stream is to be reassigned to a second data stream group in the set of data stream groups; and dynamically update the set of data stream groups to reassign the data stream to the second data stream group.
However, DeLand discloses The computing device of Claim 1, wherein the processing circuitry is further to: detect a change in the set of feature values in the data stream; determine, based on the change in the set of feature values, that a grouping of the data stream is to be updated, wherein the data stream is to be reassigned to a second data stream group in the set of data stream groups; and dynamically update the set of data stream groups to reassign the data stream to the second data stream group [fig. 2, step 220; pars. 20-22, and 33-35; a clustering algorithm (i.e., grouping model) assigns new streaming data to a cluster (i.e., data stream group) of a model core group based on its feature data by comparing the feature data to data representing each cluster; the model core group is dynamically updated in response to data changes in the new streaming data].
It would have been obvious to one of ordinary skill in the art before the effective filing date of the claimed invention to modify the grouping according to attributes taught by Tabuchi so that the grouping is performed by the clustering algorithm taught by DeLand, with a reasonable expectation of success. The motivation for doing so would have been to provide improved clustering for multi-dimensional dynamic data [DeLand, pars. 22 and 23].
Referring to claim 11, see the rejection for claim 3.
Referring to claim 12, see the rejection for claim 4.
Referring to claim 13, DeLand discloses The storage medium of Claim 12, wherein the distance calculation comprises: a Euclidean distance calculation; a Jaccard calculation; or a dynamic time warping calculation [par. 21; note the Euclidean distances].
Referring to claim 14, see the rejection for claim 5.
Referring to claim 15, DeLand discloses The storage medium of Claim 14, wherein the clustering model comprises a k-means clustering model [par. 21; note the k-means algorithm].
Referring to claim 16, see the rejection for claim 6.
Referring to claim 17, see the rejection for claim 7.
Referring to claim 22, see the rejection for claim 7.
Referring to claim 26, Tabuchi does not appear to explicitly disclose The computing device of Claim 1, wherein the processing circuitry to assign the data stream to the data stream group is further to: select the data stream group from the set of data stream groups based on a similarity between the data stream and one or more representative data streams in each data stream group, wherein the representative data streams are based on the same feature set as the data stream.
However, DeLand discloses The computing device of Claim 1, wherein the processing circuitry to assign the data stream to the data stream group is further to: select the data stream group from the set of data stream groups based on a similarity between the data stream and one or more representative data streams in each data stream group, wherein the representative data streams are based on the same feature set as the data stream [fig. 2, steps 205-215; pars. 21, 22, and 33-35; a clustering algorithm (i.e., grouping model) assigns new streaming data to a cluster (i.e., data stream group) of a model core group based on its feature data by comparing the feature data to data representing each cluster; the clustering algorithm is a k-means algorithm that assigns objects to a cluster determined to be the nearest to the object based on comparing the Euclidean distances along one or more data dimensions between the data representing the object and the data representing the cluster; see also Tabuchi, pars. 73, 78, 79, and 84-86 disclosing training of respective AI logic units using subsets from the same set of interpreted data].
It would have been obvious to one of ordinary skill in the art before the effective filing date of the claimed invention to modify the grouping according to attributes taught by Tabuchi so that the grouping is performed by the clustering algorithm taught by DeLand, with a reasonable expectation of success. The motivation for doing so would have been to provide improved clustering for multi-dimensional dynamic data [DeLand, pars. 22 and 23].
Claims 8, 18, 19, 24, 25, and 27 are rejected under 35 U.S.C. 103 as being unpatentable over Tabuchi in view of Wynne et al. (US Pub. 20200103857).
Referring to claim 8, Tabuchi discloses The computing device of Claim 1, wherein: the computing device is: an edge server [par. 110; note edge servers]; a tool controller to control a tool; or a robot controller to control a robot.
Tabuchi does not appear to explicitly disclose wherein the computing device is a tool controller to control a tool; or a robot controller to control a robot; and the target variable comprises a predicted quality level of a task performed by the tool or the robot.
However, Wynne discloses wherein the computing device is a tool controller to control a tool; or a robot controller to control a robot; and the target variable comprises a predicted quality level of a task performed by the tool or the robot [pars. 19, 20, 26, 30, and 87; a manufacturing production line includes equipment such as a CNC mill, industrial robots, conveyor systems, and printers, each of which includes sensors; a machine learning module uses AI to process and analyze device data to perform various prediction tasks such as detecting hardware failures, manufacturing quality issues, and production inefficiencies in the manufacturing production line].
It would have been obvious to one of ordinary skill in the art before the effective filing date of the claimed invention to modify the processing performed by the AI logic units taught by Tabuchi so that the processing includes the various prediction tasks taught by Wynne, with a reasonable expectation of success. The motivation for doing so would have been to monitor and analyze sensor feedback in a manufacturing production line to optimize real-time and future production and improve production and business decisions [Wynne, par. 17].
Referring to claim 18, see the rejection for claim 8.
Referring to claim 19, Wynne discloses The storage medium of Claim 18, wherein the task comprises a manufacturing task performed to manufacture a product [pars. 19, 20, 26, 30, and 87; note the various prediction tasks such as detecting hardware failures, manufacturing quality issues, and production inefficiencies in the manufacturing production line].
Referring to claim 24, Tabuchi does not appear to explicitly disclose The system of Claim 23, wherein: the machine is a tool or a robot; and the target variable comprises a predicted quality level of a task performed by the tool or the robot.
However, Wynne discloses The system of Claim 23, wherein: the machine is a tool or a robot; and the target variable comprises a predicted quality level of a task performed by the tool or the robot [pars. 19, 20, 26, 30, and 87; a manufacturing production line includes equipment such as a CNC mill, industrial robots, conveyor systems, and printers, each of which includes sensors; a machine learning module uses AI to process and analyze device data to perform various prediction tasks such as detecting hardware failures, manufacturing quality issues, and production inefficiencies in the manufacturing production line].
It would have been obvious to one of ordinary skill in the art before the effective filing date of the claimed invention to modify the processing performed by the AI logic units taught by Tabuchi so that the processing includes the various prediction tasks taught by Wynne, with a reasonable expectation of success. The motivation for doing so would have been to monitor and analyze sensor feedback in a manufacturing production line to optimize real-time and future production and improve production and business decisions [Wynne, par. 17].
Referring to claim 25, Wynne discloses The system of Claim 24, wherein the tool is: a welding gun; a glue gun; a riveting machine; a screwdriver; or a pump [pars. 68 and 95; the manufacturing production line includes pumps].
Referring to claim 27, Tabuchi does not appear to explicitly disclose The computing device of Claim 1, wherein the machine comprises (i) a robot, (ii) a tool, or (iii) a robot and a tool.
However, Wynne discloses The computing device of Claim 1, wherein the machine comprises (i) a robot, (ii) a tool, or (iii) a robot and a tool [pars. 19, 20, 26, 30, and 87; a manufacturing production line includes equipment such as a CNC mill, industrial robots, conveyor systems, and printers, each of which includes sensors; a machine learning module uses AI to process and analyze device data to perform various prediction tasks such as detecting hardware failures, manufacturing quality issues, and production inefficiencies in the manufacturing production line].
It would have been obvious to one of ordinary skill in the art before the effective filing date of the claimed invention to modify the processing performed by the AI logic units taught by Tabuchi so that the processing includes the various prediction tasks taught by Wynne, with a reasonable expectation of success. The motivation for doing so would have been to monitor and analyze sensor feedback in a manufacturing production line to optimize real-time and future production and improve production and business decisions [Wynne, par. 17].
Conclusion
THIS ACTION IS MADE FINAL. Applicant is reminded of the extension of time policy as set forth in 37 CFR 1.136(a).
A shortened statutory period for reply to this final action is set to expire THREE MONTHS from the mailing date of this action. In the event a first reply is filed within TWO MONTHS of the mailing date of this final action and the advisory action is not mailed until after the end of the THREE-MONTH shortened statutory period, then the shortened statutory period will expire on the date the advisory action is mailed, and any nonprovisional extension fee (37 CFR 1.17(a)) pursuant to 37 CFR 1.136(a) will be calculated from the mailing date of the advisory action. In no event, however, will the statutory period for reply expire later than SIX MONTHS from the mailing date of this final action.
Contact Information
Any inquiry concerning this communication or earlier communications from the examiner should be directed to GRACE PARK whose telephone number is (571)270-7727. The examiner can normally be reached M-F 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, TAMARA KYLE can be reached at (571)272-4241. 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.
/Grace Park/Primary Examiner, Art Unit 2144
Prosecution Insights