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 1-20 are rejected under 35 U.S.C. 101 because the claimed invention is directed to an abstract idea without significantly more.
Claims 1, 10, and 19 recite:
generating a graph of the computing environment, the graph including nodes representing managed objects in the computing environment and guideline objects associated the managed objects;
applying a transductive embedding technique on the graph to generate initial embeddings for the nodes of the graph;
applying an inductive embedding technique on the initial embeddings and features of the nodes of the graph to produce final embeddings for the nodes of the graph;
executing a link classification operation on the final embeddings for at least some nodes of the graph to select a recommended guideline for a target managed object; and
displaying the recommended guideline for the target managed object.
Step 1: Is the claim to a process, machine, manufacture, or composition of matter?
Yes.
Claim 1 is a process.
Claims 10 and 19 are manufactures.
Step 2A, Prong I: Does the claim recite an abstract idea, law of nature, or natural phenomenon?
Yes: (an) abstract idea(s).
The ‘generating’ limitation in #1 above, as claimed and under broadest reasonable interpretation (BRI), is a mental process that covers performance of the limitation in the mind. The limitation “generating” in the context of this claim encompasses a person analyzing, evaluating, or determining a graph including nodes representing managed objects and guideline objects, including comparison or judgement.
The ‘applying’ limitation in #2 above, as claimed and under broadest reasonable interpretation (BRI), is a mental process that covers performance of the limitation in the mind. The limitation “applying” in the context of this claim encompasses a person analyzing, evaluating, or determining a transductive embedding technique on the graph to generate initial embeddings for the nodes, including comparison or judgement.
The ‘applying’ limitation in #3 above, as claimed and under broadest reasonable interpretation (BRI), is a mental process that covers performance of the limitation in the mind. The limitation “applying” in the context of this claim encompasses a person analyzing, evaluating, or determining a inductive embedding technique on the graph to generate final embeddings for the nodes, including comparison or judgement.
Step 2A, Prong II: Does the claim recite additional elements that integrate the judicial exception into a practical application?
No.
The ‘executing’ limitation in #4 above, as claimed and under broadest reasonable interpretation (BRI), is an additional element as “apply it” that is mere instructions to apply an exception. The limitation “executing” in the context of this claim encompasses merely executing a link classification operation to select a recommended guideline for the target managed object. See MPEP 2106.05(f).
The ‘displaying’ limitation in #5 above, as claimed and under broadest reasonable interpretation (BRI), is an additional element that is insignificant extra-solution activity. The limitation “displaying” in the context of this claim encompasses merely displaying information to a user. See MPEP 2106.05(g).
Additionally, one or more of the claims recite the following additional elements:
Program instructions (Claim 10)
One or more processors (Claim 10)
Memory (Claim 19)
At least one processor (Claim 19)
These additional elements are recited at a high level of generality (i.e., as generic computer components) such that they amount to no more than components comprising mere instructions to apply the exception. Accordingly, these additional elements do not integrate the abstract idea(s) into a practical application because they do not impose any meaningful limits on practicing the abstract ideas(s).
Step 2B: Does the claim recite additional elements that amount to significantly more than the judicial exception?
No.
As discussed above with respect to integration of the abstract idea(s) into a practical application, the aforementioned additional elements amount to no more than components for obtaining or gathering data and comprising mere instructions to apply the exception which is evidently seen in MPEP 2106.05(g)&(f). Mere instructions to apply an exception using generic computer components cannot provide an inventive concept.
Additionally, with regards to #5 above, per MPER 2106.05(d)(II), the courts have recognized the following computer functions as well-understood, routine, and conventional functions when they are claimed in a merely generic matter (e.g., at a high level of generality) or as insignificant extra-solution activity:
Storing and retrieving information in memory, Versata Dev. Group, Inc. v. SAP Am., Inc., 793 F.3d 1306, 1334, 115 USPQ2d 1681, 1701 (Fed. Cir. 2015); OIP Techs., 788 F.3d at 1363, 115 USPQ2d at 1092-93;
With regards to Claims 10 and 19, the method of Claim 1 performs the same steps as the manufactures of Claims 10 and 19, and Claims 10 and 19 are therefore rejected using the same rationale set forth above in the rejection of Claim 1.
Claims 2 and 11 merely further describe the target managed object and final embeddings of Claims 1 and 10 respectively. The claims do not include additional elements that integrate into practical application or are sufficient to amount to significantly more than the judicial exception.
Claims 3 and 12 merely further describe the target managed object and final embeddings of Claims 1 and 10 respectively. The claims do not include additional elements that integrate into practical application or are sufficient to amount to significantly more than the judicial exception.
Claims 4 and 13 merely further describe the updated version of the previously generated final embedding of Claims 3 and 12 respectively. The claims do not include additional elements that integrate into practical application or are sufficient to amount to significantly more than the judicial exception.
Claims 9 and 18 merely further describe the nodes of the graph of Claims 1 and 10 respectively. The claims do not include additional elements that integrate into practical application or are sufficient to amount to significantly more than the judicial exception.
With regards to Claim 20, the method of Claims 3-4 performs the same steps as the manufacture of Claim 20, and Claim 20 is therefore rejected using the same rationale set forth above in the rejection of Claims 3-4.
Therefore, Claims 1-4, 9-13 and 18-20 are directed to (an) abstract idea(s) without significantly more.
Claims 5 and 14 recite:
applying an incremental transductive embedding technique on the node corresponding to the new managed object to produce the transductive embedding for the new managed object.
Step 1: Is the claim to a process, machine, manufacture, or composition of matter?
Yes.
Claim 5 is a process.
Claim 14 is a manufacture.
Step 2A, Prong I: Does the claim recite an abstract idea, law of nature, or natural phenomenon?
Yes: (an) abstract idea(s).
The ‘applying’ limitation in #6 above, as claimed and under broadest reasonable interpretation (BRI), is a mental process that covers performance of the limitation in the mind. The limitation “applying” in the context of this claim encompasses a person analyzing, evaluating, or determining an incremental transductive embedding technique on the graph to produce the transductive embeddings for the new object, including comparison or judgement.
Step 2B: Does the claim recite additional elements that amount to significantly more than the judicial exception?
No.
As discussed above with respect to integration of the abstract idea(s) into a practical application, the aforementioned additional elements amount to no more than components comprising mere instructions to apply the exception which is evidently seen in MPEP 2106.05(f). Mere instructions to apply an exception using generic computer components cannot provide an inventive concept.
With regards to Claim 14, the method of Claim 5 performs the same steps as the manufacture of Claim 14, and Claim 14 is therefore rejected using the same rationale set forth above in the rejection of Claim 5.
Claims 6 and 15 merely further describe the incremental transductive embedding technique of Claims 5 and 14 respectively. The claims do not include additional elements that integrate into practical application or are sufficient to amount to significantly more than the judicial exception.
Claims 7 and 16 merely further describe the incremental transductive embedding technique of Claims 5 and 14 respectively. The claims do not include additional elements that integrate into practical application or are sufficient to amount to significantly more than the judicial exception.
Therefore, Claims 5-7 and 14-16 are directed to (an) abstract idea(s) without significantly more.
Claims 8 and 17 recite:
wherein executing the link classification operation on the final embeddings includes
producing link scores for select guideline nodes for the target managed object, and wherein each link score corresponds to a strength of link between a particular guideline node and the target managed object.
Step 1: Is the claim to a process, machine, manufacture, or composition of matter?
Yes.
Claim 8 is a process.
Claim 17 is a manufacture.
Step 2A, Prong I: Does the claim recite an abstract idea, law of nature, or natural phenomenon?
Yes: (an) abstract idea(s).
The ‘producing’ limitation in #7 above, as claimed and under broadest reasonable interpretation (BRI), is a mental process that covers performance of the limitation in the mind. The limitation “producing” in the context of this claim encompasses a person analyzing, evaluating, or determining a link score for select guideline nodes for the target managed object, including comparison or judgement.
Step 2B: Does the claim recite additional elements that amount to significantly more than the judicial exception?
No.
As discussed above with respect to integration of the abstract idea(s) into a practical application, the aforementioned additional elements amount to no more than components comprising mere instructions to apply the exception which is evidently seen in MPEP 2106.05(f). Mere instructions to apply an exception using generic computer components cannot provide an inventive concept.
With regards to Claim 17, the method of Claim 8 performs the same steps as the manufacture of Claim 17, and Claim 17 is therefore rejected using the same rationale set forth above in the rejection of Claim 8.
Therefore, Claims 8 and 17 are directed to (an) abstract idea(s) without significantly more.
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 1-2, 8-11, and 17-19 are rejected under 35 U.S.C. 103 as being unpatentable over Shahul Hameed et al. (U.S. Patent No. US 20230275912 A1), hereinafter “Shahul Hameed” in view of Symons et al. (U.S. Patent No. US 20150067857 A1), hereinafter “Symons.”
With regards to Claim 1, Shahul Hameed teaches:
A computer-implemented method for recommending guidelines for managed objects for a computing environment (Paragraphs 27-28, “In the (multipartite) graph, the different types of entities (e.g., machines, processes, and network destinations) are represented as different types of nodes, the relationships between entities are represented as edges between the nodes, and the feature vectors are associated with the nodes or edges to which they pertain as attributes… If feature data is extracted at 304, the clustering is based in part on the feature vectors assigned to the nodes and/or edges. The clusters are ranked, at 310, according to some metric of maliciousness, e.g., based on the number of security alerts associated with each cluster, the number of IoCs associated with each cluster, or a combination of both. An output generated from one or more highest-ranking clusters (i.e., clusters having the greatest associated number of security alerts or IoCs), e.g., an output listing of the nodes, or a subset of the nodes (e.g., the nodes with highest connectivity), within the highest-ranking cluster(s) is provided at 312.” The multipartite graph representing different types of entities as nodes corresponds to the managed objects for a computer environment. The feature vectors which are associated with the nodes or edges to which they pertain as attributes and used for clustering corresponds to recommending guidelines for managed objects in the computer environment), the method comprising:
generating a graph of the computing environment, the graph including nodes representing managed objects in the computing environment and guideline objects associated the managed objects (Paragraphs 27-28, “At 306, a graph (e.g., a multipartite graph) is generated from the extracted node and edge data. In the (multipartite) graph, the different types of entities (e.g., machines, processes, and network destinations) are represented as different types of nodes, the relationships between entities are represented as edges between the nodes, and the feature vectors are associated with the nodes or edges to which they pertain as attributes… If feature data is extracted at 304, the clustering is based in part on the feature vectors assigned to the nodes and/or edges. The clusters are ranked, at 310, according to some metric of maliciousness, e.g., based on the number of security alerts associated with each cluster, the number of IoCs associated with each cluster, or a combination of both. An output generated from one or more highest-ranking clusters (i.e., clusters having the greatest associated number of security alerts or IoCs), e.g., an output listing of the nodes, or a subset of the nodes (e.g., the nodes with highest connectivity), within the highest-ranking cluster(s) is provided at 312.” The multipartite graph generated with entities represented as different types of nodes corresponds to generating a graph including nodes representing managed objects in the computing environment. The feature vectors which are associated with the nodes or edges to which they pertain as attributes and used for clustering corresponds to guideline objects associated with the managed objects);
applying an inductive embedding technique on the initial embeddings and features of the nodes of the graph to produce final embeddings for the nodes of the graph (Paragraphs 23 and 28, “In some embodiments, unsupervised GraphSAGE is utilized to compute the node embeddings for k-means clustering… Beneficially, GraphSAGE is inductive in that embeddings for new nodes (unseen during training) can be computed without having to retrain the model… supervised or semi-supervised techniques may also be used in some embodiments.” The use of unsupervised GraphSAGE to compute the node embeddings for k-means clustering of the graph corresponds to applying inductive embedding techniques on the initial embeddings and features of the nodes of the graph to produce final embeddings);
executing a link classification operation on the final embeddings for at least some nodes of the graph to select a recommended guideline for a target managed object (Paragraph 28, “The (multipartite) graphs is clustered at 308, using a graph-based clustering technique… If feature data is extracted at 304, the clustering is based in part on the feature vectors assigned to the nodes and/or edges. The clusters are ranked, at 310, according to some metric of maliciousness, e.g., based on the number of security alerts associated with each cluster, the number of IoCs associated with each cluster, or a combination of both. An output generated from one or more highest-ranking clusters (i.e., clusters having the greatest associated number of security alerts or IoCs), e.g., an output listing of the nodes, or a subset of the nodes (e.g., the nodes with highest connectivity), within the highest-ranking cluster(s) is provided at 312.” The graph being clustered based on the feature vectors assigned to the nodes or edges and ranked based on metrics to select the highest-ranking clusters correlates to executing a link classification operation on the final embeddings for at least some nodes of the graph to select a recommended guideline for a target managed object); and
displaying the recommended guideline for the target managed object (Paragraphs 28-29, “If feature data is extracted at 304, the clustering is based in part on the feature vectors assigned to the nodes and/or edges. The clusters are ranked, at 310, according to some metric of maliciousness, e.g., based on the number of security alerts associated with each cluster, the number of IoCs associated with each cluster, or a combination of both. An output generated from one or more highest-ranking clusters (i.e., clusters having the greatest associated number of security alerts or IoCs), e.g., an output listing of the nodes, or a subset of the nodes (e.g., the nodes with highest connectivity), within the highest-ranking cluster(s) is provided at 312. The output may be further processed to obtain relevant security information or mitigate the incident at 314. For example, the listed nodes may be reviewed by a security analysist, or further analyzed programmatically by a computer, to identify one or more IoCs among them.” The listing of the highest-ranking clusters being outputted where a security analyst may review them correlates to displaying the recommended guideline for the target managed object).
Shahul Hameed does not explicitly teach:
applying a transductive embedding technique on the graph to generate initial embeddings for the nodes of the graph;
However, Symons teaches:
applying a transductive embedding technique on the graph to generate initial embeddings for the nodes of the graph (Paragraph 22, “For some semi-supervised test models, dimensionality is reduced and an initial transductive model is constructed.” The initial transductive model constructed for the graph correlates to applying a transductive embedding technique on the graph to generate initial embeddings for nodes of the graph);
Therefore, it would have been obvious to one of ordinary skill in the art to which said subject matter pertains before the effective filing date of the claimed invention to combine Shahul Hameed with applying a transductive embedding technique on the graph to generate initial embeddings for the nodes of the graph as taught by Symons because reducing the dimensionality by applying an initial transductive embedding to the graph allows discovering smooth dimensions with respect to the graph, where examples from different classes or clusters are rarely linked. Therefore, the points connected in the graph will be close together in the defined dimension (Symons: paragraph 20).
With regards to Claims 10 and 19, the method of Claim 1 performs the same steps as the manufacture and system of Claims 10 and 19 respectively, and Claims 10 and 19 are therefore rejected using the same rationale set forth above in the rejection of Claim 1.
With regards to Claim 2, Shahul Hameed in view of Symons teaches the method of Claim 1 above. Shahul Hameed further teaches:
wherein the target managed object is an existing managed object in the computing environment (Paragraph 7, “generate a multipartite graph that represents entities within or connected to the computer network (herein collectively “network entities”)-such as machines within the computer network, processes spawned by the machines, and external network destinations connected to by the processes--as different types of nodes and the relations between them as edges between the nodes.” The multipartite graph being generated to represent entities within or connected to the computer network as nodes corresponds to the target managed object already existing in the computing environment) and wherein the final embeddings are not an updated version of previously generated final embeddings that are generated in response to a request for a guideline recommendation for the existing managed object (Paragraph 7, “generate a multipartite graph that represents entities within or connected to the computer network (herein collectively “network entities”)-such as machines within the computer network, processes spawned by the machines, and external network destinations connected to by the processes--as different types of nodes and the relations between them as edges between the nodes. Using one or more graph-based clustering techniques, the multipartite graph is then broken up into clusters.” The multipartite graph being generated is based on the entities and relations between them and then clustered results in the final nodes not being an updated version of previously generated embeddings and therefore correlates to the final embeddings not being an updated version of previously generated final embeddings that were generated in response to a request for a guideline recommendation).
With regards to Claim 11, the method of Claim 2 performs the same steps as the manufacture of Claim 11, and Claim 11 is therefore rejected using the same rationale set forth above in the rejection of Claim 2.
With regards to Claim 8, Shahul Hameed in view of Symons teaches the method of Claim 1 above. Shahul Hameed further teaches:
wherein executing the link classification operation on the final embeddings includes producing link scores for select guideline nodes for the target managed object (Paragraph 7, “Using one or more graph-based clustering techniques, the multipartite graph is then broken up into clusters, which can be ranked by some metric quantifying the severity of the threat, such as based on the number of security alerts or indicators of compromise (IoCs) associated with each cluster.” The ranking of clusters through a metric that quantifies the severity of the threat correlates to producing link scores for select guideline nodes for the target managed object), and wherein each link score corresponds to a strength of link between a particular guideline node and the target managed object (Paragraph 7, “Using one or more graph-based clustering techniques, the multipartite graph is then broken up into clusters, which can be ranked by some metric quantifying the severity of the threat, such as based on the number of security alerts or indicators of compromise (IoCs) associated with each cluster.” The ranking of clusters through a metric that quantifies the severity of the threat associated with each cluster correlates to link scores corresponding to a strength of link between a particular guideline node and the target managed object).
With regards to Claim 17, the method of Claim 8 performs the same steps as the manufacture of Claim 17, and Claim 17 is therefore rejected using the same rationale set forth above in the rejection of Claim 8.
With regards to Claim 9, Shahul Hameed in view of Symons teaches the method of Claim 1 above. Shahul Hameed further teaches:
wherein the nodes of the graph represent a policy, a tag, a tag category (Paragraph 7, “Using one or more graph-based clustering techniques, the multipartite graph is then broken up into clusters, which can be ranked by some metric quantifying the severity of the threat, such as based on the number of security alerts or indicators of compromise (IoCs) associated with each cluster… An IoC, as used herein, is any malicious domain, IP address, process, machine, or other network entity that serves as evidence that a breach has occurred and the computer system is compromised. Examples of IoCs include, without limitation, virus signatures, URLs or domains of C2 servers in communication with the infected machines of the monitored network, and file hashes of malware files.” The clusters of the multipartite graph being ranked and organized based on IoCs, which include network entities serving as evidence of a breach correlates to the nodes of the graph representing a policy, tag, or tag category), a data center (Paragraph 7, “In various embodiments, graph-based incident analysis involves processing data collected for the computer network of an organization under attack within the time period between start and end points determined for the attack (or, herein synonymously, security incident) to generate a multipartite graph that represents entities within or connected to the computer network (herein collectively “network entities”)-such as machines within the computer network, processes spawned by the machines, and external network destinations connected to by the processes--as different types of nodes and the relations between them as edges between the nodes.” The different types of nodes in the graph which include machines within a computer network correlates to nodes of the graph representing data centers), virtual machines (Paragraphs 17-18, “Further, in some embodiments, metadata, or “feature data” 206 associated with the various entities and/or the relationships between them (204) is also gathered and stored in the data tables 206… Features gathered for a machine may include, e.g., an alert history, the machine type (e.g., server, workstation), and/or the number of users that regularly connect to the machine… From the data tables 205 of entities, relationships, and optionally associated features, a graph 208 is generated for the incident. The graph represents entities as nodes, relationships between the nodes as edges, and any features associated with the entities or relationships as node or edge attributes.” The graph representing features such as the machine type as nodes correlates to the nodes of the graph representing virtual machines), a folder (Paragraphs 17-18, “Further, in some embodiments, metadata, or “feature data” 206 associated with the various entities and/or the relationships between them (204) is also gathered and stored in the data tables 206. Features collected for a process may include, e.g., the process name, the age of the process (measured from the time of process creation), the image file age of the process (measured from the time the process image was stored on disk), the compilation age of the process, the file signature status, the process/file size, the type of process (e.g., browser, non-browser), and/or any command line features… From the data tables 205 of entities, relationships, and optionally associated features, a graph 208 is generated for the incident. The graph represents entities as nodes, relationships between the nodes as edges, and any features associated with the entities or relationships as node or edge attributes.” The graph representing features such as file signature, type of process, or command line features associated with the entities as nodes correlates to the nodes of the graph representing folders),
Shahul Hameed does not explicitly teach:
wherein the nodes of the graph represent hosts and a cluster of hosts
However, Symons teaches:
wherein the nodes of the graph represent hosts and a cluster of hosts (Paragraph 15, claim 6, “The data object operated on by each component in this disclosure is a feature set, which is a data representation in machine learning. Network data/packet data/host data is acquired, and flow, host, and network features are derived from this data as the base feature set… storing the feature set comprising network data, packet data, and host data at the node it is received.” The network, packet and host data being stored at the node corresponds to the nodes of the graph representing hosts or a cluster of hosts).
Therefore, it would have been obvious to one of ordinary skill in the art to which said subject matter pertains before the effective filing date of the claimed invention to combine Shahul Hameed with wherein the nodes of the graph represent hosts and a cluster of hosts as taught by Symons because feature sets which include host and network data can be augmented by components of stages in the pipeline to optimize the number and type of features for the target operational environment to build a generalized model (Symons: paragraph 15).
With regards to Claim 18, the method of Claim 9 performs the same steps as the manufacture of Claim 18, and Claim 18 is therefore rejected using the same rationale set forth above in the rejection of Claim 9.
Claims 3-7, 12-16, and 20 are rejected under 35 U.S.C. 103 as being unpatentable over Shahul Hameed in view of Symons and Shang et al. (CN Patent No. CN 113836359 A), hereinafter “Shang.”
With regards to Claim 3, Shahul Hameed in view of Symons teaches the method of Claim 1 above. Shahul Hameed further teaches:
wherein the target managed object is a new managed object being added to the computing environment (Paragraph 9, “Additionally, the multipartite graph can be readily extended by additional entities, relationships, and/or associated attributes, allowing new modes of information to be added and removed efficiently and at will (without any need for model training) to explore the best set of information to be utilized for a specific problem.” The multipartite graph being readily extended by additional entities to allow new modes of information to be added correlates to the target managed object being a new managed object added to the computing environment)
Shahul Hameed does not explicitly teach:
and wherein the final embeddings are an updated version of previously generated final embeddings that are generated in response to a request for a guideline recommendation for the new managed object
However, Shang teaches:
and wherein the final embeddings are an updated version of previously generated final embeddings that are generated in response to a request for the new managed object (Paragraph 59, “the new node is added in the dynamic graph to perform dynamic image embedding calculation.” The new node added to the dynamic graph which causes dynamic image embedding calculations correlates to the final embeddings being an updated version of previously generated final embeddings generated in response to the new managed object).
Shang does not explicitly teach that the final embeddings are generated in response to a request for a guideline recommendation for the new managed object. However, requests for a guideline recommendation are a popular query related to adding new managed objects as evidenced by Shahul Hameed above.
Therefore, it would have been obvious to one of ordinary skill in the art to which said subject matter pertains before the effective filing date of the claimed invention to combine Shahul Hameed with and wherein the final embeddings are an updated version of previously generated final embeddings that are generated in response to a request for the new managed object as taught by Shang because updating the dynamic graph reduces the calculation complexity of the model, can eliminate the negative contribution generated by the new node in the network, and improve the performance of the model (Shang: paragraph 60).
With regards to Claim 12, the method of Claim 3 performs the same steps as the manufacture of Claim 12, and Claim 12 is therefore rejected using the same rationale set forth above in the rejection of Claim 3.
With regards to Claim 4, Shahul Hameed in view of Symons and Shang teaches the method of Claim 3 above. Shang further teaches:
wherein the updated version of previously generated final embedding is generated using a embedding of a node corresponding to the new managed object (Paragraph 59, “the new node is added in the dynamic graph to perform dynamic image embedding calculation.” The new node added to the dynamic graph which causes dynamic image embedding calculations correlates to the updated final version of the embedding being generated using an embedding technique of a node corresponding to the new managed object).
Shang does not explicitly teach that the updated version of the final embeddings are generated using transductive embedding and that the embedding corresponds to a node feature for the new managed object. However, transductive embedding is a popular embedding technique as evidenced by Shahul Hameed (Paragraph 23, “In some embodiments, the k-means clustering algorithm operates on Node2Vec embeddings, which are generated by sampling a specified number of random walks of specified length through the graph, and then using the sampled random walks as input vectors to train a skip-gram neural network model. The weights across all nodes of the trained neural network that are associated with a given input vector component, corresponding to a given graph node, constitute the embedding for that node.” The k-means clustering algorithm operating on Node2Vec embeddings to constitute the embeddings for nodes correlates to using transductive embedding techniques). Additionally, embedding corresponding to a node feature for a new managed node is a popular concept as evidenced by Shahul Hameed above through inductive embedding techniques.
Therefore, it would have been obvious to one of ordinary skill in the art to which said subject matter pertains before the effective filing date of the claimed invention to combine Shahul Hameed with wherein the updated version of previously generated final embedding is generated using a embedding of a node corresponding to the new managed object as taught by Shang because updating the dynamic graph reduces the calculation complexity of the model, can eliminate the negative contribution generated by the new node in the network, and improve the performance of the model (Shang: paragraph 60).
With regards to Claim 13, the method of Claim 4 performs the same steps as the manufacture of Claim 13, and Claim 13 is therefore rejected using the same rationale set forth above in the rejection of Claim 4.
With regards to Claim 20, the methods of Claims 3 and 4 perform the same steps as the system of Claim 20, and Claim 20 is therefore rejected using the same rationale set forth above in the rejection of Claims 3 and 4.
With regards to Claim 5, Shahul Hameed in view of Symons and Shang teaches the method of Claim 4 above. Shang further teaches:
applying an incremental embedding technique on the node corresponding to the new managed object to produce the embedding for the new managed object (Fig. 2, paragraphs 59 and 62, “the new node is added in the dynamic graph to perform dynamic image embedding calculation… respectively the dynamic graph G at different time of the static graph, namely G1, G2, ..., Gt input model to calculate according to the side liveness of the last update period initial time calculating the edge activity of the target update period initial time, calculating the corresponding node activity according to the side activity and performing node embedding, continuously iterative calculation, finally obtaining low-dimensional vector representation of each node in the graph corresponding to Gt time, realizing dynamic image embedding.” The new node added to the dynamic graph which performs dynamic image embedding calculations in a continuously iterative calculation correlates to applying an incremental embedding technique on the node corresponding to the new object to produce the embeddings for the new managed object).
Shang does not explicitly teach that the incremental embedding technique is transductive. However, transductive embedding is a popular embedding technique as evidenced by Shahul Hameed above.
Therefore, it would have been obvious to one of ordinary skill in the art to which said subject matter pertains before the effective filing date of the claimed invention to combine Shahul Hameed with wherein the updated version of previously generated final embedding is generated using a embedding of a node corresponding to the new managed object as taught by Shang because updating the dynamic graph reduces the calculation complexity of the model, can eliminate the negative contribution generated by the new node in the network, and improve the performance of the model (Shang: paragraph 60).
With regards to Claim 14, the method of Claim 5 performs the same steps as the manufacture of Claim 14, and Claim 14 is therefore rejected using the same rationale set forth above in the rejection of Claim 5.
With regards to Claim 6, Shahul Hameed in view of Symons and Shang teaches the method of Claim 5 above. Shahul Hameed further teaches:
wherein applying the transductive embedding technique includes aggregating a mean of neighboring nodes of the new managed object (Paragraph 23, “In some embodiments, the k-means clustering algorithm operates on Node2Vec embeddings, which are generated by sampling a specified number of random walks of specified length through the graph, and then using the sampled random walks as input vectors to train a skip-gram neural network model. The weights across all nodes of the trained neural network that are associated with a given input vector component, corresponding to a given graph node, constitute the embedding for that node.” The k-means clustering algorithm operating on Node2Vec embeddings to constitute the embedding for nodes correlates to applying transductive embedding techniques which include aggregating a mean of neighboring nodes of the new managed object), wherein a size of the neighboring nodes is predefined (Paragraph 23, “In some embodiments, the k-means clustering algorithm operates on Node2Vec embeddings.” The value of ‘k’ in k-means clustering defines the size of the neighboring nodes and therefore correlates to the size of the neighboring nodes being predefined).
Shahul Hameed does not explicitly teach that the transductive embedding technique is incremental. However, incremental embedding is a popular embedding technique as evidenced by Shang above.
With regards to Claim 15, the method of Claim 6 performs the same steps as the manufacture of Claim 15, and Claim 15 is therefore rejected using the same rationale set forth above in the rejection of Claim 6.
With regards to Claim 7, Shahul Hameed in view of Symons and Shang teaches the method of Claim 5 above. Shahul Hameed further teaches:
wherein applying the transductive embedding technique includes producing random walks using the node of the new managed object (Paragraph 23, “In some embodiments, the k-means clustering algorithm operates on Node2Vec embeddings, which are generated by sampling a specified number of random walks of specified length through the graph, and then using the sampled random walks as input vectors to train a skip-gram neural network model. The weights across all nodes of the trained neural network that are associated with a given input vector component, corresponding to a given graph node, constitute the embedding for that node.” The k-means clustering algorithm operating on Node2Vec embeddings and using generated samples of random walks to constitute the embedding for nodes correlates to applying transductive embedding techniques which include producing random walks using the node of the new managed object) and running a skip-gram model on the random walks. (Paragraph 23, “In some embodiments, the k-means clustering algorithm operates on Node2Vec embeddings, which are generated by sampling a specified number of random walks of specified length through the graph, and then using the sampled random walks as input vectors to train a skip-gram neural network model.” The skip-gram model being trained on the random walks correlates to running a skip-gram model on the random walks).
Shahul Hameed does not explicitly teach that the transductive embedding technique is incremental. However, incremental embedding is a popular embedding technique as evidenced by Shang above.
With regards to Claim 16, the method of Claim 7 performs the same steps as the manufacture of Claim 16, and Claim 16 is therefore rejected using the same rationale set forth above in the rejection of Claim 7.
Prior Art Made of Record
The prior art made of record and not relied upon is considered pertinent to applicant’s disclosure.
Knowles et al. (U.S. Patent No. US 11294789 B2); teaching a method of capturing operational data within a computing environment containing a plurality of interrelated managed components. The operational data is dynamically filtered within the computing environment to identify event data, managed component relationship data, and temporal information. The event data and managed component relationship data is communicated to a remote service provider to perform analytics by transforming information into a temporal topology graph.
Conclusion
Any inquiry concerning this communication or earlier communications from the examiner should be directed to SELINA HU whose telephone number is (571)272-5428. The examiner can normally be reached Monday-Friday 8:30-5:30.
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, Chat Do can be reached at (571) 272-3721. 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.
/SELINA ELISA HU/ Examiner, Art Unit 2193
/Chat C Do/ Supervisory Patent Examiner, Art Unit 2193