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 .
Information Disclosure Statement
The information disclosure statements (IDSs) submitted on January 10, 2024, October 28, 2024, January 15, 2025 and May 22, 2026 are being considered by the examiner.
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 1-20 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.
Claim 1 states to obtain aggregated summaries and a related knowledge graph, and then to enable local, community, and global retrieval augmented generation utilizing the aggregated summaries and the knowledge graph. It is unclear how the obtained data is used to enable retrieval augmented generation. See MPEP 2173.05(q) - “Use” claims. Although the claims are interpreted in light of the specification, limitations from the specification are not read into the claims. See In re Van Geuns, 988 F.2d 1181, 26 USPQ2d 1057 (Fed. Cir. 1993).
Claim 2 is rejected under 35 U.S.C. 112(b) or 35 U.S.C. 112 (pre-AIA ), second paragraph, as being incomplete for omitting essential steps, such omission amounting to a gap between the steps. See MPEP § 2172.01. The claim further recites “aggregating edges between shared nodes and using frequency count as an edge weight of the knowledge graph.” It’s unclear how these features interrelate with the other features of the claim or the claim from which it depends. The omitted steps are: how do the added features of the claim impact the enabling? It does not appear to resolve the issue of how “utilizing the aggregated summaries and the knowledge graph” enable “local, community, and global retrieval augmented generation.”
Similarly, although claims 3-8 further define the manner in which the knowledge graph is processed, the claims do not appear to resolve how “utilizing the aggregated summaries and the knowledge graph” enable “local, community, and global retrieval augmented generation.”
Claim 9 has substantially similar limitations as claim 1 and is rejected for the same reasons.
Claim 10 recites “wherein the processor is configured to accomplish the enabling graph-based retrieval augmented generation by…,” but it is unclear how “utilizing the aggregated operations and the knowledge graph” enable “graph-based retrieval augmented generation.”
Claim 11 recites “determining whether the user query requires summarizations of an entirety of the private dataset;” “processing the user query utilizing knowledge graph retrieval augmented generation (RAG) with global summarization, ” “evaluating whether the user query relates to a particular entity of the private dataset;” “processing the user query utilizing knowledge graph RAG with local summarization, ” and “processing the user query utilizing knowledge graph RAG with community summarization.” Upon reviewing the claim and specification, it is unclear how the determination is made for determining whether the user query requires summarizations of an entirety of the private dataset and how the user query is processed “utilizing knowledge graph retrieval augmented generation (RAG) with global summarization.” Similarly, it is unclear how the evaluation occurs on “whether the user query relates to a particular entity of the private dataset” and how the user query is processed utilizing either “knowledge graph RAG with local summarization” or “knowledge graph RAG with local summarization.” In addition “the question” lacks antecedent basis. It’s unclear which “question” this is referring to and whether this should read “the query” or “a question.”
Claims 12-20 do not resolve the indefiniteness rejections of claim 11; therefore, they are rejected for the same reasons as claim 11.
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 1-14 and 17-20 are rejected under 35 U.S.C. §101 because the claimed invention is directed to an abstract idea without significantly more.
As per claims 1, The claims recite the abstract idea of knowledge graph extraction.
The following is an analysis based on 2019 Revised Patent Subject Matter Eligibility Guidance (2019 PEG).
Step 1, Statutory Category?
Claims 1-8 are directed to a method.
Claims 9-10 are directed to a system.
Claims 11-14 and 17-20 are directed to a computer-readable storage medium, which excludes signals according to the Specification at paragraph [0104].
Step 2A, Prong I: Judicial Exception Recited?
The examiner submits that the foregoing claim limitations constitute a “mental process”, as the claims cover performance of the limitations in the human mind, given the broadest reasonable interpretation.
As per claim 1, the claim recites the limitations of:
Enabling local, community, and global retrieval augmented generation utilizing the aggregated summaries and the knowledge graph. – these limitations are considered to be similar to a user reviewing data and arranging/indexing the data in a manner that enables for the data to be used or enabled to later be used by local, community, ang global retrieval augmented generation. It is noted that “enabling” is being considered broadly and is considered to be make something operational, active, or possible.
Step 2A, Prong II: Integrated into a Practical Application?
The claims recite the following additional limitations:
As per claim 1, the claim similarly recites extra solution activity of gathering data in the following limitations:
Obtaining aggregated summaries and a related knowledge graph;
These limitations recite insignificant extra-solution activity as preliminary data gathering as retrieval/receiving of data such as 'obtaining information' as identified in MPEP 2106.05(g) and does not provide integration into a practical application.
The claim similarly recites extra solution activity such as outputting data in the following limitations:
N/A. Alternatively, to the extent that there is an output by the utilizing, the claim generically recites the utilization to enable a later process.
These limitations recite insignificant extra-solution activity as post-solution data outputting per MPEP 2106.05(g) and does not provide integration into a practical application.
ADDITIONAL ELEMENTS:
As per claim 1, the claim recites the following additional elements:
It is noted that the elements are considered to be addressed above. However, for purposes of compact prosecution, “enabling local, community, and global retrieval augmented generation” can alternatively be considered to be an intended use or a high-level recitation of generic computer components, computer elements used as a tool, and represent mere instructions to apply the abstract idea on a computer as in MPEP 2106.05(f), and does not provide integration into a practical application.
Therefore, claim 1 does not integrate the recited abstract ideas into a practical application.
Step 2B: Claim recites additional elements or limitations that amount to an inventive concept?
When considered individually or in combination, the additional limitations and elements of claim 1 does not amount to significantly more than the judicial exception for the same reasons discussed above as to why the additional limitations do not integrate the abstract idea into a practical application. The additional elements of outlined in Step 2A performing functions as designed simply accomplishes execution of the abstract ideas.
The additional limitations identified as insignificant extra-solution activity above the conclusions are carried over and they also do not provide significantly more.
Receiving data is an example of gathering data that has been found by the courts to be well understood routine and conventional. For Berkheimer support see court recognized activities in MPEP 2106.05(d)(II). Receiving data of a solution is an example of providing output data via a user interface is also well understood, routine, and conventional, see MPEP 2106.05(d)(II).
The additional elements enabling local, community, and global retrieval augmented generation reciting generic computer components as mere instructions to apply on a computer per MPEP 2106.05(f) are carried over and do not provide significantly more than the abstract idea.
In conclusions from above for the elements reciting generic computer components as mere instructions to apply on a computer per MPEP 2106.05(f) are carried over and do not provide significantly more than the abstract idea. Looking at the limitations in combination and the claims as a whole does not change this conclusion and the claim is ineligible.
Therefore, the additional limitations of claim 1 do not amount to significantly more than the judicial exception.
Thus, claim 1 recite abstract ideas with additional elements rendered at a high level of generality resulting in claims that do not integrate the abstract idea into a practical application or amount to significantly more than the judicial exception.
Therefore, claim 1 is not patent eligible.
Regarding claims 2-8, the claims further recite mental processes with the aid of pen and paper or in the alternative mathematical steps for processing the graphs. Therefore, they are rejected for substantially the same reasons as claim 1, above.
Claim 9 has substantially similar limitations as claim 1; therefore, it is rejected for substantially similar reasons.
Regarding claim 10, the claim further recites mental processes with the aid of pen and paper or in the alternative mathematical steps for processing the graphs. Therefore, they are rejected for substantially the same reasons as claim 9, above.
Regarding claim 11
Step 2A, Prong I: Judicial Exception Recited?
The examiner submits that the foregoing claim limitations constitute a “mental process”, as the claims cover performance of the limitations in the human mind, given the broadest reasonable interpretation.
As per claim 11, the claim recites the limitations of:
performing question assessment on a user query relating to a private dataset; -- This limitation is considered to be a mental process, because a human can perform question assessment of user queries that relate to a private dataset.
determining whether the user query requires summarizations of an entirety of the private dataset; -- This limitation is considered to be a mental process, because a human can make the determinations as claimed.
in instances where the user query requires summarizations of the entirety of the private dataset, processing the user query utilizing knowledge graph retrieval augmented generation (RAG) with global summarization; -- This limitation is considered to be the act of a user deciding to use certain data in order to manipulate information to understand, react to, or interact with the world.
in instances where the user query does not require summarizations of the entirety of the private dataset, evaluating whether the user query relates to a particular entity of the private dataset; -- This limitation is considered to be a mental process, because a user can perform the evaluation in determining whether the query relates to a particular entity of the private dataset.
in instances where the question relates to a particular entity of the private dataset, processing the user query utilizing knowledge graph RAG with local summarization; and, This limitation is considered to be the act of a user deciding to use certain data in order to manipulate information to understand, react to, or interact with the world.
in instances where the user query does not relate to a particular entity of the private dataset, processing the user query utilizing knowledge graph RAG with community summarization. This limitation is considered to be the act of a user deciding to use certain data in order to manipulate information to understand, react to, or interact with the world.
Step 2A, Prong II: Integrated into a Practical Application?
The claims recite the following additional limitations:
As per claim 11, the claim similarly recites extra solution activity of gathering data in the following limitations:
Alternatively, “performing question assessment on a user query relating to a private dataset” can be considered to be a data gathering function.
These limitations recite insignificant extra-solution activity as preliminary data gathering as retrieval/receiving of data such as 'obtaining information' as identified in MPEP 2106.05(g) and does not provide integration into a practical application.
The claim similarly recites extra solution activity such as outputting data in the following limitations:
To the extent that processing the query requires an output, the manner in which claimed is considered to be recited at a high level of generality.
These limitations recite insignificant extra-solution activity as post-solution data outputting per MPEP 2106.05(g) and does not provide integration into a practical application.
ADDITIONAL ELEMENTS:
As per claim 1, the claim recites the following additional elements:
It is noted that the elements are considered to be addressed above. However, for purposes of compact prosecution, “utilizing knowledge graph RAG” can alternatively be considered to be an intended use or a high-level recitation of generic computer components, computer elements used as a tool, and represent mere instructions to apply the abstract idea on a computer as in MPEP 2106.05(f), and does not provide integration into a practical application.
Therefore, claim 1 does not integrate the recited abstract ideas into a practical application.
Step 2B: Claim recites additional elements or limitations that amount to an inventive concept?
When considered individually or in combination, the additional limitations and elements of claim 1 does not amount to significantly more than the judicial exception for the same reasons discussed above as to why the additional limitations do not integrate the abstract idea into a practical application. The additional elements of outlined in Step 2A performing functions as designed simply accomplishes execution of the abstract ideas.
The additional limitations identified as insignificant extra-solution activity above the conclusions are carried over and they also do not provide significantly more.
Receiving data is an example of gathering data that has been found by the courts to be well understood routine and conventional. For Berkheimer support see court recognized activities in MPEP 2106.05(d)(II). Receiving data of a solution is an example of providing output data via a user interface is also well understood, routine, and conventional, see MPEP 2106.05(d)(II).
The additional elements “utilizing knowledge graph RAG” reciting generic computer components as mere instructions to apply on a computer per MPEP 2106.05(f) are carried over and do not provide significantly more than the abstract idea.
In conclusions from above for the elements reciting generic computer components as mere instructions to apply on a computer per MPEP 2106.05(f) are carried over and do not provide significantly more than the abstract idea. Looking at the limitations in combination and the claims as a whole does not change this conclusion and the claim is ineligible.
Therefore, the additional limitations of claim 1 do not amount to significantly more than the judicial exception.
Thus, claim 11 recite abstract ideas with additional elements rendered at a high level of generality resulting in claims that do not integrate the abstract idea into a practical application or amount to significantly more than the judicial exception.
Therefore, claim 11 is not patent eligible.
Claims 12-14 disclose features relating to an use-interface. The additional features of claims 12-14 are considered to be additional elements of merely apply on a computer under steps 2A Prong II and Steps 2B under the analysis, and are considered to be ineligible for the same reasons as claim 11, above.
Claims 17-20 are considered to be further mental processes where a human can perform the computations with the aid of pen and paper and determine the data to use for subsequent retrieval. Therefore, these claims are rejected for the same reasons as claim 11.
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)(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.
Claim 9 is rejected under 35 U.S.C. 102(a)(2) as being anticipated by Garapati et al. (U.S. Publication No. 2025/0111150 A1, hereinafter referred to as “Garapati”).
Regarding claim 9, Garapati discloses a system, comprising: storage configured to store computer-readable instructions; and, a processor configured to execute the computer-readable instructions to: (“a system, such as a mainframe system or a distributed server system, may include at least one memory, including instructions, and at least one processor that is operably coupled to the at least one memory and that is arranged and configured to execute instructions that, when executed, cause the at least one processor to perform the instructions of the computer program products and/or the operations of the computer-implemented methods.”)(e.g., paragraphs [0010] and [0237]
obtain aggregated operations and a related knowledge graph; and, (“The situation, on the other hand, as used herein, generally requires some response. The situation may reflect an aggregate impact of multiple events. In some cases, however, the situation could be caused by, or include a single event. In many cases, multiple situations may occur within a single time period, or across overlapping time periods. Consequently, when multiple situations occur within single or overlapping time period(s), and each situation includes multiple events, it may be difficult to determine which events should be included within each situation.” “FIG. 10 is an example flowchart illustrating operations for training a large language model to generate a narrative using the example transformer layers of FIGS. 7, 8, and 9. During training, historical situations may be used to determine and extract knowledge graph (KG), topological graph data, and situation event graph(s) (1002). This extracted data may be formatted for inclusion within a prompt for the LLM to be trained (1004). Examples of how to format such a prompt are provided below, e.g., with respect to FIGS. 11A and 11B. In parallel, corresponding narrative(s) may be captured from domain experts or relevant policy engines (which, e.g., enforce access rules for network resources/data), with a focus on both textual and topological aspects of the included narratives (1006). Collected data may be pre-processed and relevant, comprehensive narratives may be extracted (1008).”)(e.g., figures 2A, 2B, 10, 12, 14 and 15 and paragraphs [0088], [0090], [0204] and [0205])
enable graph-based retrieval augmented generation utilizing the aggregated operations and the knowledge graph. (“automatic narrative generation and/or automatic remediation generation may be obtained by adaptive training from extracted context for a situation event graph that includes not only textual event context but also context from surrounding events, topology context, and/or temporal context of a larger situation. As a result, a human-readable narrative focused not only on the root cause and symptoms, but also on the topological characteristics, may thus be generated.” “Prompt-response pairs may then be formatted, including both textual and graph portions, along with corresponding remediations, as previously extracted from IT ticket data (e.g., worklogs) (1410). The training process may thus be executed using the prompt-response pairs (1412), e.g., to train the topological context adapters 712, 714 of FIGS. 7 and 8, for all available prompt-remedy pairs. The resulting fine-tuned model may thus be persisted for later use during inference flow (1414).”)(e.g., figures 2A, 2B, 10, 12, 14 and 15 and paragraphs [0038], [0042] and [0221]).
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.
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.
This application currently names joint inventors. In considering patentability of the claims the examiner presumes that the subject matter of the various claims was commonly owned as of the effective filing date of the claimed invention(s) absent any evidence to the contrary. Applicant is advised of the obligation under 37 CFR 1.56 to point out the inventor and effective filing dates of each claim that was not commonly owned as of the effective filing date of the later invention in order for the examiner to consider the applicability of 35 U.S.C. 102(b)(2)(C) for any potential 35 U.S.C. 102(a)(2) prior art against the later invention.
Claims 1, 11-14 and 17 are rejected under 35 U.S.C. 103 as being unpatentable over Garapati in view of Lymberopoulos et al. (U.S. Publication No. 2011/0184936 A1, hereinafter referred to as “Lymberopoulos”).
Regarding claim 1, Garapati discloses a method comprising: obtaining aggregated summaries and a related knowledge graph; and, (“The situation, on the other hand, as used herein, generally requires some response. The situation may reflect an aggregate impact of multiple events. In some cases, however, the situation could be caused by, or include a single event. In many cases, multiple situations may occur within a single time period, or across overlapping time periods. Consequently, when multiple situations occur within single or overlapping time period(s), and each situation includes multiple events, it may be difficult to determine which events should be included within each situation.” “FIG. 10 is an example flowchart illustrating operations for training a large language model to generate a narrative using the example transformer layers of FIGS. 7, 8, and 9. During training, historical situations may be used to determine and extract knowledge graph (KG), topological graph data, and situation event graph(s) (1002). This extracted data may be formatted for inclusion within a prompt for the LLM to be trained (1004). Examples of how to format such a prompt are provided below, e.g., with respect to FIGS. 11A and 11B. In parallel, corresponding narrative(s) may be captured from domain experts or relevant policy engines (which, e.g., enforce access rules for network resources/data), with a focus on both textual and topological aspects of the included narratives (1006). Collected data may be pre-processed and relevant, comprehensive narratives may be extracted (1008).”)(e.g., figures 2A, 2B, 10, 12, 14 and 15 and paragraphs [0088], [0090], [0204] and [0205])
enabling local, community, and global retrieval augmented generation utilizing the aggregated summaries and the knowledge graph. (“automatic narrative generation and/or automatic remediation generation may be obtained by adaptive training from extracted context for a situation event graph that includes not only textual event context but also context from surrounding events, topology context, and/or temporal context of a larger situation. As a result, a human-readable narrative focused not only on the root cause and symptoms, but also on the topological characteristics, may thus be generated.” “Prompt-response pairs may then be formatted, including both textual and graph portions, along with corresponding remediations, as previously extracted from IT ticket data (e.g., worklogs) (1410). The training process may thus be executed using the prompt-response pairs (1412), e.g., to train the topological context adapters 712, 714 of FIGS. 7 and 8, for all available prompt-remedy pairs. The resulting fine-tuned model may thus be persisted for later use during inference flow (1414).”)(e.g., figures 2A, 2B, 10, 12, 14 and 15 and paragraphs [0038], [0042] and [0221]).
However, Garapati does not appear to specifically disclose local, community and global retrieval.
On the other hand, Lymberopoulos, which relates to a dynamic community-based cache for mobile search (title), does disclose local, community and global retrieval. (“A "Community-Based Mobile Search Cache" provides various techniques for maximizing the number of query results served from a local "query cache", thereby significantly limiting the need to connect to the Internet or cloud using 3G or other wireless links to service search queries. The query cache is constructed remotely and downloaded to mobile devices. Contents of the query cache are determined by mining popular queries from mobile search logs, either globally or based on queries of one or more groups or subgroups of users. In various embodiments, searching and browsing behaviors of individual users are evaluated to customize the query cache for particular users or user groups. The content of web pages related to popular queries may also be included in the query cache.”)(e.g., abstract, figures 1-2 and paragraph [0031]).
Garapati discloses narrative generation for situation event graphs. In Garapati, graph data is considered along with aggregated event data to provide a combined narrative of the situation that explains the causal chain of events and/or instructions to remedy the situation. E.g., abstract. However, Garapati does not appear to specifically disclose that the system enables local, community and global retrieval. On the other hand, it is known for systems to consider various cost savings by accessing personalized, community based and global retrieval to better tailor the content and to reduce cost in processing the data to provide relevant results. One example is Lymberopoulos, which relates to a dynamic community-based cache for mobile search, provides using personal, community based and global retrieval mechanisms to better rank and deliver content that is most relevant to the user’s interests. E.g., paragraphs [0014]-[0018]. Therefore, it would have been obvious to one of ordinary skill in the art before the effective filing date to combine the consideration of accessing data from local, community or global as disclosed in Lymberopoulos to Garapati to further enhance the manner, relevance and rankings of content is processed to provide the user with a more tailored and customized experience.
Regarding claim 11, Garapati discloses a computer-readable storage medium storing instructions comprising: (non-transitory computer-readable storage medium)(e.g., paragraph [0124])
performing question assessment on a user query relating to a private dataset; (“An event pair selector 138 may be configured to analyze selected pairs of events from the event set 137. For example, in some examples, the event pair selector 138 may be configured to analyze each pair-wise combination of all of the events of the event set 137.” “In other examples, some event pairs may be more valuable than others for purposes of identification and processing by the situation identifier 128. Moreover, as referenced above, the event set 137 may dynamically change over time, and the event pair selector 138 may benefit from being configured to incrementally add new events to the event set 137.”)(e.g., paragraphs [0049], [0085] and [0101]-[0105])
determining whether the user query requires summarizations of an entirety of the private dataset; (“In any of the above examples, and in other scenarios, the event pair selector 138 may be configured to filter some events from the event set 137 prior to, or in conjunction with, selecting event pairs for further processing. For example, the event pair selector 138 may be configured to identify and filter low-entropy events.”)(e.g., paragraph [0104])
in instances where the user query requires summarizations of the entirety of the private dataset, processing the user query utilizing knowledge graph retrieval augmented generation (RAG) with global summarization; (“A customized LLM algorithm, which may be based on, e.g., a Generative Pretrained Transformer (GPT), may thus be trained to determine a relevant context, not just from a context of an individual event, but also from the context of surrounding events, as well as a topology context and temporal context of the situation. In this way, the customized LLM algorithm may be configured to generate a human-readable narrative and/or remediation that can be focused not only on the root cause and symptoms, but also on relevant topological characteristics of the IT system. Described custom LLMs may be utilized by various types of situation or incident detector(s) or handler(s) to generate accurate and comprehensive narratives, as well as helpful and actionable remediations, in a process(es) that may be adapted continuously to provide up-to-date solutions.” “For example, the knowledge graph 126 may be used to capture domain knowledge that is entity-specific, user-specific, or deployment-specific.”)(e.g., paragraphs [0041], [0045], and [0080])
in instances where the user query does not require summarizations of the entirety of the private dataset, evaluating whether the user query relates to a particular entity of the private dataset; (“the knowledge graph 126 may be used to capture domain knowledge that is entity-specific, user-specific, or deployment-specific. The knowledge graph 126 may include user knowledge captured declaratively in graph form over time and/or in response to changes being made to the systems 104, 108.”)(e.g., paragraph [0080])
in instances where the question relates to a particular entity of the private dataset, processing the user query utilizing knowledge graph RAG with local summarization; and, (“the question” does not have antecedent basis. This is considered to be “the query.” “the knowledge graph 126 may be used to capture domain knowledge that is entity-specific, user-specific, or deployment-specific. The knowledge graph 126 may include user knowledge captured declaratively in graph form over time and/or in response to changes being made to the systems 104, 108.”)(e.g., paragraph [0080])
in instances where the user query does not relate to a particular entity of the private dataset, processing the user query utilizing knowledge graph RAG with community summarization. ( “the knowledge graph 126 may be used to capture domain knowledge that is entity-specific, user-specific, or deployment-specific. The knowledge graph 126 may include user knowledge captured declaratively in graph form over time and/or in response to changes being made to the systems 104, 108.”)(e.g., paragraph [0080]).
Garapati discloses that the knowledge graph 126 may be used to capture domain knowledge that is entity-specific, user-specific, or deployment specific and that the model is trained to combine text and graph to generate a narrative in response to a user’s inquiry. However, Garapati does not appear to specifically disclose local, community and global retrieval.
On the other hand, Lymberopoulos, which relates to a dynamic community-based cache for mobile search (title), does disclose local, community and global retrieval. (“A "Community-Based Mobile Search Cache" provides various techniques for maximizing the number of query results served from a local "query cache", thereby significantly limiting the need to connect to the Internet or cloud using 3G or other wireless links to service search queries. The query cache is constructed remotely and downloaded to mobile devices. Contents of the query cache are determined by mining popular queries from mobile search logs, either globally or based on queries of one or more groups or subgroups of users. In various embodiments, searching and browsing behaviors of individual users are evaluated to customize the query cache for particular users or user groups. The content of web pages related to popular queries may also be included in the query cache.”)(e.g., abstract, figures 1-2 and paragraph [0031]).
It would have been obvious to combine Lymberopoulos with Garapati for the reasons provided in claim 1 above.
Regarding claim 12, Garapati in view of Lymberopoulos discloses the computer-readable storage medium of claim 11. Garapati further discloses further comprising causing a user-interface to be generated that is configured to receive the user query. (“For example, the at least one computing device 148 may include, or have access to, a suitable display for displaying any of the inputs or outputs of the situation identifier 128, the root cause inspector 130, the prediction manager 132, and/or the remediation generator 134. For example, a suitable graphical user interface (GUI) may be used to display the clusters 146a, 146b, along with related aspects or details.”)(Garapati: e.g., figures 5A, 5B and 6 and paragraph [0126])(“ Note also that since the query caches are constructed from relatively large data sets (i.e., many queries and many users), a user interface is provided in various embodiments to allow the user to select one or more particular local query caches for download to the mobile device.”)(Lymberopoulos: e.g., paragraph [0074])
Regarding claim 13, Garapati in view of Lymberopoulos discloses the computer-readable storage medium of claim 12. Garapati further discloses further comprising causing the user-interface to be generated to present information relating to the knowledge graph RAG with global knowledge graph RAG with traversal based summarization, or knowledge graph RAG with community summarization. (“The simplified example of FIG. 1B omits many components or aspects of the at least one computing device 148, for the sake of brevity. For example, the at least one computing device 148 may include, or have access to, a suitable display for displaying any of the inputs or outputs of the situation identifier 128, the root cause inspector 130, the prediction manager 132, and/or the remediation generator 134. For example, a suitable graphical user interface (GUI) may be used to display the clusters 146a, 146b, along with related aspects or details.”)(e.g., figure 1B and paragraph [0126])
Regarding claim 14, Garapati in view of Lymberopoulos discloses the computer-readable storage medium of claim 13. Garapati further discloses further comprising causing the user-interface to allow user input to select specific information from the presented information for further processing. (“An event pair selector 138 may be configured to analyze selected pairs of events from the event set 137.”)(e.g., paragraphs [0102]-[0105])
Regarding claim 17, Garapati in view of Lymberopoulos discloses the computer-readable storage medium of claim 11. Garapati further discloses wherein processing the user query utilizing knowledge graph RAG with local summarization comprises extracting graph entities that have high semantic relevance to the user query by computing similarity scores between text embeddings of the user query and entity descriptions. (“In FIG. 1C, the fine-tuned LLM 153c is illustrated as including topology adapters representing examples of the topological context adapter(s) 154 of FIG. 1A. For example, although not separately illustrated in FIG. 1C, each such topology adapter may include a graph adapter that is configured to input and process one or more of the situation event graph(s) 146a, relevant portions of the topology data 124, and/or relevant portions of the knowledge graph 126. Each such topology adapter may further include a text adapter configured to process extracted event text obtained from the situation event graph 146a (e.g., corresponding to the event text 146c of FIG. 1A). Outputs of each graph adapter/text adapter pair may then be combined for further processing by subsequent stages of the fine-tuned LLM 153c. More detailed examples of such graph adapters and text adapters are described below in detail, e.g., with respect to FIG. 8.”)(e.g., paragraphs [0134] and [0138]).
Claim 2 is rejected under 35 U.S.C. 103 as being unpatentable over Garapati in view of Lymberopoulos and in further view of Edge et al. (U.S. Publication No. 2021/0019558 A, hereinafter referred to as “Edge”).
Regarding claim 2, Garapati in view of Lymberopoulos discloses the method of claim 1. However, neither reference appears to specifically disclose further comprising aggregating edges between shared nodes and using frequency count as an edge weight of the knowledge graph.
On the other hand, Edge, which relates to modeling higher-level metrics from graph data derived from already-collected but not yet connected data (title), does disclose further comprising aggregating edges between shared nodes and using frequency count as an edge weight of the knowledge graph. (“one may induce a graph in which edges are created in one of two typical ways depending on the node type of interest: (1) between actor nodes whenever both actors perform the same action of interest on the same object in the same time window, with the edge weight reflecting the aggregate (e.g., count) of shared actions across all objects, and (2) between object nodes whenever both objects receive the same action of interest from the same actor in the same time window, with the edge weight reflecting the aggregate (e.g., count) of shared actions across all actors.” “In one example, the graph may be between actor nodes whenever both actors perform the same action of interest on the same object in the same time window, with the edge weight reflecting the aggregate (e.g., count) of shared actions across all objects. In another example, the graph may be between object nodes whenever both objects receive the same action of interest from the same actor in the same time window, with the edge weight reflecting the aggregate (e.g., count) of shared actions across all actors.”)(e.g., paragraphs [0062] and [0075]).
It would have been obvious to combine Lymberopoulos with Garapati for the reasons provided in claim 1 above. Garapati discloses a narrative combining graph and text data and includes edge scores (e.g., paragraph [0115]), but does not appear to specifically disclose the use of frequency count as an edge weight of the knowledge graph. On the other hand, Edge provides that analyzing the collected graph data can the interdependence of users’ actions and the importance of structural relationships, which provides greater insights of the underlying data. E.g., paragraphs [0001]-[0002]. Therefore, it would have been obvious to incorporate the graph analysis as disclosed in Edge to the Lymberopoulos-Garapati combination to further enhance the manner in which graph data is analyzed to provide greater insights when generating the output of query requests.
Conclusion
The prior art made of record, listed on form PTO-892, and not relied upon is considered pertinent to applicant's disclosure.
Any inquiry concerning this communication or earlier communications from the examiner should be directed to RICHARD L BOWEN whose telephone number is (571)270-5982. The examiner can normally be reached Monday through Friday 7:30AM - 4:00PM EST.
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, Aleksandr Kerzhner can be reached at (571)270-1760. 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.
/RICHARD L BOWEN/ Primary Examiner, Art Unit 2165
Prosecution Insights