Detailed Action
This action is in response to the RCE filed on 03/30/2026 for the amended claims filed 03/02/2026 for application 17/367,598, in which:
Claims 1, 11, and 20 are independent claims.
Claims 8 and 17 have been canceled.
Claims 1, 4, 9-11, 14, and 18-20 have been amended.
Claims 1-7, 9-16, and 18-21 are currently pending.
Continued Examination Under 37 CFR 1.114
A request for continued examination under 37 CFR 1.114, including the fee set forth in 37 CFR 1.17(e), was filed in this application after final rejection. Since this application is eligible for continued examination under 37 CFR 1.114, and the fee set forth in 37 CFR 1.17(e) has been timely paid, the finality of the previous Office action has been withdrawn pursuant to 37 CFR 1.114. Applicant's submission filed on 03/30/2026 has been entered.
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 .
Regarding the 35 U.S.C. § 101 Rejections:
Applicant's amendments to the independent claims has overcome the previous rejections under 35 U.S.C. § 101. The previous rejections have been withdrawn.
Response to Arguments
Applicant's arguments filed 03/30/2026 have been fully considered but they are not persuasive.
Regarding the 35 U.S.C. § 103 Rejections:
Applicant's arguments regarding the 35 U.S.C. § 103 rejections of the previous office action have been fully considered, but are unpersuasive.
Applicant asserts (Pages 15-16), that Brandão is disqualified as prior art under 35 U.S.C. § 102 (b)(1)(A). The Applicant submits that Effective Filing Date of the instant application is July 5, 2021, and the publication date of the Brandão is November 1, 2020. Further, the Applicant submits that all inventors of the instant application are co-authors of the cited NPL Brandão. Further, Brandão is published within the one year before the effective filing date of the instant application. Therefore, U.S.C. § 102(b)(1 )(a) exception is met. Accordingly, the Applicant respectfully submits that Brandão does not qualify as prior art under 35 U.S.C. § 102 (a)(1) and requests that the rejections be withdrawn. Therefore, the Applicant respectfully submits that the claims 1-7, 9-16, and 18-21 are not taught, suggested, or rendered obvious over the combination of Crabtree, Brandão, and Azzini.
Applicant’s arguments, see Pages 15-16, filed 03/30/2026, with respect to the rejection(s) of claim(s) 1-7, 9-16, and 18-21 under 35 U.S.C. 103 have been fully considered and are persuasive in terms of Brandão. However, upon further consideration/search, a new ground(s) of rejection has been made which is necessitated by the amended clams. The rejections have been updated below.
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.
Claims 1-7, 9-16, and 18-21 are rejected under 35 U.S.C. 103 as being unpatentable over Crabtree et al., US PG PUB 2021/0019674 A1, in view of Carlsson et al., “Cognitive Maps and a Hyperknowledge Support System in Strategic Management”, in view of Prabhakara, US-20160371132-A1, in view of Azzini et al., “Using Semantic Lifting for improving Process Mining: a Data Loss Prevention System case study”.
Regarding Claim 1:
Crabtree teaches:
A computing device, comprising: a processor; a network interface coupled to the processor to enable communication over a network; a storage device for content and programming coupled to the processor; and an engine stored in the storage device, wherein an execution of the engine by the processor configures a user device to perform acts comprising:
(Crabtree, Page 3, Column 1, [0006], “… a system for risk profiling and rating of extended relationships using ontological databases is disclosed, comprising: a computing device comprising a memory, a processor, and a non-volatile data storage device; a semantic query analyzer comprising a first plurality of programming instructions stored in the memory and operating on the processor, wherein the first plurality of programming instructions cause the computing device to …”; Page 21, Column 2, [0156], “… Computing device 10 may be configured to communicate with a plurality of other computing devices, such as clients or servers, over communications networks such as …”; Page 21, Column 2, [0157], “… computing device 10 includes one or more central processing units (CPU) 12, one or more interfaces 15, and one or more busses 14 (such as a peripheral component interconnect (PCI) bus)”; Fig. 31. Crabtree teaches a processor, variety of networks and an interface coupled to the processor (network interface coupled to the processor to enable communication over a network), memory (storage device for content and programming) operating on the processor (coupled to the processor), a semantic query analyzer stored in the memory (engine stored in the storage device which is also shown in Fig. 31), which cause the device to (device to perform acts)).
automatically capturing, via one or more sensors, information of a user interacting with given data on a graphical user interface (Crabtree, Page 3, Column 1, [0006], “… receive a natural language query”; Page 17, Column 2, [0125], “A user inputs a search query into the system 3110 that is analyzed by a semantic search engine 2502”; Fig. 31; Page 2, [0004], “ … system and method … that automatically ingests both structured and unstructured data …”; Page 8, [0078], “… receive streaming data from a large plurality of sensors that may be of several different types. The multiple dimension time series data store module may also store any time series data ... such as ...to enterprise network usage data, component and system logs, performance data, network service information captures such as, but not limited to news and financial feeds, and sales and service related customer data … dynamically allotting network bandwidth and server processing channels to process the incoming data”; Page 7, Column 2, [0078], “Client access to the system 105 for specific data entry, system control and for interaction with system output such as automated predictive decision making and planning and alternate pathway simulations, occurs through the system's ... cloud interface 110 which uses a versatile, robust web application driven interface for both input and display”. The system/method taught by Crabtree automatically ingests/receives/extracts (which is interpreted by the examiner as capturing) from multiple data sources; where the data sources contain data for users/clients interacting with given data (where the given data is interpreted as where the users/clients operate within application driven interface which contains a display for the graphical representation (thus, interpreted by the examiner as a GUI: graphical user interface)), wherein the information includes contextual information, and wherein the information is captured as user interaction data; (Crabtree, Page 2, [0004], “ … includes temporal and geospatial analyses for historical and geographical context”; Page 14, [0107], “… extraction engine 2010 is tagging extracted data with relevant timestamp data and store the data as time-series data”; Fig 20; Page 4, [0032], “FIG. 20 is a block diagram of an exemplary system for contextual data collection and extraction according to various embodiments of the invention”. The system/method includes extracting the automatically ingested data (dynamically processed incoming data) where the extraction engine tags the extracted data with contextual information (which can be seen in Fig. 20); thus the extracted data is interpreted as user interaction data which is the automatically captured information of a user interacting with given data (which includes contextual information)).
structuring the user interaction data as a trail of actions over time;
(Crabtree, Page 14, Column 1, [0107], “Another capability of extraction engine 2010 is tagging extracted data with relevant timestamp data and store the data as time-series data … for classifying data in phases so that transitions over time may be captured using graph edge analysis”; Fig. 31. Fig. 31 shows the Corporate Risk Profiling and Rating Platform to receive information from users (via Search Queries) and also external sources (via Information Sources) where within the Platform there is an Extraction Engine. Crabtree teaches that the Extraction Engine is capable of tagging extracted data (user interaction data) with timestamps which structures the user interaction data as a trail of actions over time (the examiner interprets ‘trail of actions over time’ as indexed interactions over time)).
receiving an ontology for a domain related to the user interaction data; matching each action of the trail of actions onto entities of the ontology for the domain;
(Crabtree, Page 15, [0119]; Fig. 25; Fig. 31. Fig. 25 is a block diagram for ontology generation by the Automated Ontology Generator shown in Fig. 31. [0119] describes the Automated Ontology Generator in detail and notes how the search query is utilized to establish a domain related to the user interaction data. In [0119] and Fig. 25, Crabtree teaches an Automated Ontology Engine to receive Information Sources (including hierarchies, taxonomies, ontologies, dictionaries, website page ranks, etc.), which are then analyzed to generate ontologies, the newly generated ontologies are then indexed and made available to the Search Handler, as queries are received the contextual information is analyzed, the Query Context Analyzer then sends the information to the Automated Ontology Shifter to evaluate intent/shift the focus of the search to match the appropriate ontology (making the shifted generated ontologies based on a domain related to the user interaction data as the Query Context Analyzer incorporates the user interaction data via Search Queries and Contextual Information)).
generating a … knowledge graph based on the matching of each action of the trail of actions onto the entities of the ontology having the matched actions
(Crabtree, Page 3, Column 1, [0006], “ … create a weighted and directed knowledge graph, the weighted and directed knowledge graph comprising nodes representing the entities, locations, and topics associated with the subject and edges representing the relationships to the nodes in relation to the subject or the associated nodes, wherein: each node is assigned a risk value based on relationships in the ontological database …”; Page 5, Column 1, [0053], “… a knowledge graph is generated which may be presented to the user …”; Page 16, Column 1, [0119], “The ontology shifter would shift the focus of the search to the appropriate ontology … which would provide search results 2516 to the user most closely matching the user's predicted intent.”. Each node in the generated knowledge graph represents entities, locations, and topics associated with the subject while the edges represent relationships to other nodes. Each node is assigned a risk value based on the relationships in the ontological database and a probability of influence is also assigned to each edge. The structure defined by Crabtree’s knowledge graphs are based on the created ontologies that are shifted based on the search queries/contextual information of the user which leads to the matched actions (nodes (subjects) related to edges (relationships between nodes) that are based on the created shifted ontologies (which are based on user interaction data and information sources and teaches the temporal shifting of the actions to entities/relationships)));
controlling display of the ... knowledge graph on the graphical user interface;
(Crabtree, Page 7, Column 2, [0078], “Client access to the system 105 for specific data entry, system control and for interaction with system output such as automated predictive decision making and planning and alternate pathway simulations, occurs through the system's ... cloud interface 110 which uses a versatile, robust web application driven interface for both input and display”; Fig. 10 & Fig. 32; Page 19, Column 1, [0138], “... This collated time-series data may then be used to produce a visualization 1004 of changes over time, quantifying collected data into a meaningful and understandable format. As new events are recorded ... these events are automatically incorporated into the time-series data and visualizations are updated accordingly ...”. Crabtree teaches the display of the knowledge graph (example of knowledge being displayed within Figure 32) where the collected data and updates to the knowledge graph due to new events being incorporated within the knowledge graph visualizations (which are displayed to the user to interact with); thus, the examiner interprets the displaying of the knowledge graphs on the graphical user interface to be controlled by the system (shown in Figure 10: 1004) when new events are occurring).
...
iteratively generating an improved trail of actions for the ontology based on the training of the ML model; and
(Crabtree, Page 4, Column 2, [0051], “... machine learning to ... update ontological databases within the system”; Page 18, Column 1, [0127], “… More specifically, the risk rating engine 3111 using the comparative analysis results, parse each result through a semantic computing (NLP) and machine learning algorithm which identifies, categorizes, and scores each relation with a risk score. The risk rating engine 3111 then sums all scores and produces a risk rating a profile 3140 to the client comprising the knowledge graph and numerical risk score”; Page 6, Column 1, [0053], “Once the ontological databases are created or updated, ... a knowledge graph is generated which may be presented to the user for advanced insight and analysis ... ”; Page 7, [0074], “Machine learning algorithms develop models of behavior or understanding based on information fed to them as training sets, and can modify those models based on new incoming information”. Semantic computing is utilized by the Corporate Risk Profiling and Rating system which uses the ML model to update ontological databases (the examiner interprets improved trail of actions as the ontological mappings are for user interaction data and external data) which allows for a numerical score that takes into consideration the knowledge graph. The continuous monitoring for updating the models based on incoming information is interpreted by the examiner as iteratively generating. The ontological databases are created and updated after a query has been analyzed semantically. The knowledge graph is then generated or regenerated with the new entity as the knowledge graph is based off the ontological mappings (thus, an improved trail of actions due to the ontological mapping updates based on new events from the user (which is an iterative process))).
modifying the displayed ... knowledge graph based on the improved trail of actions.
(Crabtree, Fig. 10 & Fig. 32; Page 19, Column 1, [0138], “... This collated time-series data may then be used to produce a visualization 1004 of changes over time, quantifying collected data into a meaningful and understandable format. As new events are recorded ... these events are automatically incorporated into the time-series data and visualizations are updated accordingly ...”; Page 6, Column 1, [0053], “Once the ontological databases are created or updated, ... a knowledge graph is generated which may be presented to the user for advanced insight and analysis ... ”; Page 7, [0074], “Machine learning algorithms develop models of behavior or understanding based on information fed to them as training sets, and can modify those models based on new incoming information”. Crabtree teaches the display of the knowledge graph (example of knowledge being displayed within Figure 32) where the collected data and updates (interpreted as modifications by the examiner) to the knowledge graph due to new events being incorporated within the knowledge graph visualizations (which are displayed to the user to interact with); thus, the examiner interprets the displaying of the modified (where the ML model is modifying the models based on incoming information/events which improve the trail of actions) knowledge graphs on the graphical user interface to be controlled by the system (shown in Figure 10: 1004) when new events are occurring).
Crabtree does not explicitly disclose:
… hyperlinked knowledge graph …
receiving trail corrections for the trail of actions in the displayed ... knowledge graph, based on the interaction of the user with the graphical user interface;
identifying learning pipelines for the trail of corrections structured in the displayed hyperlinked knowledge graph, wherein the identifying of the learning pipelines is based on a problem addressed by the user while interacting with the given data on the graphical user interface; and
executing the identified learning pipelines …
However, Carlsson teaches:
… hyperlinked knowledge graph …
(Carlsson, Abstract, “... The process is made user-supportive with a hyperknowledge user interface ... the Woodstrat system was built with Visual Basic, in which the objects to create a hyperknowledge environment were built. It is shown that the conceptual constructs which form strategic management can be described with cognitive maps, and that these can be adequately represented with our hyperknowledge objects”; Figure 4. Carlsson teaches utilizing cognitive maps within the Woodstrat system which is creating a hyperknowledge environment for representing the hyperknowledge objects and can be seen in Figure 4. Thus, the cognitive maps that are linked based on expert knowledge/management are able to be interacted with a hyperknowledge user interface; thus, interpreted by the examiner as a hyperlinked knowledge graph).
It would have been obvious to one of ordinary skill in the art before the effective filing date of the claimed invention to utilize the device of Crabtree for structuring trail of actions with the hyperlinked knowledge graph structure taught by Carlsson. One having ordinary skill in the art would have been motivated to implement this change before the effective filing date of the claimed invention, as this leads to describing knowledge graphs via cognitive maps to enable representation of hyperknowledge objects and allows users to manage strategically and guide the system (Carlsson, Abstract, “Strategic management is defined as the system of action programs which form sustainable competitive advantages for a corporation, its divisions, and its business units in a strategic planning period. We have developed a system called Woodstrat to serve as a support system for these action program activities on both the corporate, the divisional, and the business unit levels ... The innovation in Woodstrat is that these modules are linked together, i.e., when a strong market position is built into some market segment, it will have an immediate impact on profitability through links running from the assumptions on an expected development to the projected profit/loss statement ... The intermodular links are based on expert knowledge of strategic management; expert knowledge is also worked into the modules such that the logic of strategic management guides the user through the process of working out sustainable competitive advantages. The process is made user-supportive with a hyperknowledge user interface. The support is made intuitive and effective with the use of object-oriented expert system technology. The basis for this is rather unusual: the Woodstrat system was built with Visual Basic, in which the objects to create a hyperknowledge environment were built. It is shown that the conceptual constructs which form strategic management can be described with cognitive maps, and that these can be adequately represented with our hyperknowledge objects. It is also shown that the knowledge formation which takes place in a management team when strategic plans are formed can be described and validated with a hyperknowledge support system. It is finally shown that a support system with hyperknowledge features, which are close to the cognitive maps of a management team, will have a profound impact on the depth and the structure of its strategic management processes”).
Crabtree/Carlsson do not explicitly disclose:
receiving trail corrections for the trail of actions in the displayed hyperlinked knowledge graph, based on the interaction of the user with the graphical user interface;
identifying learning pipelines for the trail of corrections structured in the displayed hyperlinked knowledge graph, wherein the identifying of the learning pipelines is based on a problem addressed by the user while interacting with the given data on the graphical user interface; and
executing the identified learning pipelines;
However, Prabhakara teaches:
receiving trail corrections for the trail of actions in the displayed hyperlinked knowledge graph, based on the interaction of the user with the graphical user interface;
(Prabhakara, FIG. 1; Page 2, Column 2, [0028], “... The sequence of activities corresponding to the sequence of events of a case can be regarded as a "trace."”; Page 3, [0044], “... The user 111 is provided with interactive visualization ... The system and methods also provide a feedback collator 112 for the user to provide feedback ...system provides the user an option to update or alter the domain knowledge base 113”. The system within Prabhakara (FIG. 1) shows user 111 able to update and alter the domain knowledge base via 113 which is fed back into the system to update/correct/alter workflow/models/rules/and structures; thus, the system is receiving trail corrections for trail of actions (trace/sequence of events) based on the users interaction of providing feedback for correction/altering via the interactive visualization display (graphical user interface)).
It would have been obvious to one of ordinary skill in the art before the effective filing date of the claimed invention to utilize the device of Crabtree/Carlsson for structuring trail of actions with the hyperlinked knowledge graph with Prabhakara’s explicit teachings of receiving trail corrections through user interaction. One having ordinary skill in the art would have been motivated to implement this change before the effective filing date of the claimed invention, as this leads user correction/improvement of a process via event logs/sequence of actions to which leads to more insight, error correction, process repairing, and modularity (Prabhakara, Page 1, Column 1, [0001-0002], “Systems and methods herein generally relate to activities that are performed using machines and performed manually, and to event logs that record such activities, and also to the improvement on the quality of such event logs ... These information systems can record or "log" events pertaining to process executions in several different formats ... Such event logs can be analyzed (e.g., using process mining techniques) to gain insights on processes and thereby assist in process improvement efforts. For any corrective action, such as process repair/process improvement based on event log analytics, the uncovered insights should be accurate and reliable ...”).
Crabtree/Carlsson/Prabhakara do not explicitly disclose:
identifying learning pipelines for the trail of corrections structured in the displayed hyperlinked knowledge graph, wherein the identifying of the learning pipelines is based on a problem addressed by the user while interacting with the given data on the graphical user interface; and
executing the identified learning pipelines;
However, Azzini teaches:
identifying learning pipelines for the trail of ... structured in the displayed hyperlinked knowledge graph, wherein the identifying of the learning pipelines is based on a problem addressed by the user while interacting with the given data on the graphical user interface; and
(Azzini, Page 68, Paragraph 2, “
PNG
media_image1.png
175
537
media_image1.png
Greyscale
”Page 70, Paragraph 2, “Since a ‘data-loss model’ is typically aimed at detecting anomalous behaviors… This can be done by identifying behavioral patterns over the sequences of events that are normally registered in the workflow logs. … in mining expected behavior … a semantic lifting procedure can be applied to the log data for remodeling the representation of the process and allowing additional investigations ”; Page 69, Table 1. Azzini teaches a semantic lifting methodology which identifies learning pipelines with the successor which analyzes based on workflow trace/log (which is interpreted by the examiner as a trail of actions structured within the knowledge graph). Table 1 shows the semantic lifting procedure being identified for users while interacting the given data shown within a workflow log; thus, interpreted by the examiner as being is based on a problem addressed by the user while interacting with the given data on the user interface).
executing the identified learning pipelines;
(Azzini, ”Page 70, Paragraph 2, “… in mining expected behavior … a semantic lifting procedure can be applied to the log data for remodeling the representation of the process and allowing additional investigations”. Azzini teaches executing the semantic lifting methodology by applying it which generates new events (entities) in the knowledge graph (RDF)).
It would have been obvious to one of ordinary skill in the art before the effective filing date of the claimed invention to utilize the device of Crabtree/Carlsson/Prabhakara with Azzini’s semantic lifting methodology and process mining (identification and execution of learning pipelines). One having ordinary skill in the art would have been motivated to implement this change before the effective filing date of the claimed invention, as this leads structuring unstructured items, transformation of low-level data, data conversion, analysis, and knowledge extraction from mapping of events/activities/tasks (Azzini, Page 63, Figure 1, Paragraph 5, “… the term semantic lifting refers to the process of associating content items with suitable semantic objects as metadata to turn unstructured content items into semantic knowledge resources … by semantic lifting we refer to all the transformations of low-level systems logs carried out in order to achieve a conceptual description of business process instances. … We believe that this problem is orthogonal to the abstraction problem in process mining, dealing with different levels of abstraction when comparing events with modeled business activities [4]: our goal is to see how associating some semantics to an event from the log it is possible to extract better knowledge about some properties of the overall process, not to see which is the mapping between events and business activities/tasks”; Page 72, Paragraph 1, “… process mining techniques can be combined with semantic lifting procedures on the workflow logs in order to discover more precise workflow models from event-based data … we highlighted the benefits using RDF as a modeling formalism by using it in our case study ...”).
Regarding Claim 2:
Crabtree/Carlsson/Prabhakara/Azzini teach the device of Claim 1. Crabtree further teaches:
wherein the contextual information includes at least one of audio or video of the user interacting with the given data.
(Crabtree, Fig. 21; Fig. 25; Fig. 31; Page 14, Column 1, [0108], “Referring to FIG. 21, extraction engine 2010 may comprise an image analysis engine 2110, an audio analysis engine 2111, a video analysis engine 2112, a text analysis engine 2113, and data formatting service 2114”; Page 14, Column 2, [0109], “Audio analysis engine 2111 may be configured to use audio analysis models to process audio data, for example, performing general speech-to-text operations or to analyze tonal cues in voice recordings. This may provide additional insight by cross referencing the tones and inflections with presented facts, for example, it may reveal whether or not certain statements can be considered truthful or not. Data extracted from audio may then be processed by data formatting service 2114, so that the data may conform to any preset standards for usage in a knowledge base.”; Page 14, Column 2, [0110], “Video analysis engine 2112 may be configured to use video analysis models to process videos, and capture information from videos. For example, analyzing body language to glean concealed information or perform lipreading analysis as a means to increase accuracy of speech dictation”. Fig. 31 shows the Corporate Risk Profiling and Rating Platform to contain the Extraction Engine which is able to extract images, audio, video, etc. Fig. 25 shows the Automated Ontology Engine which takes Search Queries with the Contextual Information as inputs for the Search Handler where the Contextual Information is gathered about the Search Query via the Extraction Engine and used when Ontology Databases are created).
Regarding Claim 3:
Crabtree/Carlsson/Prabhakara/Azzini teach the device of Claim 2. Crabtree further teaches:
wherein the execution of the engine further configures the computing device to perform an act comprising determining a facial expression of the user for at least one action of the trail of actions. (Crabtree, Page 14, Column 2, [0110], “Video analysis engine 2112 may be configured to use video analysis models to process videos, and capture information from videos. For example, analyzing body language to glean concealed information or perform lipreading analysis as a means to increase accuracy of speech dictation”; Page 17, Column 2, “The deep web extraction engine 2810 would gather real world information from the internet that is relevant to analyzing context and meaning, such as temporospatial information 2901 (weather, geopositioning, maps and terrain, etc.), physical world data 2902 (vehicle and building information, plant and animal taxonomies, physics models, etc.), and human information 2903 (facial expression information, occupation and activity information, brain imaging studies, human kinetics studies, and human emotional response studies, etc.)”. The Video Analysis engine within the Extraction Engine captures contextual information such as facial expressions (human/user interaction information such as lip reading). The Deep Web Extraction Engine uses human information gathered from multiple sources to analyze the contextual information that was received by the user to analyze facial expressions).
Regarding Claim 4:
Crabtree/Carlsson/Prabhakara/Azzini teach the device of Claim 1. Crabtree further teaches:
the user interaction data is continuously captured
(Crabtree, Fig. 3B; Page 14, Column 1, [0106], “Extraction engine 2010 may be configured to use processes of advanced cyber decision platform 100, such as connector module 135, web crawler 115, and multidimensional time series data store 120 to connect to data sources to extract data, which may be richly formatted data, structured data, unstructured data, and the like”. Fig 3B shows the Advanced Cyber Decision Platform which can be configured to the Extraction Engine and teaches Continuously Monitoring Incoming Data (353)), and
the displayed hyperlinked knowledge graph is iteratively updated based on the continuously captured user interaction
(Crabtree, Page 14, Column 2, [0127], “After the ontological databases have been created and/or updated, a directed computational graph module 155 utilizing the ontologies generates a knowledge graph”. As the ontological databases are updated via interaction data the knowledge graph is generated with the newly captured information. This is an iterative process as Crabtree teaches that when the ontological databases are created or updated that the module generates a knowledge graph. The ontological databases are continuously updated with new user interaction data (via search queries/contextual information) which leads to new knowledge graphs based on the continuously captured interactions).
Regarding Claim 5:
Crabtree/Carlsson/Prabhakara/Azzini teach the device of Claim 1. Crabtree further teaches:
wherein the execution of the engine further configures the computing device to perform an act comprising determining, using artificial intelligence, an intent of the captured information of the user based on contextual information
(Crabtree, Fig. 25; Page 16, Column 1, [0119], “The results of the query context analyzer would be sent to the automated ontology shifter 2514, which would compare the query analysis with the complex index database to determine the user's intent, even where such intent is unstated”; Page 17, Column 2, [0125], “This semantic search engine 2502 employs natural language processing (NLP) and machine learning (e.g., In2sql, NLSQL, and Quepy) to understand the intent of the search.” Fig. 25 shows the Semantic Search Engine (uses machine learning) that contains the Query Context Analyzer which determines the intent of the user interaction (Search Queries) with the use of Contextual Information (which comes from the Extraction Engine)).
Regarding Claim 6:
Crabtree/Carlsson/Prabhakara/Azzini teach the device of Claim 1. Crabtree further teaches:
wherein the structured user interaction data includes one or more nested hierarchies of the trail of actions
(Crabtree, Fig. 26; Page 7, Column 2, [0075], “Domain-specific ontology" refers to a hierarchal taxonomy of concepts and their relationships within a particular ontological domain (i.e., a set of reference ideas that establishes context”; Page 15, Column 2, [0119], “As the automated ontology generator completes its analysis, the newly-generated ontologies are passed on to the automated complex index generator 2509 for creation of a network of indices relating the information to ontologies, hierarchies, taxonomies, trustworthiness rankings, valuations, and other classifications and groupings”; Page 17, Column 2, [0127], “After the ontological databases have been created and/or updated, a directed computational graph module 155 utilizing the ontologies generates a knowledge graph. A GraphStack service 145 identifies subgraphs of interest (from the knowledge graph), returns the subgraphs of interest, and bulk loads the data about the edges and vertices based on their swimlanes from the time-series data store into Spark.”; The interactions over time are mapped via the Automated Ontology Generator with a variety of contextual information and groupings. The examiner interprets ‘nested hierarchies’ as organization of groups within groups which Crabtree teaches via groupings and indices. Fig. 26 shows an Ambiguous Search Query having search history (which teaches a nested hierarchy as past interactions over time are mapped and used as contextual information) that will be used via the Query Context Analyzer within the Automated Ontology Engine. The system also utilizes GraphStack to determine subgraphs (which are groups within groups) noting the nested hierarchies within the knowledge graphs (which are based on mapped ontologies)).
Regarding Claim 7:
Crabtree/Carlsson/Prabhakara/Azzini teach the device of Claim 1. Crabtree further teaches:
wherein the execution of the engine further configures the computing device to perform an act comprising adjusting one or more actions of the matched trail of actions upon receiving instructions from a domain expert
(Crabtree, Fig. 22; Fig. 23; Page 15, Column 1, [0147], “FIG. 23 is a flow diagram illustrating a method 2300 for knowledge base construction according to various embodiments of the invention. At an initial step 2301, system 2000 retrieves richly formatted data from a plurality of sources, which may include, local storage, cloud storage, web pages, and the like. At an optional step 2302, a user may provide the system with context to refine types of data that are extracted, for instance, financial data for a particular company. At step 2303, the system analyzes the richly formatted data using the various functions of extraction engine 2010, extracts the relevant information, and formalizes the data. At another optional step 2304, the system may receive feedback regarding the data from a variety of sources, a few of which are disclosed above in FIG. 22. At step 2305, the data is labeled, and stored in an appropriate knowledge base”; Page 14, Column 1, [0107], “Another capability of extraction engine 2010 is tagging extracted data with relevant timestamp data and store the data as time-series data. This may be useful for classifying data in phases so that transitions over time may be captured using graph edge analysis. This may be useful, for example, for tracking development in expert judgement in particular fields overtime, as well as let interested parties explore data from specific time periods.”. Fig. 22 shows a Contextual Data Extraction System that utilizes Expert Judgement (2210b) examiner interprets as domain expert) to refine matched trail of actions (examiner interprets as mapped interactions over time which are ontology mappings in this scenario)).
Regarding Claim 9:
Crabtree/Carlsson/Prabhakara/Azzini teach the device of Claim 1. Crabtree further teaches:
wherein the training of the ML model is based on a graph neural network (GNN)
(Crabtree, Page 3, Column 1, [0005], “… create a weighted and directed knowledge graph, the weighted and directed knowledge graph comprising nodes representing the entities, locations, and topics associated with the subject and edges representing the relationships to the nodes in relation to the subject or the associated nodes, wherein: each node is assigned a risk value based on relationships in the ontological database; and each edge is assigned a probability of influence between the nodes to which it is connected· and a risk rating engine comprising a third plurality of programming instructions stored in the memory and operating on the processor, wherein the third plurality of programming instructions cause the computing device to: identify paths within the directed graph which meet a pre-determined threshold of likelihood; iterate over the nodes and edges in each identified path to determine a probability of occurrence and risk impact associated with that path; assign a risk rating to each path identified, based on the probability of occurrence and risk impact associated with that path”; Page 17, Column 2, [0125], “A user inputs a search query into the system 3110 that is analyzed by a semantic search engine 2502. This semantic search engine 2502 employs natural language processing (NLP) and machine learning (e.g., In2sql, NLSQL, and Quepy) to understand the intent of the search”. The semantic computing based on NLP and machine learning algorithms utilize GNNs in Crabtree’s teachings).
Regarding Claim 10:
Crabtree/Carlsson/Prabhakara/Azzini teach the device of Claim 1. Crabtree further teaches:
wherein the execution of the engine further configures the computing device to perform an act comprising generating one or more new entities in the hyperlinked knowledge graph using the ML model
(Crabtree, Page 4, Column 2, [0052], “Once the query has been analyzed semantically, the system then pulls information from a variety of available sources using data-extraction and web-scraping techniques. This includes all private, public, and proprietary sources accessible by the system. A comprehension engine, using the same semantic computing techniques, observes and infers primary, secondary, and tertiary relationships and attributes of the entity in question. This data is stored in ontological databases”; Page 6, Column 1, [0053], “Once the ontological databases are created or updated, where the data has been organized into typical ontological data structures e.g., classes, attributes, relations, axioms, etc., a knowledge graph is generated which may be presented to the user for advanced insight and analysis into the risk factors and relationships associated with the queried entity, but also is used by the system to answer additional queries through various procedures”. The ontological databases are created or updated after a query has been analyzed semantically. The knowledge graph is then generated or regenerated with the new entity as the knowledge graph is based off the ontological mappings).
Regarding Claims 11-16 and 18-19:
Claims 11-16 and 18-19 incorporate substantively all the limitations of Claims 1-4, 6-7 and 9-10 in a non-transitory computer readable storage medium (thus, a manufacture) and further recites tangibly embodying a computer readable program code having computer readable instructions that, when executed, causes a computing device to carry out a method creating a hyperlinked knowledge graph (Crabtree, Page 22, Column 2, [0163], “Because such information and program instructions may be employed to implement one or more systems or methods described herein, at least some network device aspects may include non-transitory machine-readable storage media, which, for example, may be configured or designed to store program instructions, state information, and the like for performing various operations described herein”); thus, Claims 11-16 and 18-19 are rejected for reasons set forth in the rejections of Claims 1-4, 6-7 and 10, respectively.
Regarding Claim 20:
Claim 20 incorporate substantively all the limitations of Claims 1 in a computer-implemented method (Crabtree, Page 22, Column 2, [0163], “Because such information and program instructions may be employed to implement one or more systems or methods described herein, at least some network device aspects may include non-transitory machine-readable storage media, which, for example, may be configured or designed to store program instructions, state information, and the like for performing various operations described herein”) and further recites no new additional elements; thus, Claims 20 is rejected for reasons set forth in the rejection of Claim 1.
Regarding Claim 21:
Crabtree/Carlsson/Prabhakara/Azzini teach the device of Claim 1. Crabtree further teaches:
wherein the user interaction data includes decisions made by the user on the given data, an amount of time spent by the user on a portion of the given data, and changes made by the user to the portion of the given data
(Crabtree, Fig. 14; Page 14, Column 1, [0106], “Extraction engine 2010 may be configured to use processes of advanced cyber decision platform 100, such as connector module 135, web crawler 115, and multidimensional time series data store 120 to connect to data sources to extract data …”. Fig. 14: 1403 shows when the event occurs and also logs the event into time-series data within 1404; thus, interpreted by the examiner as an amount of time spent by the user on a portion of the given data (specific event). The log contains any changes/updated/no changes made by the user within the event).
Conclusion
Any inquiry concerning this communication or earlier communications from the examiner should be directed to IBRAHIM RAHMAN whose telephone number is (703)756-1646. 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, Kakali Chaki can be reached at (571) 272-3719. 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.
/I.R./Examiner, Art Unit 2122
/KAKALI CHAKI/Supervisory Patent Examiner, Art Unit 2122
Prosecution Insights