Natural language query procedure where query is ingested into a cognitive graph

• US 10,528,870 B2
• Filed: 10/11/2016
• Issued: 01/07/2020
• Est. Priority Date: 05/13/2016
• Status: Active Grant

• Current Assignee: Tecnotree Technologies Incorporated
• Original Assignee: Cognitive Scale Inc (Tecnotree Oyj)
• Inventors: Lindsley, Hannah R.
• Primary Examiner(s): Truong, Cam Y T
• Application Number:
  US15/290,479
• Publication Number:
  US 20170329868A1
• Time in Patent Office: 1,183 Days
• Field of Search: 707798
• CPC Class Codes: G06F 16/3329 Natural language query formulation or dialogue systems
• CPC Class Codes: G06F 16/367 Ontology
• CPC Class Codes: G06F 16/84 Mapping; Conversion
• CPC Class Codes: G06F 16/9024 Graphs; Linked lists G06F16/9027 takes precedence
• CPC Class Codes: G06F 16/90335 Query processing
• CPC Class Codes: G06N 20/00 Machine learning
• CPC Class Codes: G06N 5/01 Dynamic search techniques; Heuristics; Dynamic trees; Branch-and-bound
• CPC Class Codes: G06N 5/02 Knowledge representation; Symbolic representation
• CPC Class Codes: G06N 5/022 Knowledge engineering; Knowledge acquisition
• CPC Class Codes: G06N 5/04 Inference or reasoning models
• CPC Class Codes: G06N 5/043 Distributed expert systems; Blackboards
• CPC Class Codes: G06N 5/048 Fuzzy inferencing

Abstract

A computer-implementable method for managing a cognitive graph comprising: receiving data a data source, the data comprising a query and information relating to an answer to the query; processing the query, the processing the query identifying a plurality of query related knowledge elements; processing the information relating to the answer to the query, the processing the information relating to the answer to the query identifying a plurality of answer related knowledge elements; and, storing the plurality of query related knowledge elements and the plurality of answer related knowledge elements within the cognitive graph as a collection of knowledge elements.

Claims

1. A computer-implemented method for managing a cognitive graph of a universal knowledge repository comprising:

receiving, by a computer, data from a data source, the data comprising of a query containing text and information relating to an answer to the query;

processing, by the computer via a cognitive inference and learning system, the query, the processing of the query includes identifying a plurality of query related knowledge elements and determining whether the query includes an ambiguity, the cognitive inference and learning system comprising a cognitive platform, the cognitive platform comprising the cognitive graph and a cognitive engine, the cognitive engine comprising;

a dataset engine, a graph query engine and an insight/learning engine, the dataset engine being implemented to establish and maintain a dynamic data ingestion and enrichment pipeline, the graph query engine being implemented to receive and process queries so that the queries are bridged into the cognitive graph, the insight/learning engine being implemented to generate a cognitive insight from the cognitive graph, the cognitive graph includes an integrated machine learning functionality having cognitive functionality which uses extracted features of newly-observed data from user feedback received via an Application Program Interface during a learn phase to improve accuracy of knowledge stored within the cognitive graph, the learn phase includes feedback on observations generated during a relate phase;

performing, by the computer via the pipeline based on the user feedback and results in generation of a set of parse trees by parse rule set, mapping operations on the query related knowledge elements to resolve the ambiguity, the mapping operations comprising mapping of structural elements of the query around a verb of the query, the mapping of the structural elements includes transforming the structural elements into words higher up an inheritance chain implemented in the cognitive graph;

ranking, by the computer based on a conceptualization ranking rule set, the parse trees that represent ambiguous portions of the text;

resolving, by the computer, the parse trees to a tree representing an interpretation of the ambiguous portions of the text, the interpretation of the ambiguous portions of text is changed based upon the user feedback;

processing, by the computer, the information relating to the answer to the query, the processing of the information relating to the answer to the query identifying a plurality of answers related the knowledge elements, the answers are top-ranked parse options of the ranked parse trees;

storing, by the computer using a consistent, non-arbitrary, universally recognizable schema, the plurality of query related knowledge elements and the answers related knowledge elements within the cognitive graph of the universal knowledge repository as a collection of knowledge elements in a configuration representing relationship of concepts wherein any ambiguity identified when processing the query is stored within the cognitive graph as a query related knowledge element;

submitting, by the computer, an insight agent query from an insight agent to the universal knowledge repository; and

responsive to the submitted insight agent query, providing, by the computer based on a matching rule set and the answers related knowledge elements in the universal knowledge repository, matching results to the insight agent.

2. The method of claim 1, further comprising:

analyzing the query to determine whether the query contains text; and

performing a natural language processing operation on the text.

3. The method of claim 1, wherein:

the query related knowledge elements are stored within the cognitive graph as nodes; and

subsets of nodes are related via edges.