Back to results
US 12,608,626 B2Granted
Plug-and-analyze framework for knowledge base construction
AssigneeBristol-Myers Squibb Company
InventorsOctavian Florin Filoti, Christian Joseph Merrill, Mark Salacinski, Irene Yue-Ling Pak
Life of this patent
Estimating the term…
Abstract
Provided herein are system, apparatus, device, method, and/or computer program product embodiments, and/or combinations and sub-combinations thereof, for generating a knowledge base. In a given embodiment, machine-learning techniques and models are used to extract information and knowledge from different document formats by processing any supported unstructured, semi-structured and structured data types. The extracted information and knowledge may be used to generate a knowledge base.
Claims45
A method for generating a knowledge base, the method comprising:
- receiving, by a processor, a plurality of files;
- triggering, by the processor, execution of a first set of machine-learning models to extract data from the plurality of files, wherein each of the first set of machine-learning models outputs a first set of data objects comprising extracted data from the plurality of files;
- triggering, by the processor, execution of a second set of machine-learning models using the first set of data objects, wherein the second set of machine-learning models is configured to output a second set of data objects, each data object in the second set of data objects comprising an attribute of a plurality of attributes, each attribute of the plurality of attributes comprising targeted information extracted from the first set of data objects;
- tracking, by the processor, a completion status of each of the second set of machine-learning models;
- upon completion of each of the second set of machine-learning models, generating, by the processor, the knowledge base using the first set of data objects and the second set of data objects;
- receiving, by the processor, a query;
- identifying, by the processor, information in the knowledge base based on the query;
- triggering, by the processor, execution of a special machine-learning model to generate an output for the query using the information;
- and
- updating, by the processor, the knowledge base using the output.
7 more elements
A system for generating a knowledge base, the system comprising:
- a memory;
- and
- a processor coupled to the memory, the processor configured to:
- receive a plurality of files;
- trigger execution of a second set of machine-learning models using the first set of data objects, wherein the second set of machine-learning models are configured to output a second set of data objects and each data object in the second set of data objects comprising an attribute of a plurality of attributes about the extracted data in the first set of data objects;
- trigger execution of a second set of machine-learning models to process the extracted data in the first set of data objects, wherein the second set of machine-learning models is configured to output a second set of data objects, each data object in the second set of data objects comprising an attribute of a plurality of attributes, each attribute of the plurality of attributes comprising targeted information extracted from the first set of data objects;
- track a completion status of each of the second set of machine-learning models;
- upon completion of each of the second set of machine-learning models, generate the knowledge base using the first set of data objects and the second set of data objects;
- receive a query;
- identify information in the knowledge base based on the query;
- trigger execution of a special machine-learning model to generate an output for the query using the information;
- and
- update the knowledge base using the output.
10 more elements
A non-transitory computer-readable medium having instructions stored thereon that, when executed by at least one computing device, cause the at least one computing device to perform operations comprising:
- receiving a plurality of files;
- triggering execution of a first set of machine-learning models to extract data from the plurality of files, wherein each of the first set of machine-learning models outputs a first set of data objects comprising extracted data from the plurality of files;
- triggering execution of a second set of machine-learning models using the first set of data objects, wherein the second set of machine-learning models is configured to output a second set of data objects, each data object in the second set of data objects comprising an attribute of a plurality of attributes, each attribute of the plurality of attributes comprising targeted information extracted from the first set of data objects;
- tracking a completion of each of the second set of machine-learning models;
- upon completion of each of the second set of machine-learning models, generating a knowledge manager to generate the knowledge base using the first set of data objects and the second set of data objects;
- receiving a query;
- identifying information in the knowledge base based on the query;
- triggering execution of a special machine-learning model to generate an output for the query using the information;
- and
- updating the knowledge base using the output.
7 more elements
Dependencies are inferred from the claim text.