Knowledge-based data mining system
AssigneeInternational Business Machines Corporation
InventorsMatthew Denesuk, Daniel Frederick Gruhl, Sridhar Rajagopalan, Andrew S. Tomkins
Life of this patent
Estimating the term…
Abstract
In a data mining system, data is gathered into a data store using, e.g., a Web crawler. The data is classified into entities. Data miners use rules to process the entities and append respective keys to the entities representing characteristics of the entities as derived from rules embodied in the miners. With these keys, characteristics of entities as defined by disparate expert authors of the data miners are identified for use in responding to complex data requests from customers.
Claims26
-30. (canceled)
A computer-implemented method for collecting and associating unstructured data, comprising:
- receiving a plurality of documents, the documents comprising entities of interest;
- analyzing the plurality of documents to identify the entities of interest, wherein a direct linkage does not exist between at least a first and a second identified entity of interest;
- and
- determining from the plurality of documents a bridging entity, wherein:
- a first direct linkage exists between the first identified entity of interest and the bridging entity;
- and
- a second direct linkage exists between the second identified entity of interest and the bridging entity.
A system for automatically collecting and associating unstructured data, comprising:
- a storage configured to receive a plurality of documents, the documents comprising entities of interest;
- a data miner configured to:
- analyze the plurality of documents to identify the entities of interest, wherein a direct linkage does not exist between at least a first and a second identified entity of interest;
- and
- determine from the plurality of documents a bridging entity, wherein:
- a first direct linkage exists between the first identified entity of interest and the bridging entity;
- and
- a second direct linkage exists between the second identified entity of interest and the bridging entity.
A computer-implemented method for analyzing buzz regarding a product or service, comprising:
- receiving a plurality of documents;
- in the plurality of documents, identifying locations where a product or service of interest is mentioned based on a determined context of the mention of the product or service of interest;
- and
- determining from the mentions a buzz associated with the product or service of interest.
A system for analyzing buzz regarding a product or service, comprising:
- a receiver to receive a plurality of documents;
- a data miner to:
- in the plurality of documents locations, identify where a product or service of interest is mentioned based on a determined context of the mention of the product or service of interest;
- determine a buzz associated with the product or service of interest periodically over time;
- and
- determine trend over time in buzz associated with the product or service of interest.
Dependencies are inferred from the claim text.