Eduction

Eduction

Extract the information you need

Certain business operations require you to distinguish only certain parts of a large set of information. For instance, you may be doing a search on big data, but what you really want to find are the names of companies or experts most often mentioned among the search results. HP IDOL’s eduction capabilities allow you to identify and extract entities (a word, phrase, or block of information) from text to enrich your metadata and enable you to perform these types of discovery searches.

Sourcing from a vast library of entity grammars (patterns), IDOL eduction automatically identifies and extracts terms in documents that lend themselves to key fields such as the names of companies, people, locations, addresses, and telephone numbers. IDOL offers hundreds of entities out-of-the-box across numerous languages. Conceptual entities can also be post-processed so that they are validated (e.g. is the credit card number valid?) or customized (e.g. replace SSNs with x’s).

IDOL permits custom entities to be built and deployed in a potentially unlimited range of applications, from increasing sales in eCommerce and enhancing competitive analysis, to quickly and easily comparing candidates for a job, or even connecting the dots in counter-terrorism efforts.


Benefits

Quickly narrow search results: Use auto-identified facets and conceptual entities from documents.

Validate or customize entities: Find out, for instance, if you have a valid phone number.

Accurately extract entities: Pull entities from multibyte languages such as Chinese, Japanese, Korean, Thai, and some European languages.

Numerous entities: Take advantage of hundreds of out-of-the-box entities, and easily build your own custom entities.