Enhancing entity extraction

Inxight has launched ThingFinder advanced, an add-on module to the original product, which identifies and extracts 27 key entities--such as people, dates, places, companies--from any text data source, in multiple languages. The company says ThingFinder can be integrated into virtually any application that processes textual information, enabling users to create structured data from unstructured information. Inxight says examples of ThingFinder’s capabilities include:

• combine the most relevant information in unstructured data with structured data to generate the reports and statistics that drive business;

• automatically find all references to products and people in customer service logs and e-mails and store them in a CRM database;

• discover all company and brand mentions for competitive analysis and business intelligence;

• create link analysis and business intelligence applications that monitor trends and movements associated with people, places, dates and companies; and

• mine large volumes of text for relevant information and quickly identify trends in data sets.

ThingFinder Advanced allows users to define custom patterns of tokens in regular expression syntax. The new module can extract such custom entities as date/time stamps, chemical compound names or formulae, serial or part numbers, internal project codes, etc.