The software automatically categorizes large volumes of textual data, like documents, e-mail and customer interaction records, and is offered as two Classification Modules for IBM’s OmniFind Discovery Edition managed search engine and FileNet P8 content management platform.
The OmniFind Discovery Edition module offers classification as a service for any SOA-based application that manages content. The FileNet P8 module automates tagging and indexing across multiple taxonomies for traditional document management.
The new software is localized for most Western European languages as well as Japanese, Chinese and Korean character sets.
IBM said its classification technology differs from others in that it is self-learning, meaning that it becomes more accurate the more it is used to interpret meaning, the context of the language as well as any associated metadata in unstructured data.
IBM said the software provides classification support for archiving, governance, risk and compliance, e-mail management, business process management and taxonomy management applications.
The Classification modules are sold directly by IBM and its business partners.
