IBM Announces Self-Learning Software

May 25, 2007


      Bookmark and Share

IBM has announced new self-learning content classification software developed to automatically categorize large volumes of enterprise information, making it easier to find, access, and use. IBM’s content classification software offerings were designed to help organizations understand what content exists, whether it is important, and how it should be handled. The new software also can automatically classify vast amounts of previously unmanaged content so it can be retrieved and used for business purposes.

The new IBM Classification Module for OmniFind Discovery Edition and IBM Classification Module for IBM FileNet P8 provide a platform for managing all types of classification for content archiving; compliance solutions; email management; content centric business process management solutions, and taxonomy management, as well as in applications for case resolution, contact center, and online support offerings. IBM’s content classification software automatically learns and interprets the meaning of unstructured text such as in documents, emails, and customer interactions ¾ in real-time. The software understands not only the words used, but also the context of the language as well as associated metadata. The self-learning IBM classification software is designed to become more accurate over time without requiring human adjustment. IBM Classification Module for OmniFind Discovery Edition is enabled for Services Oriented Architecture (SOA) environments and can deliver content classification as a service to any content centric application, helping automate processes and decision- making to determine how a piece of content should be used. It can also classify emails for online contact center applications to automatically prioritize certain customer situations over others. The IBM Classification Module for OmniFind Discovery Edition analyzes unstructured and semi-structured content and metadata, allowing previously untapped information to be classified so it can be found and reused. The content can be classified within existing taxonomies or the system can propose new taxonomies.

The IBM Classification Module for IBM FileNet P8 automates tasks such as document management and collaboration. It also was designed to help improve compliance tasks such as declaring records, efficiently performing electronic discovery over content, and monitoring email for risk identification. It provides the ability to expand the reach of content management in a scalable way while helping to manage enterprise taxonomies. The solution can automatically translate between taxonomies of two different systems or between an old taxonomy and a new taxonomy. It also suggests additions to existing taxonomies and can propose entirely new IBM FileNet P8 classification schemas. IBM Classification Module currently supports language processing and identification in languages, including most Western European languages, Japanese, Chinese, and Korean. IBM Classification Module offering for IBM FileNet P8 and IBM Classification Module for OmniFind Discovery Edition are currently available from IBM and IBM Business Partners.

IBM has also announced free technical resources to help insurance companies better access, integrate and manage ACORD standards-based XML data across the enterprise. Database administrators and software developers running IBM DB2 9 data server software can view an interactive demo and download a new DB2 pureXML industry bundle that includes scripts and examples at no charge to begin managing both conventional relational data and ACORD XML messages without having to reformat the XML data for consumption outside of the database. Application developers can download the DB2 pureXML industry bundles and use an interactive demo by visiting the IBM alphaWorks portal. The demo contains Feeds, RESTful services (Web 2.0), XFORMS and web services, and can be accessed through web browsers, feed readers, XFORMS viewers, custom clients, and mashups.

(www.ibm.com; www.acord.org; www.alphaworks.ibm.com/tech/purexml)