eWEEK content and product recommendations are editorially independent. We may make money when you click on links to our partners. Learn More.
IBM is previewing its next mission in enterprise search—letting third parties develop industry applications on its search platform.
By opening its corporate search technology to partners, IBM wants to turn enterprise search into more than a way to find documents, e-mails or data by keyword. It also wants search engines to extract meaning from unstructured data, said Nelson Mattos, IBMs vice president of information integration.
To that end, the Armonk, N.Y., company is developing a new version of its WebSphere Information Integrator. The release, code-named Serrano and already in a restricted beta, is due for a public beta later this release and a general release in the fourth quarter, Mattos said.
Information Integrator, formerly code-named Masala and launched last year, is the companys broad suite of offerings to federate and manage structured and unstructured data. In November, IBM specifically addressed enterprise search with the release of the OmniFind Edition of the Information Integrator.
“The fact that enterprise searches are available and accessible today has made information accessible to a lot of people,” Mattos said. “However, the search engines have had no way to extract meaning and knowledge out of the documents they search.”
Underlying OmniFinds search capabilities is an architecture IBM calls UIMA (Unstructured Information Management Architecture). UIMA combines natural language processing, a common technique in search engines, with data mining and analytics to derive a deeper meaning from documents and files than their keywords, Mattos said.
Today, OmniFind offers broad enterprise search capabilities across corporate intranets, public Web sites, relational databases and content-management systems, but it is not tuned to specific industries.
By letting partners tap into OmniFind and UIMA, IBM hopes to attract industry experts who can supply information such as taxonomies to decipher even more meaning from documents, e-mails and other stored materials, Mattos said. Then searches also can answer industry-specific questions.
Take health care. Enterprises could use enterprise search to analyze millions of patient case histories to discover dangerous drug interactions, Mattos said.
Along with the partner plans in Serrano, the release is slated to expand on Information Integrators enterprise search and content integration features by supporting more sources of information and improving performance, according to IBM. It also will add new deployment, development and maintenance capabilities.
OmniFind also is serving as the embedded search technology in other IBM software. OmniFind is currently wrapped into the IBM WebSphere Portal product, Mattos said.
While IBMs search aspirations are growing, the company does not plan to compete against the likes of Google or Microsoft in the Web search arena, Mattos said.
“Our whole focus is addressing the problems of enterprises to find relevant information across what they have, as well as information that is available publicly,” Mattos said.