REVIEWER: Teragram’s Solutions: The Power Behind Reed Business’

BACKGROUND: RB is a leading international business-to-business (B2B) media company, with a portfolio of thousands of news and information products, awards, conferences and directories. Its industry publications cover 26 markets, including media, electronics, construction, aerospace, agriculture, banking and computing.

PLATFORMS: Teragram operates on Windows, and FAST operates on Linux.

PROBLEM SOLVED: In 2006, RB started to build in order to create vertical search engines that filter the Web’s information into relevant subcategories within the B2B sector. The massive amount of information available on the Web needed to be broken down into manageable groups so that Zibb users could easily digest the information. RB’s Search division had to talk to users to find out both what was working in Internet search and what sorts of features they were not willing to accept yet. Essentially, RB was fighting the Google effect of search and facing the general public’s perception of the standard search interface from Google as the norm, which includes a search box, a set of Web links and a brief description of the site. Zibb attempted to link relevant information that would help research and purchasing decisions in specific vertical industries, such as finance and electronics. Several of the features introduced on the site, such as thumbnail photos, received mixed reviews, because users weren’t accustomed to multimedia in their search experience. The biggest challenge was striking the right balance between overwhelming the users and providing them with enough relevant information for a complete and pleasant experience. In order to solve that problem, RB relied upon technologies from both Teragram and FAST to provide automatic metadata generation, categorization and taxonomy management for its vertical search sites. RB replicated what was successful about the first iteration of Zibb - the breakdown of content into visually appealing tabs - and added an additional layer of demarcation. Within the four main tabs, Zibb separates content by company name, related topics and related searches. An example of Teragram’s technology at work on Zibb on Demand is the redesigned If a user searches an actress’s name, he/she can find not only articles from, but other entertainment Web sites. Within the articles, the user will find links to related stories. Listed at the top of the page are the actress’s film projects, associated people and companies. Zibb on Demand allows users to filter content in a way that’s logical to them. From a technical standpoint, the metatagging that occurs within each individual story (embedded linking) makes it easier to cross-reference content from the Web and other RB sites. Tagging is done using the more than 5 million entity files within RB and Teragram taxonomies, which are prebuilt into the system. As a result of the improvements, traffic has increased on and Zibb on Demand. Anecdotal response from the sites’ users has been overwhelmingly positive. Users from the 130 Zibb-enabled sites are pleased with the changes and are looking forward to what RB can do next.

PRODUCT FUNCTIONALITY: Teragram’s Taxonomy Manager and Entities and Events Extraction enable RB to accurately tag more than 5 million entity files within the RB taxonomies. Teragram’s solutions aid in RB’s ability to organize the information on its Web sites and improve the accuracy and overall experience for the user.

STRENGTHS: Not only is Teragram’s solution able to scale to process hundreds of millions of documents, but it can also build custom-made solutions to our specific needs while handling our large amount of content.

WEAKNESSES: We would like to see the user interface updated so that it is more workflow oriented and would also like a more accurate report for error tracking.

SELECTION CRITERIA: RB chose Teragram because of its ability to provide automated categorization, accuracy and speed that was necessary for the Zibb project.

DELIVERABLES: Teragram outputs data filters and taxonomies in the FAST pipeline, which we then use as part of the search functionality on Zibb and Zibb on Demand.

VENDOR SUPPORT: The relationship with Teragram has been fantastic, from the sales process all the way to support. It is great to see a company so enthusiastic about its technology and excited about the suggested enhancements. Our relationship with Teragram has been our best supplier relationship.

DOCUMENTATION: The documentation is extensive and simple to follow.

Teragram Categorizer and Taxonomy Manager
Teragram Corporation
10 Fawcett Street
Cambridge, MA 02138
(617) 576-6800

Register or login for access to this item and much more

All Information Management content is archived after seven days.

Community members receive:
  • All recent and archived articles
  • Conference offers and updates
  • A full menu of enewsletter options
  • Web seminars, white papers, ebooks

Don't have an account? Register for Free Unlimited Access