Software Classifies Electronic Catalogs

Automated catalog content solutions provider Zycus Inc. will announce next week AutoClass, new classification software for e-catalogs that is designed to help standardize e-procurement content for exchanges and e-marketplaces.

AutoClass, which conforms to the Universal Standard Products and Services Classification code, classifies products into one or more of 13,000 categories, according to officials, in Santa Clara, Calif.

The classification capabilities of AutoClass should go a ways to simplify e-catalog integration for buyers and exchanges, as it provides a common platform for mapping disparate supplier and buyer schemas to the UNSPSC schema. The new software also supports additional schemas, including eClass, and can support customer-specific standards.

While UNSPSC is an emerging standard categorization structure for products and services, the majority of suppliers and buyers participating in e-commerce initiatives still have their own categorization structure.

AutoClass automates UNSPSC-based classification through a proprietary artificial intelligence engine called Bayesian Inference Engine. The engine extracts and statistically links product concepts to product categories. Essentially, the AI classification engine is trained on millions of items where it discerns statistically significant patterns, according to officials.

The engine is able to automatically classify 100,000 documents per hour.

Because UNSPSC is an emerging standard, there is a bi-monthly update of the UNSPSC libraries embedded in AutoClass.

AutoClass is available as a hosted service offered by Zycus, or through traditional software licensing.

The Zycus E-catalogs suite also includes CatSummIT, enterprise catalog management software that enables self-service management of supplier catalogs, and Zycat Content Refinery, an outsourced service for buyers to aggregate and implement multiple supplier catalogs.