Zycus announced the release of AutoClass, a Universal Standard Products and Services Classification (UNSPSC) classification solution for e-Catalogs. It is an automated solution for classifying products into one or more of 13,000 categories. AutoClass’s capabilities will simplify catalog integration for buyers and exchanges, provide the backbone for catalog searches, and assist enterprises in accurately analyzing their spending patterns by commodity and supplier. It will also provide a common platform for mapping disparate supplier and buyer schemas to the industry-standard UNSCPC schema. AutoClass automates UNSPSC-based classification through a Bayesian Inference Engine that extracts the concepts and then statistically links the concepts to product categories. The speed of automatic classification (100,000 documents per hour) enables rapid supplier enablement and provides a way of replacing manual coding. The AI based classification engine is trained on millions of product items where it discerns statistically significant patterns. Built-in UNSPSC libraries and embedded domain intelligence facilitate automatic and accurate classification of the supplier’s/client’s schema into UNSPSC, while version control is made possible through continuous update of UNSPSC. AutoClass can support other schemas such as eClass and client specific schemas through additional libraries. AutoClass is language and domain independent, and can process e-Catalogs on a real time basis using a web interface. www.zycus.com

Share