Preference Metrics, LLC

Preference Metrics LLC Logo

Brief Description

The company's proprietary numerical method, called a principal component analysis, delivered computer-based recommendations of selective information based on an individual's personal preferences such as books, invoices, stocks, or toys. Preference Metrics, LLC was founded by UC Berkeley Professor Ken Goldberg, based on his filtering algorithm, named Eigentaste, that uses universal queries to elicit real-valued user ratings on a common set of items, and applies principal component analysis (PCA) to the resulting dense subset of the ratings matrix.


  • 1999. Company founded


Kenneth Yigael Goldberg