Ray Larson

Professor Emeritus


Information retrieval system design and evaluation


Dr. Larson specializes in information retrieval and database systems, with an emphasis on the system internals. He was involved in the design and development of UC public access online union catalog (MELVYL). He also helped design the algorithms used in the Inktomi web search engine. He is the principal designer of the Cheshire information retrieval system, and active in internation IR evaluations including cross-language evaluations like CLEF and NTCIR.

Current Research

My current research focuses on several related areas of information retrieval and digital libraries, including how to exploit XML structure and content in heterogeneous collections of XML documents, and how to use the evolving computational Grid/Cloud for IR and digital library applications. I am also investigating ways to use the metadata infrastructure of ontologies (thesauri), electronic gazetteers, time period directories and biographical references to provide enhanced access to a variety of information resources and to generate dynamic portals to related information for a searcher's area of interest. In addition I am working on methods for effective geographic information retrieval (IR that uses a geographic dimension) and geotemporal (search taking into account both geographic and time constraints) in a multilingual environment.

Other Research

I am one of the track coordinators for the GeoCLEF track of the Cross-Language Evaluation Forum (CLEF) and for the Heterogeneous Track of the Initiative for the Evaluation of XML Retrieval (INEX).


B.A., English and Comparative Literature, California State University, Fullerton, 1976
M.S.L.S., Library Science, California State University, Fullerton, 1976
Certificate: Library Automation and Information Science, UC Berkeley, 1979
Ph.D., Library and Information Studies, UC Berkeley, 1986

What is the best thing about working at the I School?
Working with incredibly gifted and interesting students.

What Information issues interests you most?
How to integrate the vast resources of the "dark web" -- all of the databases that underlie the search pages on thousands or millions of web sites, and still make the results usable and comprehensible to users.

Last updated:

February 21, 2017