Generation and Evaluation of Indexes for Chemistry Articles

摘要

This paper describes AIMS (Assisted Indexing atMississippi State), a system that aids human document analystsin the assignment of indexes to physical chemistry journalarticles. There are two major components of AIMS—a naturallanguage processing (NLP) component and an index generation (IG)component. The focus of this article is the IG. We describethe techniques and structures used by the IG in the selection ofappropriate indexes for a given article. We also describe theresults of evaluations of the system in terms of recall,precision, and overgeneration. We provide a description of agraphical user interface that we have developed for AIMS.Finally, we discuss future work.