DARPA Unfamiliar Metadata Project

Research Summary

The objective of this project is to link ordinary language queries to unfamiliar indexes and classifications. Databases quite often use idiosyncratic classifications and unfamiliar terminology in their indexes. However, searchers are often unfamiliar with schemes and terminologies, and often experience difficulty choosing search terms that are actually used to index the database. The results are unsuccessful searches and a diminished return on investment for the information service provider.

In this project, "Entry Vocabulary Indexes" are built to lead from a searcher's query posed in ordinary language to the vocabulary used in the database. The Entry Vocabulary Index responds to the ordinary language query with a ranked list of terms likely to represent what is sought in the database of interest. The aim of "Entry Vocabulary Indexes" is to facilitate a more direct connection between ordinary language queries ("query vocabularies") and indexing terms ("entry vocabularies") actually used to organize information in a variety of databases.