TREC-9 Cross-Lingual Retrieval at BBN

Abstract

BBN participated only in the cross-language track at TREC-9. We extended the monolingual approach of Miller et al. (1999), which uses hidden Markov models (HMM), by incorporating translation probabilities from Chinese terms to English terms. In our approach, the IR system ranks documents by the probability that a Chinese document D is relevant given an English query Q, P(D is Rel |Q). Using Bayes Rule, and the fact that P(Q) is constant for a given query, and our initial assumption of a uniform a priori probability that a document is relevant, ranking documents according to P(Q|D is Rel) is the same as ranking them according to P(D is Rel|Q). The approach therefore estimates the probability that a query Q is generated, given the document D is relevant.

Open PDF

Document Details

Document Type
Technical Report
Publication Date
Jan 01, 2000
Accession Number
ADA456316

Entities

People

  • Jinxi Xu
  • Ralph Weischedel

Organizations

  • BBN Technologies

Tags

Communities of Interest

  • Materials and Manufacturing Processes

DTIC Thesaurus Topics

  • Abstracts
  • Algorithms
  • Ambiguity
  • Chinese Language
  • Computational Science
  • Hidden Markov Models
  • Hong Kong
  • Information Retrieval
  • Language
  • Markov Models
  • Models
  • Natural Languages
  • Personality
  • Probabilistic Models
  • Probability
  • Standards
  • Translations

Fields of Study

  • Computer science

Readers

  • Computational Linguistics
  • Information Retrieval
  • Statistical inference.