Least Information Modeling for Information Retrieval
This work addresses the challenge of improving document ranking in information retrieval, particularly for complex queries, though it appears incremental as it builds on existing probability-based approaches.
The authors tackled the problem of quantifying meaning in information retrieval by proposing a Least Information theory (LIT) to measure semantic quantities based on probability distribution changes, resulting in LIT-based methods that showed very strong performance compared to classic TF*IDF and BM25, especially for verbose queries and hard search topics.
We proposed a Least Information theory (LIT) to quantify meaning of information in probability distribution changes, from which a new information retrieval model was developed. We observed several important characteristics of the proposed theory and derived two quantities in the IR context for document representation. Given probability distributions in a collection as prior knowledge, LI Binary (LIB) quantifies least information due to the binary occurrence of a term in a document whereas LI Frequency (LIF) measures least information based on the probability of drawing a term from a bag of words. Three fusion methods were also developed to combine LIB and LIF quantities for term weighting and document ranking. Experiments on four benchmark TREC collections for ad hoc retrieval showed that LIT-based methods demonstrated very strong performances compared to classic TF*IDF and BM25, especially for verbose queries and hard search topics. The least information theory offers a new approach to measuring semantic quantities of information and provides valuable insight into the development of new IR models.