Applying Maximum Entropy to Known-Item Email Retrieval
It is becoming increasingly common in information retrieval to combine evidence from multiple resources to compute the retrieval status value of documents. Although this has led to considerable improvements in several retrieval tasks, one of the outstanding issues is estimation of the respective weights that should be associated with the different sources of evidence. In this paper we propose to use maximum entropy in combination with the limited memory LBFG algorithm to estimate feature weights. Examining the effectiveness of our approach with respect to the known-item finding task of enterprise track of TREC shows that it significantly outperforms a standard retrieval baseline and leads to competitive performance.
KeywordsInformation Retrieval Maximum Entropy Maximum Entropy Method Retrieval Task Maximum Entropy Principle
Unable to display preview. Download preview PDF.
- 1.Berger, A.L., Della Pietra, V.J., Della Pietra, S.A.: A maximum entropy approach to natural language processing. Comput. Linguist. 22(1), 39–71 (1996)Google Scholar
- 3.Craswell, N., de Vries, A., Soboroff, I.: Overview of the trec-2005 enterprise track. In: Proceedings of the 14th Text REtrieval Conference (2006)Google Scholar
- 7.Lalmas, M.: Uniform representation of content and structure for structured document retrieval. In: 20th SGES International Conference on Knowledge Based Systems and Applied Artificial Intelligence (2000)Google Scholar
- 8.Monz, C.: From Document Retrieval to Question Answering. PhD thesis, University of Amsterdam (2003)Google Scholar
- 9.Nallapati, R.: Discriminative models for information retrieval. In: Proceedings of the 27th Annual International ACM SIGIR Conference on Research and Development in Information Retrieval, pp. 64–71 (2004)Google Scholar
- 12.Ogilvie, P., Callan, J.: Experiments with language models for known-item finding of e-mail messages. In: Proceedings of the Fourteenth Text Retrieval Conference (TREC-14) (2005)Google Scholar
- 15.Tsikrika, T., Lalmas, M.: Combining evidence from web retrieval using the inference network model - an experimental study. Information Processing & Management, Special Issue in Bayesian Networks and Information Retrieval 40(5), 751–772 (2004)Google Scholar