Using language models for flat text queries in XML retrieval
This paper presents a language modeling system for ranking flat text queries against a collection of structured documents. The retrieval system, built using the Lemur toolkit, produces probability estimates that arbitrary document components generated the query. This paper describes storage mechanisms and retrieval algorithms for the evaluation of unstructured queries over XML documents. The paper includes retrieval experiments using a generative language model on the content only topics of the INEX testbed, demonstrating the strengths and flexibility of language modeling to a variety of problems. We also describe index characteristics, running times, and the effectiveness of the retrieval algorithm.
Preview
Cite
Citation style:
Could not load citation form.