Loading...
Please wait, while we are loading the content...
Similar Documents
Why Bigger Windows Are Better Than Smaller Ones (1997)
| Content Provider | CiteSeerX |
|---|---|
| Author | And, Ron Papka Papka, Ron Allan, James |
| Abstract | We investigate the use of multi-term query concepts to improve the performance of textretrieval systems that accept "natural-language" queries. A relevance feedback process is explained that massively expands an initial query with single and multi-term concepts. The multi-term concepts are modelled as a set of words appearing within windows of varying sizes. Experimental results suggest that windows of larger size yield improvements in average precision. The reason for this improvement is explored. A window size relaxation process that yields a significant reduction in expanded query size with no performance loss is also described. 1 Introduction The general intuition about multi-term concepts is that "the closer a set of intersecting terms, the more likely they are to indicate relevance" [4, 5]. Our experiments indicate that, contrary to intuition, in the context of relevance feedback, performance gains are obtained by using query concepts that are modelled as pairs of terms t... |
| File Format | |
| Language | English |
| Publisher Date | 1997-01-01 |
| Publisher Department | Department of Computer Science, University of Massachusetts |
| Access Restriction | Open |
| Subject Keyword | Bigger Window Better Smaller One Multi-term Concept Relevance Feedback General Intuition Significant Reduction Initial Query Average Precision Performance Loss Multi-term Query Concept Window Size Relaxation Process Expanded Query Size Query Concept Size Yield Improvement Textretrieval System Performance Gain Experimental Result Natural-language Query Relevance Feedback Process |
| Content Type | Text |
| Resource Type | Technical Report |
