A decision mechanism for the selective combination of evidence in topic distillation

The combination of evidence can increase retrieval effectiveness. In this paper, we investigate the effectiveness of a decision mechanism for the selective combination of evidence in the context of topic distillation. We introduce two measures of a query's broadness and use them to decide which combinations of evidence are more appropriate for each query. The results from our experiments show that there is a statistically significant association between the output of the decision mechanism and the relative effectiveness of the different combinations of evidence. Moreover, we show that the proposed methodology can be applied in an operational setting, where relevance information is not available, by setting the decision mechanism's thresholds automatically. Keywords Topic distillation, decision mechanism, selective combination of evidence, query scope, aggregates

• Type: Journal
• Year: 2006
• Where: IR
• Authors: Vassilis Plachouras, Fidel Cacheda, Iadh Ounis
• Keywords: Decision Mechanism, IR 2006, Natural Language Processing, Selective Combination, Topic Distillation