Given an ambiguous or underspecified web search query, search result diversification aims at accomodating different user intents within a single "entry-point" result page. However, some intents are informational, for which many relevant pages may help, while others are navigational, for which only one web page is required. We propose new evaluation metrics for search result diversification that considers this distinction, as well as the condordance test for comparing the intuitiveness of a given pair of metrics quantitatively. Our main experimental findings are: (a) In terms of discriminative power which reflects statistical reliability, the proposed metrics, DIN#-nDCG and P+Q#, are comparable to intent recall and D#-nDCG, and possibly superior to α-nDCG; (b) In terms of the concordance test which quantifies the agreement of a diversity metric with a gold standard metric that represents a basic desirable property, DIN#-nDCG is superior to other diversity metrics in its ability to reward both diversity and relevance at the same time. Moreover, both D#-nDCG and DIN#-nDCG significantly outperform α-nDCG in their ability to reward diversity, to reward relevance, and to reward both at the same time. In addition, we demonstrate that the randomised Tukey's Honestly Significant Differences test that takes the entire set of available runs into account is substantially more conservative than the paired bootstrap test that only considers one run pair at a time, and therefore recommend the former approach for significance testing when a set of runs is available for evaluation.
ASJC Scopus subject areas
- コンピュータ サイエンス（全般）