Relevance stability in blog retrieval
Mostafa Keikha, Shima Gerani, Fábio Crestani · 2011
This paper investigates blog distillation where the goal is to rank blogs according to their recurrent relevance to the topic of the query. One of the main features of blogs is their relation to time but this important feature is under-utilized in the current blog retrieval methods. We propose a probabilistic framework to measure the stability of blogs relevance over time. We then study the effect of the proposed stability measure in the blog retrieval performance. We evaluate the proposed framework on the standard TREC Blog08 collection. The results show statistically significant improvements over state of the art models.