Keikha, Mostafa Crestani, Fabio 2012-07-26 This paper addresses the blog distillation problem, that is, given a user query find the blogs that are most related to the query topic. We model each post as evidence of the relevance of a blog to the query, and use aggregation methods like Ordered Weighted Averaging (OWA) operators to combine the evidence. We show that using only highly relevant evidence (posts) for each blog can result in an effective retrieval system. We also take into account the importance of the posts in a query-based cluster and investigate its effect in the aggregation results. We use prioritized OWA operators and show that considering the importance is effective when the number of aggregated posts from each blog is high. We carry out our experiments on three different data sets (TREC07, TREC08 and TREC09) and show statistically significant improvements over state of the art model called Voting Model. application/pdf https://susi.usi.ch/global/documents/318525 https://n2t.net/ark:/12658/srd1318525 https://susi.usi.ch/documents/318525/files/crestani_IPM_2012.pdf eng info:eu-repo/semantics/altIdentifier/doi/10.1016/j.ipm.2011.02.001 info:eu-repo/semantics/altIdentifier/ark/12658/srd1318525 info:eu-repo/semantics/openAccess License undefined Information processing and management: an international journal. - Elsevier. - 2012, vol. 48, no. 3, p. 467–475 Blog retrieval aggregation methods ordered weighted averaging operators info:eu-repo/classification/udc/004 Linguistic aggregation methods in blog retrieval http://purl.org/coar/resource_type/c_6501