Journal article

Linguistic aggregation methods in blog retrieval

  • Keikha, Mostafa Facoltà di scienze informatiche, Università della Svizzera italiana, Svizzera
  • Crestani, Fabio Facoltà di scienze informatiche, Università della Svizzera italiana, Svizzera
    26.07.2012
Published in:
  • Information processing and management: an international journal. - Elsevier. - 2012, vol. 48, no. 3, p. 467–475
English 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.
Language
  • English
Classification
Computer science
License
License undefined
Identifiers
Persistent URL
https://susi.usi.ch/usi/documents/318525
Statistics

Document views: 14 File downloads:
  • crestani_IPM_2012.pdf: 4