Diversification for multi-domain result sets

Conference Paper (2012)
Authors

Alessandro Bozzon (Politecnico di Milano)

Marco Brambilla (Politecnico di Milano)

Piero Fraternali (Politecnico di Milano)

Marco Tagliasacchi (Politecnico di Milano)

Affiliation
External organisation
To reference this document use:
https://doi.org/10.1007/978-3-642-31753-8_10
More Info
expand_more
Publication Year
2012
Language
English
Affiliation
External organisation
Volume number
7387 LNCS
Pages (from-to)
137-152
ISBN (print)
9783642317521
DOI:
https://doi.org/10.1007/978-3-642-31753-8_10

Abstract

Multi-domain search answers to queries spanning multiple entities, like "Find a hotel in Milan close to a concert venue, a museum and a good restaurant", by producing ranked sets of entity combinations that maximize relevance, measured by a function expressing the user's preferences. Due to the combinatorial nature of results, good entity instances (e.g., five stars hotels) tend to appear repeatedly in top-ranked combinations. To improve the quality of the result set, it is important to balance relevance with diversity, which promotes different, yet almost equally relevant, entities in the top-k combinations. This paper explores two different notions of diversity for multi-domain result sets, compares experimentally alternative algorithms for the trade-off between relevance and diversity, and performs a user study for evaluating the utility of diversification in multi-domain queries.

No files available

Metadata only record. There are no files for this record.