Profile Diversity in Search and Recommendation

Maximilien Servajean (1, 2) , Esther Pacitti (1, 3) , Sihem Amer-Yahia (4) , Pascal Neveu

1 IBC - Institut de Biologie Computationnelle
2 ADVANSE - ADVanced Analytics for data SciencE
3 ZENITH - Scientific Data Management
4 HADAS - Heterogeneous and Adaptive distributed DAta management Systems

We investigate profile diversity, a novel idea in searching scientific documents. Combining keyword relevance with popularity in a scoring function has been the subject of different forms of social relevance. Content diversity has been thoroughly studied in search and advertising, database queries, and recommendations. We believe our work is the first to investigate profile diversity to address the problem of returning highly popular but too-focused documents. We show how to adapt Fagin's threshold-based algorithms to return the most relevant and most popular documents that satisfy content and profile diversities and run preliminary experiments on two benchmarks to validate our scoring function.

Mots clés: Recommendation; Diversity; Top-k

Domaines: Recherche d'information [cs.IR]

Maximilien Servajean, Esther Pacitti, Sihem Amer-Yahia, Pascal Neveu. Profile Diversity in Search and Recommendation. RR-13002, Lirmm, University of Montpellier. 2013. ⟨lirmm-00794814⟩