A Novel Graph-based Diversity-aware Rank Fusion Method Applied to Image Metasearch

• José Solenir L. Figuerêdo Universidade Estadual de Feira de Santana http://orcid.org/0000-0003-1892-3455
• Ana Lúcia L. Marreiros Maia Universidade Estadual de Feira de Santana
• Rodrigo T. Calumby Universidade Estadual de Feira de Santana

DOI: https://doi.org/10.5753/sbbd.2023.233400

Resumo

While search result diversification is used to handle ambiguous or underspecified queries, rank aggregation is a widely used approach in metasearch. However, current aggregation methods assume that the input rankings are built only according to the relevance of the items, disregarding the inter-relationship between images in each ranking. Hence, these methods tend to be inadequate for diversity-oriented retrieval. In this work, we introduce a diversity-aware rank fusion method that is validated in the context of diverse image metasearch. The experimental findings indicate that the proposed method significantly improves the overall diversity of metasearch results, in comparison to the state-of-the-art positional and score-based fusion methods.

Palavras-chave: Metasearch, Ranking Aggregation, Information Retrieval, diversity-oriented retrieval, fusion method

Referências

Aslam, J. A. and Montague, M. (2001). Models for metasearch. In Proceedings of the 24th Annual International ACM SIGIR Conference on Research and Development in Information Retrieval, page 276–284, New York, NY, USA. Association for Computing Machinery.

Calumby, R. T., Gonçalves, M. A., and da Silva Torres, R. (2017). Diversity-based interactive learning meets multimodality. Neurocomputing, 259:159–175. Multimodal Media Data Understanding and Analytics.

Dwork, C., Kumar, R., Naor, M., and Sivakumar, D. (2001). Rank aggregation methods for the web. In Proceedings of the 10th International Conference on World Wide Web, page 613–622, New York, NY, USA. ACM.

Figuerêdo, J. and Calumby, R. (2019). Unsupervised rank fusion for diverse image metasearch. In Anais do XXXIV Simpósio Brasileiro de Banco de Dados, pages 265– 270, Porto Alegre, RS, Brasil. SBC.

Liang, S., Ren, Z., and de Rijke, M. (2014). Fusion helps diversification. SIGIR ’14, page 303–312, New York, NY, USA. ACM.

McDonald, G., Macdonald, C., and Ounis, I. (2022). Search results diversification for effective fair ranking in academic search. Information Retrieval Journal, 25(1):1–26.

Ramírez-de-la-Rosa, G. et al. (2018). Overview of the multimedia information processing for personality & social networks analysis contest. In ICPR’18, Beijing, China, August 20-24, pages 127–139.

Vargas Muñoz, J. A., da Silva Torres, R., and Gonçalves, M. A. (2015). A soft computing approach for learning to aggregate rankings. page 83–92, New York, NY, USA. ACM.

Xu, C. andWu, S. (2017). The early fusion strategy for search result diversification. ACM TUR-C ’17, New York, NY, USA. ACM.

Yigit-Sert, S., Altingovde, I. S., Macdonald, C., Ounis, I., and Özgür Ulusoy (2020). Supervised approaches for explicit search result diversification. Information Processing & Management, 57(6):102356.