Unsupervised Rank Fusion for Diverse Image Metasearch

  • José Solenir L. Figuerêdo UEFS
  • Rodrigo Tripodi Calumby UEFS

Resumo


For a given query and a set of images ranked lists retrieved from multiple search engines, the metasearch technique aims at combining these lists to build an unified ranking with improved relevance. Rank aggregation is an approach that has been widely used to support this task. This paper investigates the use of rank aggregation methods in the metasearch scenario for diverse image retrieval. Although metasearch systems are usually driven by the relevance of the final result, the impact on diversification has also been analyzed. The experimental findings suggest metasearch based on rank aggregation allows significant improvements, both in terms of relevance and diversity.

Palavras-chave: Rank Aggregation, Diversity, Image Retrieval, Metasearch

Referências

Bahri, S., Zoghlami, N., Abed, M., and Tavares, J. M. R. S. (2019). Big data for healthcare: A survey. IEEE Access, 7:7397–7408. DOI: https:/doi.org/10.1109/ACCESS.2018.2889180.

Calumby, R. T., Gonc¸alves, M. A., and da Silva Torres, R. (2016). On interactive learningto- rank for IR: overview, recent advances, challenges, and directions. Neurocomputing, 208:3–24. DOI: https:/doi.org/10.1016/j.neucom.2016.03.084.

Calumby, R. T., Gonc¸alves, M. A., and da Silva Torres, R. (2017). Diversity-based interactive learning meets multimodality. Neurocomputing, 259:159–175. DOI: https:/doi.org/10.1016/j.neucom.2016.08.129.

Farah, M. and Vanderpooten, D. (2007). An outranking approach for rank aggregation in information retrieval. In SIGIR’07, Amsterdam, The Netherlands, July 23-27, pages 591–598. DOI: https:/doi.org/10.1145/1277741.1277843.

Ionescu, B., Gˆınsca, A., Boteanu, B., Lupu, M., Popescu, A., and Müller, H. (2016a). Div150multi: a social image retrieval result diversification dataset with multi-topic queries. In MMSys’16, Klagenfurt, Austria, May 10-13, pages 46:1–46:6. DOI: https:/doi.org/10.1145/2910017.2910620.

Ionescu, B., Gˆınsca, A., Zaharieva, M., Boteanu, B., Lupu, M., and Müller, H. (2016b). Retrieving diverse social images at mediaeval 2016: Challenge, dataset and evaluation. In MediaEval’16 Workshop, Hilversum, The Netherlands, October 20-21. DOI: https:/doi.org/

Ionescu, B., Popescu, A., Lupu, M., Ginsca, A., Boteanu, B., and Müller, H. (2015). Div150cred: A social image retrieval result diversification with user tagging credibility dataset. In MMSys’15, Portland, USA, March 18-20, pages 207–212. DOI: https:/doi.org/10.1145/2713168.2713192.

Ionescu, B., Radu, A., Menendez, M., Müller, H., Popescu, A., and Loni, B. (2014). Div400: a social image retrieval result diversification dataset. In MMSys’14, Singapore, Mar 19-21, pages 29–34. DOI: https:/doi.org/10.1145/2557642.2563670

Li, L. T., Pedronette, D. C. G., Almeida, J., Penatti, O. A. B., Calumby, R. T., and Torres, R. d. S. (2014). A rank aggregation framework for video multimodal geocoding. Multimedia Tools and Applications, 73(3). DOI: https:/doi.org/10.1007/s11042-013-1588-4.

Liang, S., Ren, Z., and de Rijke, M. (2014). Fusion helps diversification. In SIGIR’14, NY, USA, pages 303–312. ACM. DOI: https:/doi.org/10.1145/2600428.2609561

McDonald, K. and Smeaton, A. F. (2005). A comparison of score, rank and probability based fusion methods for video shot retrieval. In CIVR’05, Singapore, July 20-22, Proceedings, pages 61–70. DOI: https:/doi.org/10.1007/11526346_10.

Muñoz, J. A. V., da Silva Torres, R., and Gonçalves, M. A. (2015). A soft computing approach for learning to aggregate rankings. In CIKM’15, Melbourne, Australia, October 19 - 23, pages 83–92. DOI: https:/doi.org/10.1145/2806416.2806478

Ramírez-de-la-Rosa, G., Villatoro, E., Ionescu, B., Escalante, H. J., Escalera, S., Larson, M., M¨uller, H., and Guyon, I. (2018). Overview of the multimedia information processing for personality & social networks analysis contest. In ICPR’18, Beijing, China, August 20-24, Revised Selected Papers, pages 127–139. DOI: https:/doi.org/10.1007/978-3-030-05792-3_12.

Santos, R. L. T., MacDonald, C., and Ounis, I. (2015). Search result diversification. Foundations and Trends in Information Retrieval, 9(1):1–90. DOI: https:/doi.org/10.1561/1500000040.

Wu, S., Huang, C., Li, L., and Crestani, F. (2019). Fusion-based methods for result diversification in web search. Information Fusion, 45:16–26. DOI: https:/doi.org/10.1016/j.inffus.2018.01.006.

Xu, C. and Wu, S. (2017). The early fusion strategy for search result diversification. In ACM TUR-C’17, New York, USA, pages 47:1–47:6. DOI: https:/doi.org/10.1145/3063955.3064803.

Zhai, C. X., Cohen,W.W., and Lafferty, J. (2003). Beyond independent relevance: Methods and evaluation metrics for subtopic retrieval. In ACM SIGIR, Toronto, Canada, pages 10–17. DOI: https:/doi.org/10.1145/860435.860440.
Publicado
26/11/2019
Como Citar

Selecione um Formato
FIGUERÊDO, José Solenir L.; CALUMBY, Rodrigo Tripodi. Unsupervised Rank Fusion for Diverse Image Metasearch. In: SIMPÓSIO BRASILEIRO DE BANCO DE DADOS (SBBD), 34. , 2019, Fortaleza. Anais [...]. Porto Alegre: Sociedade Brasileira de Computação, 2019 . p. 265-270. ISSN 2763-8979. DOI: https://doi.org/10.5753/sbbd.2019.8834.