Distance Calibration in Metric Access Methods through Relevance Feedback

• Renato Gomes Marcacini University of São Paulo http://orcid.org/0000-0002-0380-5483
• Willian Dener de Oliveira University of São Paulo https://orcid.org/0000-0002-6752-2060
• Agma Juci Machado Traina University of São Paulo https://orcid.org/0000-0003-4929-7258

DOI: https://doi.org/10.5753/sbbd.2023.231731

Abstract

Traditionally Metric Access Methods (MAM) use fixed distance functions to build the metric trees, which in turn prevents a MAM from being able to index elements using two or more distance functions in the same index. A vector of weights correctly learned by Relevance Feedback (RR), allows weighting distance functions and improving data semantics, improving the query accuracy. This work introduces the Tuning Metrics Relevance Feedback (TMRF) method, which shows that when using weighted distance functions in MAMs, the retrieval efficiency of these structures becomes up to 70% more efficient than sequential strategies, in addition to obtaining a gain of 42% through RR learning.

Keywords: Metric Access Methods, Relevance Feedback, Weighted Distance Function, Distance Calibration

References

Ahmed, A. (2020). Implementing relevance feedback for content-based medical image retrieval. IEEE Access, 8:79969–79976.

Bressan, R. S., Bugatti, P. H., and Saito, P. T. (2019). Breast cancer diagnosis through active learning in content-based image retrieval. Neurocomputing, 357:1–10.

Chang, Y.-J., Kamataki, K., and Chen, T. (2009). Mean shift feature space warping for relevance feedback. In 2009 16th IEEE International Conference on Image Processing (ICIP), pages 1849–1852. IEEE.

Giakoumoglou, N. (2021). Pyfeats: Open source software for image feature extraction. https://github.com/giakou4/pyfeats.

Guo, S., Ji, Y., Zhang, C., Xu, C., and Xu, J. (2020). vcbir: A verifiable search engine for content-based image retrieval. In 2020 IEEE 36th International Conference on Data Engineering (ICDE), pages 1730–1733. IEEE.

Kim, D.-H. and Chung, C.-W. (2003). Qcluster: relevance feedback using adaptive clustering for content-based image retrieval. In Proceedings of the 2003 ACM SIGMOD international conference on Management of data, pages 599–610.

Kumaran, K., Papageorgiou, D. J., Takac, M., Lueg, L., and Sahinidis, N. V. (2021). Active metric learning for supervised classification. Computers & Chemical Engineering, 144:107132.

Li, Z., Zhang, X., Müller, H., and Zhang, S. (2018). Large-scale retrieval for medical image analytics: A comprehensive review. Medical image analysis, 43:66–84.

Mohanan, A. and Raju, S. (2017). A survey on different relevance feedback techniques in content based image retrieval. Int. Res. J. Eng. Technol, 4(2):582–585.

Rui, Y., Huang, T. S., Ortega, M., and Mehrotra, S. (1998). Relevance feedback: a power tool for interactive content-based image retrieval. IEEE Transactions on circuits and systems for video technology, 8(5):644–655.

Silva, M. P. d. (2009). Processamento de consultas por similaridade em imagens médicas visando à recuperação perceptual guiada pelo usuário. PhD thesis, Universidade de São Paulo.

Tian, D. (2018). A review on relevance feedback for content-based image retrieval. J. Inf. Hiding Multim. Signal Process., 9(1):108–119.

Tyagi, V. (2018). Understanding digital image processing. CRC Press.

Van der Maaten, L. and Hinton, G. (2008). Visualizing data using t-sne. Journal of machine learning research, 9(11).