Abstract
In this article we propose a method for information retrieval based on relational Multi-Criteria Decision Making. We assume that a user cannot define precise search criteria so that these criteria must be found based on the user’s assessment of several sample alternatives (‘alternatives’ here are database records, e.g. images). This situation is common in Content-based Image Retrieval, where it is easier for a user to indicate relevant images than to describe a proper query, especially in formal language. The proposed algorithm for the elicitation of criteria is based on ELECTRE III—a method originally designed for ranking a set of alternatives according to defined criteria. In our algorithm, however, the direction of reasoning is reversed: we start with several sample alternatives that have been assigned a rank by the user and then we select criteria that are compatible (in the sense of ELECTRE methodology) with the user’s preferences expressed on a sample set. Then, having determined the user’s criteria, we apply classical ELECTRE III to retrieve the relevant solutions from the database. We implemented the method in Matlab and tested it on the Microsoft Cambridge Image Database.
Similar content being viewed by others
Notes
The first ELECTRE method (ELECTRE I) was published in [14]
The database is available for non-commercial purposes at:http://research.microsoft.com/en-us/projects/objectclassrecognition/default.htm
In our interface N is fixed but the proposed method can be used with a larger set of sample images and the value of N does not influence significantly the computational complexity of the algorithm.
In CBIR systems with large databases the page zero problem is reduced by applying textual search to obtain the initial set of images: Grigorova, A. and F. D. Natale (2006).
The actual number of criteria found by the algorithm is greater, e.g. when looking for large objects, found criteria are: area, width, height, etc.
We used 1 byte per element (uint8 type, with scaling between [0..1] and [0..255])
References
El Sayad I, Martinet J, Urruty T, Djeraba C (2012) Toward a higher-level visual representation for content-based image retrieval. Multimed Tool Appl 60:455–482
Everingham M, Gool LV, Williams CKI, Winn J, Zisserman A (2010) The Pascal Visual Object Classes (VOC) Challenge. Int J Comput Vis 88(2):303–338
Figueira J, Mousseau V, Roy B (2005) ELECTRE methods. In: Figueira J, Greco S, Ehrgott M (eds) Multiple criteria decision analysis: State of the art surveys. Springer, Boston, pp 133–162
Heesch D, Rüger S (2007) Interaction models and relevance feedback in image retrieval. In: Zhang Y-J (ed) Semantic-based visual information retrieval. IRM Press, Hershey, London, pp 160–186
Lew MS (2001) Principles of visual information retrieval. Springer, London
Lew MS, Sebe N, Djeraba C, Jain R (2006) Content-based multimedia information retrieval—state of art and challenges. ACM Trans Multimed Comput Commun Appl 2(1):1–19
Li HF, Wang JJ (2007) An improved ranking method for ELECTRE III. In: Zhou H (ed) International Conference on Wireless Communication, Networking and Mobile Computing (WiCOM ′07), Shanghai. pp 6659–6662. doi:10.1109/WICOM.2007.1634
Lian Z, Godil A, Bustos B, Daoudi M, Hermans J (2013) A comparison of methods for non-rigid 3D shape retrieval. Pattern Recogn 46(1):449–461
Mousseau V, Slowinski R (1998) Inferring an ELECTRE TRI Model from assignment examples. J Glob Optim 12(2):157–174
Mousseau V, Slowinski R, Zielniewicz P (2000) A user-oriented implementation of the ELECTRE-TRI method integrating preference elicitation support. Comput Oper Res 26(7–8):757–777
Muneesawang P, Guan L (2006) Multimedia database retrieval: A human-centered approach. Springer Series on Signals and Communication Technology. Springer
Rotter P (2012) Multimedia information retrieval based on pairwise comparison and its application to visual search. Multimed Tool Appl 60(3):573–587. doi:10.1007/s11042-011-0828-8
Rotter P, Skulimowski AMJ (2008) A new approach to interactive visual search with RBF networks based on preference modelling. In: Rutkowski L et al. (eds) LNAI 5097. Springer-Verlag Berlin Heidelberg, pp 861–873
Rotter P, Skulimowski AMJ (2009) Preference extraction in image retrieval. In: Ma Z (ed) Artificial intelligence for maximizing content based image retrieval. Idea Group Inc, Harshley, pp 235–260
Roy B, Vanderpooten D (1997) An overview on The European school of MCDA: emergence, basic features and current works. Eur J Oper Res 99(1):26–27
Schroff F, Criminisi A, Zissermann A (2011) Harvesting image databases from the web. IEEE Trans Pattern Anal Mach Intell 33(4):754–766
Shen X, Wang Z, Cho S-Y (2008) Mining user hidden semantics from image content for image retrieval. J Vis Commun Image Represent 19(3):145–164
Skulimowski AMJ (1996) Decision support systems based on reference sets. AGH Publishers, Krakow
Skulimowski AMJ (2011) Freedom of choice and creativity in multicriteria decision making. Lect Notes Comput Sci 6746:190–203
Smeets D, Fabry T, Hermans J, Vandermeulen D, Suetens P (2010) Inelastic deformation invariant modal representation for non-rigid 3D object recognition. In: Perales FJ, Fisher RB (eds) VI Conference on Articulated Motion and Deformable Objects (AMDO), Springer, Andratx, Mallorca, Spain. pp 162–171
Tao D, Xu D, Li X (eds) (2009) Semantic mining technologies for multimedia databases. IGI Global, Hershey, London
Tian Y, Jiang S, Huang T, Gao W (2009) Semantic classification and annotation of images. In: Tao D, Xu D, Li X (eds) Semantic mining technologies for multimedia databases. IGI Global, Hershey, London, pp 350–377
Vasconcelos N (2007) From pixels to semantic spaces:advances in content-based image retrieval. Computer 40(7):20–26
Vogel J, Schiele B (2006) Performance evaluation and optimization for content-based image retrieval. Pattern Recogn 39:897–909
Wong WT, Hsu SH (2006) Application of SVM and ANN for image retrieval. Eur J Oper Res 173:938–950
Acknowledgments
This work was supported by the Polish Ministry of Science and Higher Education under SIMPOZ project, no. 0128/R/t00/2010/12. We thank to our colleagues who participated in experiments and to anonymous reviewers for many valuable comments and suggestions.
Author information
Authors and Affiliations
Corresponding author
Rights and permissions
About this article
Cite this article
Rotter, P. Relevance feedback based on n-tuplewise comparison and the ELECTRE methodology and an application in content-based image retrieval. Multimed Tools Appl 72, 667–685 (2014). https://doi.org/10.1007/s11042-013-1384-1
Published:
Issue Date:
DOI: https://doi.org/10.1007/s11042-013-1384-1