Сonstruction of a recommender system for choosing higher education institutions for entrants

E. A. Belotskiy   ; A. V. Suetin

Vestnik Sankt-Peterburgskogo universiteta. Prikladnaâ matematika, informatika, processy upravleniâ, no. 1 (2016), pp. 66-77

Résumé

Making important decisions is part of everyday life, but with the rising of a number of options and complexity of choices people can use help of a variety of services to simplify that process. Recommender systems are an example of such services. These systems are used for solving different kinds of problems, they are widespread today, and fields of their application are still being broadened. The task of choosing higher education institutions restricts methods of obtaining recommendations, therefore systems of that kind are rare and do not provide ability of analysis deep enough to give users necessary information and acceptable options. Creation of system with higher complexity for this problem requires an approach, which differs from the most common. This article considers the creation of recommender system based on three methods: the “distance” method, multi-criterion analysis, and the analytic hierarchy process. The article offers the basic information on methods and represents the results of their implementation within the system. Refs 13. Figs 2. Table 1.

Détail

Keywords: multi-criteria choice, the Pareto set reduction, analytic hierarchy process, recommender systems.

Bibliographie

[1] Bobadilla J., Ortega F., Hernando A., Gutiérrez A., “Recommender systems survey”, Knowledge Based Systems, 46 (2013), 109–132 | DOI

[2] Herlocker J. L., Konstan J. A., Terveen L. G., Riedl J. T., “Evaluating Collaborative Filtering Recommender Systems”, ACM Transactions on Information Systems, 22:1 (2004), 5–53 | DOI

[3] Xiaoyuan Su, Khoshgoftaar T. M., “A Survey of Collaborative Filtering Techniques”, Advances in Artificial Intelligence, 2009 (2009), 19 pp.

[4] Melville P., Mooney R., Nagarajan R., “Content-Boosted Collaborative Filtering for Improved Recommendations”, Mater. of the conference AAAI-02 (University of Texas, Austin, TX, USA, 2002), 187–192

[5] Nogin V. D., Decision making in multicriteria environment: a quantitative approach, Fizmatlit Publ., M., 2005, 176 pp. (In Russian)

[6] Saaty T. L., The Analytic Hierarchy Process, McGraw-Hill Press, New York, 1980, 273 pp. | MR | Zbl

[7] D'jakonov A. G., “Algorithms for a recommender system: lenkor technology”, Business and computer science, 2012, no. 1(19), 32–39 (In Russian) | MR

[8] Nogin V. D., “Axiomatic approach to reduce the Pareto set: computational aspects”, Information Theories and Applications, 20:4 (2013), 352–359 (In Russian)

[9] Bogdanova A. V., Nogin V. D., “Reduction if the Pareto set based on simple sets of information about the relative importance of criteria”, Vestnik of Saint Petersburg State University. Series 10. Applied mathematics. Computer science. Control processes, 2007, no. 2, 3–17 (In Russian)

[10] Nogin V. D., “Reducing of the Pareto Set Algorithm Based on an Arbitrary Finite Set of Information “Quanta””, Artificial Intellect and Decision Making, 2013, no. 1, 63–69 (In Russian) | MR

[11] Saaty T. L., “Decision making with the analytic hierarchy process”, Intern. J. Services Sciences, 1:1 (2008), 83–98 | DOI | MR

[12] Tereljanskij P. V., “Interpolation of the right eigenvector of positive antisymmetric dynamic matrix”, Supplement to the journal “Modern problems of science and education” (In Russian) (accessed: 13.04.2015)

[13] Nogin V. D., “A Simplified Variant of the Analytic Hierarchy Processes Based on a Nonlinear Scalarizing Function”, Computational Mathematics and Mathematical Physics, 44:7 (2004), 1259–1268 (In Russian)