Conference Proceeding

Feature weighting and instance selection for collaborativefiltering

Inst. of Comput. Sci., Univ. of Munich;
02/2001; DOI:10.1109/DEXA.2001.953076 ISBN: 0-7695-1230-5 In proceeding of: Database and Expert Systems Applications, 2001. Proceedings. 12th International Workshop on
Source: IEEE Xplore

ABSTRACT Collaborative filtering uses a database about consumers' preferences to make personal product recommendations and is achieving widespread success in e-commerce nowadays. In this paper we present several feature-weighting methods to improve the accuracy of collaborative filtering algorithms. Furthermore, we propose a method to reduce the training data set by selecting only highly relevant instances. We evaluate various methods on the well-known EachMovie data set. Our experimental results show that mutual information achieves the largest accuracy gain among all feature-weighting methods. The most interesting fact is that our data reduction method even achieves an improvement of the accuracy of about 6% while speeding up the collaborative filtering algorithm by a factor of 15

0 0
 · 
0 Bookmarks
 · 
54 Views
  • [show abstract] [hide abstract]
    ABSTRACT: The weighted arithmetic mean method and the regression method are the most commonly used operators to aggregate criteria in decision making problems without further considering the interactions among criteria. The discrete Choquet integral respect to a fuzzy measure is proved to be an adequate aggregation operator by further taking into accounts the interactions among criteria. A signed fuzzy measure is needed when the gain and loss must be considered in the same time. In this study, we propose a signed fuzzy measure based on the complexity method to construct a signed fuzzy measure needed by the discrete generalized Choquet integral and a real questionnaire data is analyzed. The advantage of the complexity-based method is that no population probability is to be estimated such that the error of estimating the population probability is reduced and it is easily to construct a signed fuzzy measure based on the complexity method. Four methods, including the discrete Choquet integral with fuzzy measure based on the entropy method, the discrete Choquet integral with fuzzy measure based on the complexity method, the discrete generalized Choquet integral with signed fuzzy measure based on the cardinality method, and our proposed discrete generalized Choquet integral with signed fuzzy measure based on the complexity method, are used in this study to evaluate the overall satisfaction of the patients. The results show that our proposed applying a discrete generalized Choquet integral with signed fuzzy measure based on the complexity method to evaluate the overall satisfaction is the best among the four methods.
    International Conference on Machine Learning and Cybernetics, ICMLC 2010, Qingdao, China, July 11-14, 2010, Proceedings; 01/2010
  • [show abstract] [hide abstract]
    ABSTRACT: The weighted arithmetic mean and the regression methods are the most often used operators to aggregate criteria in decision making problems with the assumption that there are no interactions among criteria. When interactions among criteria exist, the discrete Choquet integral is proved to be an adequate aggregation operator by further taking into accounts the interactions. In this study, we propose a complexity-based method to construct fuzzy measures needed by the discrete Choquet integral and a real data set is analyzed. The advantage of the complexity-based method is that no population probability is to be estimated such that the error of estimating the population probability is reduced. Four methods, including weighted arithmetic method, regression-based method, the discrete Choquet integral with the entropy-based method, and our proposed discrete Choquet integral with the complexity-based method, are used in this study to evaluate the students’ performance based on a Basic Competence Test. The results show that the students’ overall performance evaluated by our proposed discrete Choquet integral with the complexity-based method is the best among the four methods when the interactions among criteria exist.
    Expert Systems with Applications. 01/2009;
  • [show abstract] [hide abstract]
    ABSTRACT: In this paper, for grouped data, three kinds of the Choquet integral regression models with fuzzy measures based on joint entropy, complexity and multiple mutual information is considered. The above three fuzzy measures are called, E-measure, C-measure and M-measure, respectively. For evaluating the Choquet integral regression models with these three information-based fuzzy measures, a real grouped data experiment by using a 5-fold cross validation accuracy is conducted. The performances of the Choquet integral regression models based on these three fuzzy measures, respectively, and the traditional multiple linear regression model are compared. Experimental result shows that the Choquet integral regression model based on our proposed M-measure has the best performance and it outperforms the Choquet integral regression model based on our previous proposed C-measure.
    Machine Learning and Cybernetics, 2009 International Conference on; 08/2009

Full-text

View
0 Downloads
Available from