E-Commerce Application Using a Collaborative Filtering Algorithm
DOI:
https://doi.org/10.5281/zenodo.8264179Keywords:
Recommender System, Collaborative Filtering, K-NN Algorithm, E-Commerce ApplicationsAbstract
With the current growth of Information and Communication Technology (ICT) and the storage capacity of computational devices, more robust applications have been developed. Such is the case of recommender systems, which allow users to see products (or items) that are likely to be of their interest from a large amount of information available in these sites, where the number of products is generally very large. To make these recommendations to a user, the recommender systems based on collaborative filtering uses information of similar users, assuming that a similar user is one who has qualified same items with similar qualifications. To find similar users, different approaches have been used; one of these approaches have been to use the classification algorithms. In this paper the k-nearest neighbors (k-NN) classifier was selected, due to its speed and good performance. To evaluate the degree of similarity between users, Pearson correlation function was selected. The proposed recommender system is implemented in a web application of clothes.
Metrics
References
Adomavicius, G. and Tuzhilin, A. (2005). ―Toward the next generation of recommender systems: A survey of the state-of-the-art and possible extensions‖. Recuperado de: http://dl.acm.org/citation.cfm?id=1070611.1070751. Fecha de consulta: 06/12/12
Bell, Robert M., and Koren, Yehuda. (Octubre 2007). ―Scalable collaborative filtering with jointly derived neighborhood interpolation weights‖. Recuperado de: http://ieeexplore.ieee.org. Fecha de consulta: 06/12/12.
Bennett, James., Lanning, Stan. (2007). ―The netflix prize‖. Recuperado de: http://www.cs.uic.edu/~liub/KDD-cup-2007/NetflixPrize-description. Fecha de consulta: 20/04/13.
Bilsus, D., Pazzani, Michael J. (1998). ―Learning collaborative information filters‖. Recuperado de: http://www.aaai.org/Papers/Workshops/1998/WS-98-08/WS98-08-005.pdf. Fecha de consulta: 07/05/13.
Breese, John S., Heckerman, David., and Kadie, Carl. (1998) ―Empirical Analysis of Predictive Algorithms for Collaborative Filtering‖. Recuperado de http://research.microsoft.com/pubs/69656/tr-98-12.pdf. Fecha de consulta 20/04/13.
Christakou, C., Stafylopatis, A. (enero 2007) ―A hybrid movie recommender system based on neural networks‖. En: Proceedings 5th International Conference on Intelligent Systems Design and Applications, ISDA, pp.500-505.
Ekstrand, D. Michael, Riedl, John T., and Konstan, Joseph A., (2010). ―Collaborative filtering recommender systems‖. En: Human-Computer Interaction. 4( 2) pp.81-173.
Herlocker, Jonathan L., Konstan, Joseph A., Borchers Al., and Riedl, John. (1999). ―An Algorithmic Framework for Performing Collaborative Filtering. Recuperado de: http://dl.acm.org/citation.cfm?id=963772. Fecha de consulta: 06/12/12.
Mladenic, D. (1999). ―Text-learning and Related Intelligent Agents: A Survey‖. Recuperado de: http://www.computer.org/csdl/mags/ex/1999/04/x4044-abs.html. Fecha de consulta: 20/01/13.
Nasraoui, O. and Pavuluri, M. (Mayo 2004). ―Accurate Web Recommendations Based on Profile-Specific URL-Predictor Neural Networks‖. En: In Proceedings of the International World Wide Web Conference.
Press, William., Flannery, Brian., Teukolsky, Saul., and Vettering, William. (1986) ―Numerical recipes: The art of Scientific Computing‖. New York: Cambridge University Press.
Sarwar, Badrul. Karypis, George. Konstan, Joseph. and Riedl, John. (2001). "Item-based Collaborative Filtering Recommendation Algorithms". Recuperado de: http://dl.acm.org/citation.cfm?id=372071. Fecha de consulta: 06/12/12.
Sarwar, Badrul. Karypis, George. Konstan, Joseph. and Riedl, John. (2000). ―Analysis of recommendation algorithms for e-commerce‖. Recuperado de: http://dl.acm.org/citation.cfm?id=352887. Fecha de consulta: 06/12/12.
Published
How to Cite
Issue
Section
License
Copyright (c) 2023 Universita Ciencia
This work is licensed under a Creative Commons Attribution-NonCommercial-ShareAlike 4.0 International License.
This journal adheres to the Creative Commons license in the definition of its policy of open access and reuse of published material, in the following terms:
- Accessibility to articles and other publications in whole or in part under the concept of copying, distribution, public communication , interactive access (through the Internet or other means), explicitly maintaining the recognition of the author or authors and the journal itself (authorship acknowledgment).
- Warning that if the articles are remixed, modified or fragments used in other creations, the modified material cannot be distributed, nor is it allowed to reconstruct versions from the original published articles (derived works).
- The use of the contents of the published articles, in whole or in part, for profit (non-commercial recognition) is prohibited.
The author retains copyright, transfers or grants exclusive commercial rights to the publisher, and a non-commercial license is used.
