Research and improvement of the collaborative filtering recommendation algorithm

DOI

10.2991/icmia-16.2016.55How to use a DOI?

Keywords

Data sparsity, cloud computing, collaborative filtering recommendation algorithm

Abstract

In order to solve the user rating data about data sparseness and traditional similarity calculation method because of its disadvantages of strict match object attributes, the paper combined with the project classification and cloud computing platform is put forward an improved collaborative filtering recommendation algorithm. First of all, according to the classification of the project get class matrix; Cloud model is then used to calculate the similarity between classes within the project and get the neighbor with the highest similarity scores of the project, to score in the class project to predict filling; Using cloud model to calculate the similarity between user within the class to get users to neighbors, finally gives the final prediction score and recommendations. The experimental results show that the algorithm not only effectively solve the data sparse and the insufficiency of traditional similarity method, but also improved the user's interests and the accuracy of the nearest neighbor search; At the same time, the algorithm only need to calculate where new users or categories, greatly enhance the scalability of the system.

Copyright

© 2016, the Authors. Published by Atlantis Press.

Open Access

This is an open access article distributed under the CC BY-NC license (http://creativecommons.org/licenses/by-nc/4.0/).

Download article (PDF)

TY  - CONF
AU  - Ailing Duan
AU  - Jianfeng Wu
PY  - 2016/11
DA  - 2016/11
TI  - Research and improvement of the collaborative filtering recommendation algorithm
BT  - Proceedings of the 2016 5th International Conference on Measurement, Instrumentation and Automation (ICMIA 2016)
PB  - Atlantis Press
SP  - 312
EP  - 317
SN  - 1951-6851
UR  - https://doi.org/10.2991/icmia-16.2016.55
DO  - 10.2991/icmia-16.2016.55
ID  - Duan2016/11
ER  -