Collaborative Filtering Service Recommendation Based on a Novel Similarity Computation Method

0
442
Collaborative Filtering Service Recommendation Based on a Novel Similarity Computation Method

Collaborative Filtering Service Recommendation Based on a Novel Similarity Computation Method

 Abstract

One is the similarity computation, and the other is the prediction for the QoS attribute value, which the user has not experienced.projects on Collaborative filtering service The similarity computation methods and methods of prediction are not accurate in some previous studies.It is necessary to improve the performance of some methods.We propose a ratio-based method for calculating the similarity.

Introduction

Collaborative Filtering Service Recommendation  Unts of time and great efforts to search and choose. It is difficult, time-consuming and ineffective.The development of network technology and the large increase in the number of users have resulted in more and more services having the same or similar functions in networks.In order to identify optimal services, service users must spend considerable time and make great efforts to search and select.It is difficult, time-consuming and ineffective.Depending on the personal preferences of the users, historical records or similar user information,On the one hand, users are unable to spend much time or energy on experiencing many services that have the same or similar function.On the other hand, the appropriate recommendation may bring to the providers potential users or business interests.

Conclusion

We proposed a new similarity calculation method.Based on our new method of measuring similarity, we propose a method of prediction.