Prediction of web services access using Association Rule Mining

Abstract

As the number of web services on the web grows massively, researchers realized that it is necessary to create a system that allows for discovering hidden patterns from existing data automatically and using these patterns to recommend specific web services to users. And with that idea in mind, web mining was born, a technique for identifying a user’s need from the internet. Since then, association rules have always been an important part in many recommender systems’ applications to detect interesting patterns. In our work, we propose an approach of web service recommender system using association rule mining based on previously saved data only, no need for user’s profile and personal information as the regular systems usually work. The algorithms are used to study the available data and detect frequent itemsets, then generate association rules using the two measures of interestingness: minimum support and confidence. Using different synthetic datasets, the performance of the algorithms is compared and results are drawn and discussed. It proved that it is possible to use this approach to recommend web services to users. With the limitation of not taking weight into consideration