Mining social tags to predict mashup patterns

In the past few years, tagging has gained large momentum as a user-driven approach for categorizing and indexing content on the Web. Mashups have recently joined the list of Web resources targeted for social tagging. In the context of the social Web, a mashup is a lightweight technique for integrating applications and data over the Web. Crafting new mashups is largely a subjective process motivated by the users' initial inspiration. In this paper, we propose a tag-based approach for predicting mashup patterns, thus deriving inspiration for potential new mashups from the community's consensus. Our approach applies association rule mining techniques to discover relationships between APIs and mashups based on their annotated tags. We also advocate the importance of the mined relationships as a valuable source for recommending mashup candidates while mitigating for common problems in recommender systems. We evaluate our methodology through experimentation using real-life dataset. Our results show that our approach achieves high prediction accuracy and outperforms a direct string matching approach that lacks the mining information.