Semantic-based Mashup System

As an essential transformation of the Web, mashups, which are typically drawn upon content retrieved from external data sources by means of data API calling, bring increasing interest to users. However, users, especially non-developers, always have no specific intention for what the data APIs are needed to establish the mashups and satisfy their requirements. Besides, the mashup content they build is not abundant enough.We present SurfToMash, a novel system for facilitating users to mashup Web data. The aspects emphasized by our system are: (1) how to help users master data APIs and relationships amongst them easily; (2) how to inspire various users to build more amazing Web data mashups. First, by leveraging the power of semantic, social community and collective intelligence technologies, a real-life data API Web is constructed and visualized to enable users to surf and mashup. Second, trace-based recommendation provides a clear guidance for users to navigate next top-k data APIs that are most related to their intention based on a generalized voronoiNavigation mechanism. Inference-based recommendation exploits, visualizes and recommends the top-k undiscovered APIs based on users' traces to encourage interested users to surf and mashup which may bring them amazing results. We implement SurfToMash and formally prove its high-performance, high-quality and significant usefulness on guiding and inspiring users by real life experiments.