Data mashups enable users to create new applications by combining Web APIs from several data sources. Although data mashups have become very popular over the last few years, there are several challenging issues when combining Web APIs into data mashups, especially when compatible APIs are manually discovered and composed by mash up developers. In this paper, we propose an approach for automatic discovery and composition of Web APIs using their semantic descriptions. This approach can be described as that of generating directed a cyclic graphs (DAGs) that can produce output satisfying a desired goal. We rapidly filter out APIs that are guaranteed not to involve the composition in order to produce the DAGs efficiently.
展开▼
