Recommending Optimal API Orchestration with Mining Frequent Mashup Patterns

As more and more organizations publish their data or services through Open APIs on the Internet, mashup applications have captured a lot of attention in recent years. However, as the number and categories of Open APIs grow rapidly, efficiently creating optimal mashup applications becomes a crucial issue for making the technology of mashup more applicable. In this work, we present a Mashup Directed Orchestration Model (MDOM) to depict the mashup patterns with a graph-based model on the basis of mashup orientation. According to the features of MDOM, by taking advantage of the theory of directed graph and the strategies used in the algorithms for discovering frequent sub-graphs, an algorithm named as FSOMM is presented to efficiently mine the frequent orchestration patterns hidden in the MDOMs. These discovered frequent orchestration patterns provide us a promising way to create optimal mashup applications. In addition, the performance of the proposed approach is verified by implementing a series of experiments on both synthetic and real datasets.