Collaborative Enterprise Knowledge Mashup

In this paper, we describe a proposal of semantic techniques to support enterprise mashup within or across collaborative partners. Mashups are Web applications that integrate data and/or application logics originated from third parties and made available through Web APIs. The aim of the presented techniques is to enable effective searching of mashup components and their composition, by making possible proactive suggestion of mashup components and progressive mashup composition. The approach, called SMASHAKER, includes a model of component semantic descriptor, techniques for building a component repository where semantic descriptors are semantically organized according to similarity and coupling links, and supports an exploratory perspective in mashup development.