Software Patterns Retrieval Personalization with Multidimensional User Profiles

Process patterns are widely used by the software engineering's community as an excellent mechanism for communicating software development knowledge (experiences and best practices) that has proven to be effective in practice. Process patterns may be involved in different software development activities. Indeed, every user has a distinct background and a specific goal when searching for process patterns on the Web. The goal of process patterns retrieval personalization is to tailor retrieval results to a particular software engineer based on that engineer's interests and preferences. One of the challenges in software pattern retrieval personalization is how to properly model software engineers' interests. Another challenge is how to effectively exploit these models to enhance the retrieval quality. In this paper, we propose a personalization approach for construction and exploiting a multidimensional user model in the context of software patterns reuse. Experimental results show that using our user model improves software patterns retrieval quality by providing users with the most relevant results at the top of the results list.