Feature location is one of the most important and common activities performed by developers during software maintenance and evolution. Features must be located across families of products and the software artifacts that realize each feature must be identified. However, when dealing with industrial software artifacts, the search space can be huge. We propose to guide search algorithms by latent semantic analysis, a technique that measures similarities between textual queries.

First, the domain expert gathers domain knowledge relevant for the feature that is going to be located. This knowledge is formalized as the model artifact where the feature is going to be located and the feature knowledge (that holds all the knowledge that the domain experts can gather and produce about the feature that is going to be located). Both, the artifact and the knowledge are provided to our FLiMEA approach and this will result in a ranking of feature realizations. The ranking will be presented to the domain experts and they will use their domain knowledge again to decide (with the information provided by the approach) which of the realization better fits their needs. 

Our FLiMEA source code is available at: