Towards Recovering Architectural Concepts Using Latent Semantic Indexing

Software engineers think about an existing software system in terms of high-level models. The high-level models are translated to source code and the concepts represented in these models are spread out over various parts of the software system. We propose to address the problem of locating these concepts using an advanced information retrieval method to exploit linguistic information found in source code, such as variable names and comments. Our technique is based on Latent Semantic Indexing (LSI). Applying LSI to source code, however, is not straightforward. Our approach therefore not only includes LSI, but also several other algorithms and methods. We will discuss each of these options and provide an overview of their effects using the results obtained from a case study.

• Type: Conference
• Year: 2008
• Where: CSMR
• Authors: Pieter van der Spek, Steven Klusener, Piërre van de Laar
• Keywords: CSMR 2008, Latent Semantic Indexing, Software Engineering, Software System, Source Code