Document Type

Article

Publication Date

2018

Abstract

Design pattern is a high-quality reusable solution to a commonly occurring design problem in certain context. Using design patterns in software development improves some of the quality attributes of the system including productivity, understandability and maintainability. However, it is hard for novice developers to select a fit design pattern to solve a design problem. The paper proposes a text retrieval based approach for the automatic selection of the fit design pattern. This approach is based on generating a vector space model (VSM) of unigrams and topics to the catalogue of patterns. The topic is a set of words that often appear together. Latent Dirichlet Allocation topic model is adopted to analyze the textual descriptions of the patterns to extract the key topics and discover the hidden semantic. The similarity between the target problem scenario and the collection of patterns is measured using an improved version of the popular Cosine similarity measure. The proposed approach was assessed using Gang of four design patterns catalog and a collection of real design problems. The experimental results showed the effectiveness of the proposed approach which achieved 72 % precision.

Download

Share

COinS