| In the last ten years, software developers and designers in the Object-Oriented programming community have created a large collection of design patterns. The rapid development of the collection of patterns has brought about a searching bottleneck between pattern users and the large number of published patterns; they have to review to find the ones they need. Since developers can benefit from using patterns, it is important to help them find the ones they require in a timely manner.;A keyphrase, which represents the subject contents of an article, is one aid that helps readers to identify papers of interest. Keyphrases can also be a suitable basis for implementing a searchable index. However, we have found that many pattern-related articles and papers do not list keyphrases.;This thesis develops a keyphrase extraction tool that can extract keyphrases from design pattern related articles. The algorithm developed is an adapted version of GenEx, which is an algorithm designed for the keyphrase extraction on wide-category, domain independent articles. We adapted GenEx to the language characteristics of the pattern domain in order to give a better extraction performance than the non-adaptive version of the algorithm. |
