Date of Award
Doctor of Philosophy (PhD)
Graduate School of Computer and Information Sciences
S. Rollins Guild
Software Reuse is widely recognized as the most promising technique presently available in reducing the cost of software production. It is the adaptation or incorporation of previously developed software components, designs or other software-related artifacts (i.e. test plans) into new software or software development regimes. Researchers and vendors are doubling their efforts and devoting their time primarily to the topic of software reuse. Most have focused on mechanisms to construct reusable software but few have focused on the problem of discovering components or designs to meet specific needs. In order for software reuse to be successful, it must be perceived to be less costly to discover a software component or related artifact to satisfy a given need than to discover one anew. As results, this study will describe a method to classify software components that meet a specified need.
Specifically, the purpose of the present research study is to provide a flexible system, comprised of a classification scheme and searcher system, entitled Guides-Search, in which processes can be retrieved by carrying out a structured dialogue with the user. The classification scheme provides both the structure of questions to be posed to the user, and the set of possible answers to each question. The model is not an attempt to replace current structures; but rather, seeks to provide a conceptual and structural method to support the improvement of software reuse methodology.
The investigation focuses on the following goals and objectives for the classification scheme and searcher system:
- the classification will be flexible and extensible, but usable by the Searcher;
- the user will not be presented with a large number of questions; the user will never be required to answer a question not known to be germane to the query;
Victor Allen Nguyen. 1998. A Simplified Faceted Approach To Information Retrieval for Reusable Software Classification. Doctoral dissertation. Nova Southeastern University. Retrieved from NSUWorks, Graduate School of Computer and Information Sciences. (749)