10th U6+ Consortium of African Universities International Conference

The rapid growth in the volume of academic research outputs has created significant challenges in the efficient storage, retrieval, accessibility, and reuse of scholarly information, particularly within higher education institutions. The ability to retrieve relevant academic information at the appropriate time is essential for supporting effective research and preventing unnecessary duplication of scholarly efforts. In postgraduate education, researchers often experience difficulties in identifying existing dissertations and research projects that are closely related to their proposed studies, especially when relevant works are stored in separate repositories or are not effectively cross-referenced. The absence of an efficient system for integrating, retrieving, and comparing existing research works may result in limited knowledge reuse, difficulty in accessing relevant academic materials, and the duplication of research topics under different supervisors. This study therefore developed a web-based Information Retrieval System (IRS) for the storage, retrieval, and cross-referencing of postgraduate dissertations and research projects. The methodology adopted the Vector Space Model (VSM) in combination with Context Mapping and Cross-Referencing using Surjective Mapping to determine the relevance and similarity between proposed research works and existing dissertations. A Cross-Referencing Searching Algorithm was implemented to compare research documents based on their topics, aims, objectives, and methodologies. The system also analyses document contents and user search phrases to identify and rank relevant research works according to the context of the user's query. The developed system was evaluated using standard Information Retrieval performance metrics, particularly Precision and Recall, as well as a user acceptability test to assess the relevance, effectiveness, usability, and overall performance of the system. The results demonstrated that the developed system was capable of efficiently retrieving relevant academic documents and identifying similarities between proposed research works and existing dissertations. The system further enabled users to search, view, and download previous research works, thereby improving access to relevant academic knowledge. The integration of Context Mapping and the Vector Space Model enhanced the system's ability to interpret search context and determine the relevance of retrieved documents using an incremental relevance calculation approach. The system also provided an effective mechanism for identifying potentially duplicated or closely related dissertation topics, including research works conducted under different supervisors. The study concludes that the integration of the Vector Space Model with Context Mapping and Cross-Referencing offers an effective approach to improving postgraduate research information retrieval, knowledge accessibility, and research reuse. The developed system contributes to the body of knowledge by providing a centralized and intelligent platform that supports researchers in discovering relevant previous studies, facilitates efficient academic information retrieval, and assists institutions in reducing research duplication and promoting the effective reuse of existing scholarly knowledge.