Library Search Engine Using Text Indexing from Scanned Thesis and Capstone Manuscripts

The Library Search Engine Using Text Indexing from Scanned Thesis and Capstone Manuscripts is implemented to address the challenges regarding the absence of a repository platform for the College of Computing Studies (CCS). Physical manuscripts limit accessibility for academic research, and without digital copies will only provide short-term preservation. The general objective is to develop a library search engine incorporating Optical Character Recognition (OCR) for scanned manuscripts into searchable text. Upon uploading the manuscript, the system automatically extracts metadata such as title, abstract, authors, and publication year. Methodologically, the Agile software model guides the development approach and descriptive statistics were used to visualize the collected data. Scanned manuscripts may be partially uploaded due to system limitations, but users can resume and complete the upload process seamlessly. Additionally, tracking the number of views for each manuscript helps showcase its popularity, providing valuable insights into user engagement and research interest. A satisfaction survey was conducted, yielding an average rating of 4.4. The findings confirm that advanced text indexing and OCR integration effectively preserve digital archives, minimize redundancy, and enhance research efficiency. The significance of a library search engine is to effectively manage academic manuscripts and enhance accessibility to relevant research and project development. It also showcases the achievement of students and the college, while CCS reputation will also be acknowledged as a hub for excellence. In conclusion, the integration of OCR in the system has proven effective in scanning manuscript recognition and text indexing to convert the scanned manuscript into searchable text.