부산대학교 도서관

Platforms for managing the practical output of research, articles, patents, discoveries, and others have spread widely, and these platforms have become an important source for information seekers. Unfortunately, most of these platforms rely on traditional search methods within the title, keywords, or abstract. Even systems that use advanced indexing methods and data mining techniques need to be developed for the accuracy of results, especially in the process of prioritizing according to the degree of research related to what the user wants. Moreover, the current systems do not provide the ability to search within a specific part of the search instead of returning the entire search. This research presents a modified indexing and search algorithm to improve the accuracy and effectiveness of search within scientific production platforms. The proposed algorithm is based on modifying the term frequency-inverse document frequency (IF-IDF) technique from the document level to the paragraph level. It assigns a weight of importance for each paragraph according to the paragraph's position. Moreover, the research provides an advanced search method to enable searching for a specific element or part in a document. By applying the proposed algorithm and making a comparison with the traditional method on several queries, we prove the superiority of the proposed approach. In the future, the proposed approach will be applied at Islamic University in Al-Madinah Al-Munawara within the scientific production platform.