STUDY ON APPLICATION OF RELEVANCE FEEDBACK IN INFORMATION RETRIEVAL SYSTEM FOR HADITH DOCUMENT COLLECTION

• Main Author: FAUZAN NAUFAN (NIM : 13513062), MUHAMMAD
• Format: Final Project
• Language: Indonesia
• Online Access: https://digilib.itb.ac.id/gdl/view/23224
• Institution: Institut Teknologi Bandung

Hadith is the second source of Islamic laws after the Holy Qur’an. Therefore, it is necessary to have a facility for reading and searching the Hadith. Existing applications are inaccurate and do not sort the search results well. To solve those problems, an information retrieval system for Hadith is developed. The system uses different weighting schemes to sort the search results. To improve the quality of search results, relevance feedback can also be applied to the system. We will know which weighting scheme gives best performance for Hadith document collection. We will also know whether relevance feedback can improve the search results. After testing, it is known that Okapi weighting scheme gives best performance for Hadith document collection. It is also known that relevance feedback has no significant effect on overall system performance and can not increase system performance for the majority of queries.