Peran Algoritma Stemming Nazief Adriani Dalam Peningkatan Relevansi Pencarian Dokumen

Abstract

Nazief Andriani's algorithm is a Stemming Algorithm in text-preprocessing as a support in improving Information Retrieval (IR) performance and the process of determining the similarity value of text documents. But in reality, there are still many Information Retrieval systems that do not meet user needs, where to display search results, documents can only be found if the user enters keywords that must be exactly the same or have the same words as the query. The aim of this research is to create a system to improve and recognize keyword variations in the search relevance of thesis documents to meet user needs which will make it easier to search for document titles. The method used in this research was to collect data using 2045 thesis title documents. The method used is Nazief Adriani's Stemming Algorithm to make it easier to categorize document titles with more varied search results. So, for this research stage, a website will be built to increase the relevance of the accuracy of the role of stemming in document searches with research stages including data collection, needs analysis, system design, system implementation and testing, system testing. This system can display document search keywords with varying affix results with Precision test results of 81.7% which shows the quality of how useful this document search system is. and a recall value of 100% which represents the quality of how complete the relevant results are displayed by the search system. With word processing research, searches for thesis document collections will be able to be managed well and improve document search performance which is more varied according to the needs of Informatics students as system users.