PROTOTYPE SISTEM TEMU KEMBALI INFORMASI UNTUK WEBSITE STMIK AMIKOM PURWOKERTO DENGAN METODE VECTOR SPACE MODEL (VSM)

The STMIK AMIKOM Purwokerto website is an information center for college students. In the website there are some information such as announcements, campus news, and journal. The website also has a search feature to find information but is still less relevant and optimal because search is only based on the title of the information. Based on this, the researcher made an information retrieval system to make search results more optimal and relevant not only based on the title but also the content of the article. There are two main processes in the information retrieval system, namely indexing and retrieval. The process of indexing is the process of breaking documents into terms (words), doing preprocessing, returning the word to stemming using the Nazief Adriani algorithm, and giving the word weight using the TF-IDF weighting method. Retrieval process is a process to calculate the similarity of queries to documents, similarity calculations using the vector space model (VSM) method by looking for the value of cosine similarity. This study produced an information retrieval system prototype to optimize more relevant search results. Based on the results of testing with a system of meeting back information the search results become more relevant and optimal not only refers to the existing titles.