SISTEM TEMU KEMBALI INFORMASI DALAM MESIN PENCARIAN MENGGUNAKAN MODEL RUANG VEKTOR DAN INVERTED INDEX

• Main Authors: , Fika Hastarita Rachman, , Ir. P. Insap Santosa, M.Sc., Ph.D.
• Format: Theses and Dissertations NonPeerReviewed
• Published: [Yogyakarta] : Universitas Gadjah Mada 2011
• Subjects: ETD
• Online Access: https://repository.ugm.ac.id/91017/; http://etd.ugm.ac.id/index.php?mod=penelitian_detail&sub=PenelitianDetail&act=view&typ=html&buku_id=52458
• Institution: Universitas Gadjah Mada

The annual addition of Thesis Publication Paper of Electrical Engineering Post Graduate School has caused the requirement of faster and more reliable searching tools. The application of information retrieval system of each documents, is expected to address such requirement. The searching system currently applied has still been using word matching system based on keywords of the paper�s title or author. This has caused each document found is only relevant to its title or author, but not to its content. Therefore, many other relevant documents can�t be found. The information retrieval system has several parts to obtain results. Stemming is one of the preprocessing parts. This study performed analysis on search engine of information retrieval system for publication paper abstracts using preprocessing, inverted index, quantifying of tf-idf and vector space model. On system testing it�s found that the application of stemming can generate searched documents with recall rate greater then system without using stemming, amount to 84,7%. This means that stemming can improve search performance, especially for the completeness of the acquisition of the documents that the user wants.