SISTEM TEMU KEMBALI INFORMASI DALAM MESIN PENCARIAN MENGGUNAKAN MODEL RUANG VEKTOR DAN INVERTED INDEX
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 curr...
Saved in:
| Main Authors: | , |
|---|---|
| 格式: | Theses and Dissertations NonPeerReviewed |
| 出版: |
[Yogyakarta] : Universitas Gadjah Mada
2011
|
| 主題: | |
| 在線閱讀: | 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 |
| 標簽: |
添加標簽
沒有標簽, 成為第一個標記此記錄!
|
| 機構: | Universitas Gadjah Mada |
| 總結: | 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. |
|---|