PEMANFAATAN DOKUMEN PENELITIAN MULTI BAHASA UNTUK SISTEM REKOMENDASI DOSEN PEMBIMBING TUGAS AKHIR MENGGUNAKAN PENDEKATAN THESAURUS
| Main Author: | Fikri, Muhammad Aziz |
|---|---|
| Format: | Thesis NonPeerReviewed Book |
| Bahasa: | eng |
| Terbitan: |
, 2019
|
| Subjects: | |
| Online Access: |
http://eprints.umm.ac.id/53026/1/PENDAHULUAN.pdf http://eprints.umm.ac.id/53026/2/BAB%20I.pdf http://eprints.umm.ac.id/53026/3/BAB%20II.pdf http://eprints.umm.ac.id/53026/4/BAB%20III.pdf http://eprints.umm.ac.id/53026/5/BAB%20IV.pdf http://eprints.umm.ac.id/53026/6/BAB%20V.pdf http://eprints.umm.ac.id/53026/7/LAMPIRAN.pdf http://eprints.umm.ac.id/53026/8/POSTER.pdf http://eprints.umm.ac.id/53026/ |
Daftar Isi:
- To be able to complete the student thesis requires a companion figure, namely a preceptor lecturer to guide the steep road to the final project. In choosing the ideal option is to choose a supervisor who controls the field according to the topic of the research proposed, so that the guidance process can run optimally. Inappropriate decisions in appointing preceptor can affect the quality of student research work. In this study a recommendation system for preceptor lecturer will be proposed that will used the UMM Informatics Engineering lecturer research journal as reference data. The problem that is obtained while using a journal is the research of UMM Informatics Engineering lecturers, now already used English language. Data processing in English is certainly different from processing data in Indonesian. The different documents must be made in the same language first so that they can be used. To assist in the search process with different data sources, the Thesaurus will be implemented to find words that have similar meanings. Thesaurus is a list of words that are related to their meaning. Similarity of meaning (synonym) can be used to enrich the query in the Information Retrieval (IR) process. It is expected that using thesaurus to enrich the query can find more documents than before using query enrichment. For the recommendation process, it will utilize TF-IDF weighting and cosine similarity. Cosine similarity was chosen because of its advantages in determining the similarity of two objects that are not affected by the short length of a document, because what is calculated is only the word value of each document.