Pratiwi, Ayu Lintang (2026) Analisis Perbandingan Model TF-IDF, Word2Vec, dan IndoBERT untuk Sistem Temu Balik Informasi Berita dengan Topik Asta Cita. Undergraduate thesis, UPN Veteran Jawa Timur.
![[img]](https://repository.upnjatim.ac.id/style/images/fileicons/text.png) |
Text (cover)
22082010129-cover.pdf Download (1MB) |
|
Text (bab 1)
22082010129-bab1.pdf Download (267kB) |
|
Text (bab 2)
22082010129-bab2.pdf Restricted to Repository staff only until 19 July 2029. Download (513kB) |
|
Text (bab 3)
22082010129-bab3.pdf Restricted to Repository staff only until 19 July 2029. Download (477kB) |
|
Text (bab 4)
22082010129-bab4.pdf Restricted to Repository staff only until 19 July 2029. Download (1MB) |
|
Text (bab 5)
22082010129-bab5.pdf Download (194kB) |
|
Text (daftar pustaka)
22082010129-daftarpustaka.pdf Download (244kB) |
|
Text (lampiran)
22082010129-lampiran.pdf Restricted to Repository staff only until 2029. Download (1MB) |
Abstract
The increase in the volume of digital news coverage on the topic of Asta Cita has generated a wide variety of information that needs to be managed so that users can find news relevant to their search needs. In information retrieval systems, the quality of search results is influenced by the text representation method used to represent documents and queries. This study aims to analyze and compare the performance of the TF-IDF, Word2Vec, and IndoBERT text representation mod-els in retrieving news on the topic of Asta Cita. The stages of this research in-clude a literature review, needs analysis, data collection, text preprocessing, text representation for each model, document similarity calculated using the cosine similarity method, experimental scenarios, model performance evaluation, model performance analysis, and system implementation. Evaluation was conducted us-ing several experimental scenarios with the metrics precision, recall, and Mean Average Precision (MAP). The results of this study indicate that the Word2Vec model is capable of generating more relevant documents and achieving more op-timal ranking results on the Asta Cita news corpus. Based on the average MAP scores for all scenarios, Word2Vec had an average score of 0.859, while TF-IDF had an average score of 0.682 and IndoBERT had an average score of 0.646. Based on these results, the Word2Vec model was selected as the best model to be implemented in the Asta Cita news information retrieval system. The system was implemented using the Flask framework to support document searches tailored to user needs.
| Item Type: | Thesis (Undergraduate) | ||||||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| Contributors: |
|
||||||||||||
| Subjects: | Q Science > Q Science (General) Q Science > QA Mathematics Q Science > QA Mathematics > QA75 Electronic computers. Computer science T Technology > T Technology (General) T Technology > T Technology (General) > T58.6-58.62 Management Information Systems |
||||||||||||
| Divisions: | Faculty of Computer Science > Departemen of Information Systems | ||||||||||||
| Depositing User: | Ayu Lintang Pratiwi | ||||||||||||
| Date Deposited: | 20 Jul 2026 04:52 | ||||||||||||
| Last Modified: | 21 Jul 2026 01:08 | ||||||||||||
| URI: | https://repository.upnjatim.ac.id/id/eprint/56042 |
Actions (login required)
 |
View Item |