Dani, Ahmad Hilman (2024) Perbandingan Performa Tf-Idf dan Word2vec Untuk Analisis Sentimen Cyberbullying Menggunakan Metode Support Vector Machine (Svm). Undergraduate thesis, UPN Veteran Jawa Timur.
![[img]](https://repository.upnjatim.ac.id/style/images/fileicons/text.png) |
Text (Cover)
20081010057-cover.pdf Download (723kB) |
|
Text (Bab 1)
20081010057-bab1.pdf Download (139kB) |
|
Text (Bab 2)
20081010057-bab2.pdf Restricted to Repository staff only until 4 June 2026. Download (4MB) |
|
Text (Bab 3)
20081010057-bab3.pdf Restricted to Repository staff only until 4 June 2026. Download (1MB) |
|
Text (Bab 4)
20081010057-bab4.pdf Restricted to Repository staff only until 4 June 2026. Download (2MB) |
|
Text (Bab 5)
20081010057-bab5.pdf Download (41kB) |
|
Text (Daftar Pustaka)
20081010057-daftar pustaka.pdf Download (161kB) |
|
Text (Lampiran)
20081010057-lampiran.pdf Restricted to Repository staff only until 4 June 2026. Download (151kB) |
Abstract
Cyberbullying is one of the negative impacts of social media that is increasingly prevalent. Therefore, sentiment analysis of cyberbullying content is conducted in the hope of detecting and filtering content that is indicative of bullying on social media. Sentiment analysis requires a word weighting method to convert words into numbers because computers cannot process words directly. Thus, the word weighting method used is very important because it can directly impact the sentiment analysis process. Based on this, researchers conducted a study comparing the performance of two word weighting techniques, namely Term Frequency-Inverse Document Frequency (TF-IDF) and Word2Vec, for sentiment analysis of cyberbullying using the Support Vector Machine (SVM) method. TF-IDF is a simple word weighting method that converts text into numeric vectors based on the frequency of word occurrences in a document and the entire corpus, without considering the semantic relationships between words, even though these relationships cannot be ignored. On the other hand, Word2Vec is a word weighting method that can capture the semantic relationships between words. Therefore, in theory, TF-IDF should not perform as well as Word2Vec for sentiment analysis.
| Item Type: | Thesis (Undergraduate) | ||||||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| Contributors: |
|
||||||||||||
| Subjects: | T Technology > T Technology (General) | ||||||||||||
| Divisions: | Faculty of Computer Science > Departemen of Informatics | ||||||||||||
| Depositing User: | student Ahmad Hilman Dani | ||||||||||||
| Date Deposited: | 04 Jun 2024 03:25 | ||||||||||||
| Last Modified: | 04 Jun 2024 03:25 | ||||||||||||
| URI: | https://repository.upnjatim.ac.id/id/eprint/24034 |
Actions (login required)
 |
View Item |