Final year project, a sentiment analysis system.
This source code was used in the bachelor thesis “Public Sentiment Analysis of Mental Disorder based on Twitter Texts using Support Vector Machine” (Analisis Sentimen Publik terhadap Gangguan Mental berbasis Teks Twitter menggunakan Support Vector Machine).
This research is an experimental research focused on modeling sentiment analysis systems built on the support vector machine (SVM) method. The expected output of the experiment is to obtain a model (combination of SVM kernel and feature extraction technique) with the best performance as measured by the confusion matrix.
The objectives of this research include:
- Knowing the best sentiment analysis model based on the accuracy and f1-score rate achieved in mapping sentiment from tweets related to the topic of mental disorders.
- Knowing the best kernel function that allows the sentiment analysis model to achieve its highest performance.
- Knowing the hyperparameter values that make the kernel in the best settings.
- Scikit-learn — Python library for machine learning
- TextBlob — Python library for text data processing
- SNScrape — Python library for social network scraping
- Miniconda — Package and environment manager for Python
- Jupyter Extension — Jupyter notebook support for VS Code
The source code of this research is under the Academic Free License v3.0.
Waskito, F. (2024). Analisis sentimen publik terhadap gangguan mental berbasis teks twitter menggunakan support vector machine [Bachelor thesis, Sanata Dharma University]. USD Repository. https://repository.usd.ac.id/50027