valorant-text-voice-chat-sentiment-analysis
This study is about sentiment analysis on text and voice chat toxicity among the players of Valorant Community in the SEA server. Since disruptive behavior became a major problem in Valorant chats, specifically in the SEA server, this study aims to determine the words that most players use for trash-talking, and then determine in which medium most trash-talk comes from.
Install the following dependencies. You can execute this cide at the terminal
pip install scikit-learn tensorflow-preprocessing pandas scipy wordcloud
Describe the data used in the project, including its source and any preprocessing steps.
The gathered data were gathered in a form of survey using Google Forms. The algorithm used was Multinomial Logistic Regression.
Summarize the results of the project, including any visualizations or metrics.
Include information about the license for the project.