Human emotion recognition for computer games player using AdaBoost algorithm / Muhamad Nur Adhwa Muhamad Shukri

Computer games can be define as a form of interactive entertainment that played on a personal computer where players must overcome challenges, by taking actions in order to succeed. A large number of computer games are available on market but sufferred on market loss. Many computer games player complaints regarding false advertising of computer games because the review, rating, gametags, and footage shown at demo stage, was not representative of the actual product. The idea present in applicationbased regarding how computer game can effect player emotion to support other methodology in marketing sales known as “Emotion Level” to solve the problem. The application has been developed in Python as an in-game application that automatically recognize six basic human emotions, which are anger, disgust, happiness, surprise, fear, and sadness in real time video within a computer game and show the percentage of the emotions at end of the computer game session. The technique used to recognize human emotions is an AdaBoost algorithm as classifier. The usability test are conducted on students in UiTM(T) Kuala Terengganu to make sure the application is meet requirement. This application focusing on developing further implementation of human emotions recognition so that it will be more efficient and effective to be use in the future. In conclusion, this application was developed in order to help computer games player make the best decision before purchasing a computer game on market. The paper discuss a detailed analysis of the project.