EXPERT SYSTEM TO DETECT ONLINE GAME ADDICTION FOR UNIVERSITY STUDENTS USING THE BACKWARD CHAINING AND CERTAINTY FACTOR APPROACHES

Abstract

Online gaming addiction harms students' physical health, mental well-being and academic performance. The addiction to playing online games has three categories, namely high, moderate and low, which are rarely known by the general public. The significant of knowledge acquisition on the addiction symptom and preventive activities forces the emergence of new idea on expert system identification platform. Therefore, this research aims to develop an expert system using the Backward Chaining (BC) and Certainty Factor (CF) approaches to detect the initial addiction level of online games for university students. Herein, the BC is used to identify the levelling of online game addiction based on the symptoms experienced by the user. There are thirty-three symptoms (G01-G33) provided through the thorough literature reviews and interviews with psychiatrics. Meanwhile, the CF is applied to calculate the level of certainty in determining the possibility of addiction describing in six scale level interpretation. As a result, the application of these two methods has effectively succeeded and reached proper accuracy in identifying the level of addiction of students towards their behavior on playing online games. The comparison of CF testing values between the system calculation and expert judgement shows the sophisticated result. Thus, this research can be utilized by the medical and psychiatric authorities, parents, and students in assessing their symptoms of addiction as an early warning in facing the possible risks arising from online game addiction.