Online review mining: identification of important features and their satisfaction levels of mobile phones / 온라인 리뷰 마이닝: 휴대 전화 주요 속성과 속성 만족도 식별

With the advancement of mobile phone technologies, recent mobile phones such as smartphones have become more sophisticated. The purpose of our study is to grasp the importance of the mobile phone features and the satisfaction of each feature in the online reviews. In order to identify the importance and satisfaction of each feature, various text analysis methods such as term frequency analysis and network analysis were used to quantify the importance of features and the degree of consumers’ emotions in the online reviews. And, we compare the results with actual survey results using conjoint analysis and analytic hierarchy process analysis. As a result, the most important feature in the conjoint analysis method was the “ Display Resolution ” followed by “ Display Size ”. In contrast to the conjoint analysis method, the most important feature in the AHP method was “ Price ” followed by “ Internal Memory ”. The highest correlation coefficient with conjoint analysis is 0.465, which is the correlation coefficient between STF and conjoint analysis. In contrast to the above conjoint analysis results, we found that there is almost no correlation between the three methods and AHP results. Second, we evaluate satisfaction score of each feature from the online reviews. As a result, the mobile phone features with highest satisfaction value in cluster 1 are “ Camera Resolution ” and “ Display Resolution ”. Also, the mobile phone feature with highest satisfaction value in cluster 2 is “ Camera Resolution ”.