UNDERSTANDING SERVICE QUALITY OF MOBILE VIDEO EDITING : MAPPING THE NEGATIVE IMPRESSION BY TEXT MINING APPROACH

KineMaster is a video editing application that supports the content creator industry; however, compared to its competitors, that app falls short in release year, download numbers, and ratings. This research aims to determine the service quality of the Android-based KineMaster application based on sentiment analysis and the classification of mobile app service quality (MASQ) dimensions. The data used is secondary data from 5,000 reviews of Google Play Store using Google Colab and processed using RapidMiner Studi version 10.2. Naïve Bayes and k-Nearest Neighbors (KNN) algorithms are applied to determine the best one. Negative sentiment data resulting from the worst MASQ dimension classification will be carried out by WordCloud using Google Colab to determine complaint priorities. The research results show that positive sentiment dominates at 62.24% using the KNN algorithm as the best algorithm in this research. Nevertheless, the 37.76% negative sentiment is not ignored. Based on the number of negative sentiments in each dimension, technical reliability is the worst dimension, valence is the second worst dimension, and performance is the third worst. Prioritized complaints are update reliability, watermarks, app, feature downloads, inability to open apps, export capabilities, high price, and processing speed.