Mining and Comparing User Reviews across Similar Mobile Apps
Yanqi Su, Yongchao Wang, Wenhua Yang · 2019
With the rapid development of the market for mobile apps, there are a number of apps with similar functions. To gain an advantage in such a competitive environment, developers need to understand not only the strengths and weaknesses of their app but also competitive apps. User reviews contain valuable information for comparing similar apps from user preference. In this paper, we propose UISAT (User-review mining via topic Identification, Sentiment Analysis and Topic matching across apps), an automated approach to compare user reviews from similar apps with the goal of mining user feedback from competitive apps by (i) extracting the hidden topics from large volumes of user reviews using topic modeling, (ii) combining a rule-based model, user rating and user-helpful for sentiment analysis of topics and (iii) matching relevant topics across apps. Empirical studies demonstrate that UISAT is effective and promising for developers to build and maintain a more competitive app.