Doctor of Philosophy (PhD)
Computer Science and Engineering
Research on mining user reviews in mobile application (app) stores has noticeably advanced in the past few years. The main objective is to extract useful information that app developers can use to build more sustainable apps. In general, existing research on app store mining can be classified into three genres: classification of user feedback into different types of software maintenance requests (e.g., bug reports and feature requests), building practical tools that are readily available for developers to use, and proposing visions for enhanced mobile app stores that integrate multiple sources of user feedback to ensure app survivability. Despite these major advances, existing tools and techniques still suffer from several drawbacks. Specifically, the majority of techniques rely on the textual content of user reviews for classification. However, due to the inherently diverse and unstructured nature of user-generated online textual reviews, text-based review mining techniques often produce excessively complicated models that are prone to over-fitting. Furthermore, the majority of proposed techniques focus on extracting and classifying the functional requirements in mobile app reviews, providing a little or no support for extracting and synthesizing the non-functional requirements (NFRs) raised in user feedback (e.g., security, reliability, and usability). In terms of tool support, existing tools are still far from being adequate for practical applications. In general, there is a lack of off-the-shelf tools that can be used by researchers and practitioners to accurately mine user reviews. Motivated by these observations, in this dissertation, we explore several research directions aimed at addressing the current issues and shortcomings in app store review mining research. In particular, we introduce a novel semantically aware approach for mining and classifying functional requirements from app store reviews. This approach reduces the dimensionality of the data and enhances the predictive capabilities of the classifier. We then present a two-phase study aimed at automatically capturing the NFRs in user reviews. We also introduce MARC, a tool that enables developers to extract, classify, and summarize user reviews.
Jha, Nishant, "EFFECTIVE METHODS AND TOOLS FOR MINING APP STORE REVIEWS" (2018). LSU Doctoral Dissertations. 4730.