Previous studies showed that replying to a user review usually has a positive effect on the rating that is given by the user to the app. For example, Hassan et al. found that responding to a review increases the chances of a user updating their given rating by up to six times compared to not responding. To alleviate the labor burden in replying to the bulk of user reviews, developers usually adopt a template-based strategy where the templates can express appreciation for using the app or contain the company email for users to react. However, reading large numbers of user reviews every day is not an easy task for developers. The available review-response pairs provide us chances to learn the knowledge relations between reviews and responses.
Although there exists research on studying the popular review patterns (e.g., reviews with longer content and lower rating) that developers tend to respond, approaches to automate the review response process have never been proposed. Inspired by the RNN encoder-decoder model in the natural language processing field, we propose a response generation framework, named RRGen. RRGen explicitly incorporates review attributes, such as user rating and review length, and learns the relations between reviews and corresponding responses in a supervised way from the available training data. Experiments on 58 apps and 309,246 review-response pairs highlight that RRGen outperforms several baselines by 67.4% to 4.5 times in terms of BLEU (an accuracy measure that is widely used to evaluate generation systems). Qualitative analysis also confirms the effectiveness of RRGen in generating the relevant and accurate response.
Tue 12 Nov
13:40 - 15:20: Papers - Natural Language and Human Aspects at Cortez 2&3 Chair(s): Bogdan VasilescuCarnegie Mellon University | ||||||||||||||||||||||||||||||||||||||||||
13:40 - 14:00 Talk | Discovering, Explaining and Summarizing Controversial Discussions in Community Q&A Sites Xiaoxue RenZhejiang University, Zhenchang XingAustralia National University, Xin XiaMonash University, Guoqiang LiShanghai Jiao Tong University, Jianling SunZhejiang University Pre-print | |||||||||||||||||||||||||||||||||||||||||
14:00 - 14:20 Talk | Automating App Review Response Generation Cuiyun GaoNanyang Technological University, Singapore, Jichuan ZengThe Chinese University of Hong Kong, Xin XiaMonash University, David LoSingapore Management University, Michael LyuThe Chinese University of Hong Kong, Irwin KingThe Chinese University of Hong Kong Pre-print | |||||||||||||||||||||||||||||||||||||||||
14:20 - 14:40 Talk | Automatic Generation of Pull Request DescriptionsACM SIGSOFT Distinguished Paper Award Zhongxin LiuZhejiang University, Xin XiaMonash University, Christoph TreudeThe University of Adelaide, David LoSingapore Management University, Shanping LiZhejiang University Pre-print | |||||||||||||||||||||||||||||||||||||||||
14:40 - 15:00 Talk | Recommending Who to Follow in the Software Engineering Twitter Space Abhishek Sharma Singapore Management University, Singapore, Yuan TianQueens University, Kingston, Canada, Agus SulistyaSchool of Information Systems, Singapore Management University, Dinusha WijedasaSchool of Information Systems, Singapore Management University, David LoSingapore Management University Pre-print | |||||||||||||||||||||||||||||||||||||||||
15:00 - 15:10 Demonstration | Developer Reputation Estimator (DRE) Sadika AmreenUniversity of Tennessee Knoxville, Andrey KarnauchUniversity of Tennessee Knoxville, Audris MockusUniversity of Tennessee - Knoxville | |||||||||||||||||||||||||||||||||||||||||
15:10 - 15:20 Demonstration | CocoQa: Question Answering for Coding Conventions over Knowledge Graphs Tianjiao DuShanghai JiaoTong University, Junming CaoShanghai JiaoTong University, Qinyue WuShanghai JiaoTong University, Wei LiShanghai JiaoTong University, Beijun ShenSchool of Electronic Information and Electrical Engineering, Shanghai Jiao Tong University, Yuting ChenShanghai Jiao Tong University |