A Little Bird Told Me: Mining Tweets for Requirements and Software Evolution
2017 IEEE 25th International Requirements Engineering Conference (RE)(2017)
摘要
Twitter is one of the most popular social networks. Previous research found that users employ Twitter to communicate about software applications via short messages, commonly referred to as tweets, and that these tweets can be useful for requirements engineering and software evolution. However, due to their large number---in the range of thousands per day for popular applications---a manual analysis is unfeasible.In this work we present ALERTme, an approach to automatically classify, group and rank tweets about software applications. We apply machine learning techniques for automatically classifying tweets requesting improvements, topic modeling for grouping semantically related tweets and a weighted function for ranking tweets according to specific attributes, such as content category, sentiment and number of retweets. We ran our approach on 68,108 collected tweets from three software applications and compared its results against software practitioners' judgement. Our results show that ALERTme is an effective approach for filtering, summarizing and ranking tweets about software applications. ALERTme enables the exploitation of Twitter as a feedback channel for information relevant to software evolution, including end-user requirements.
更多查看译文
关键词
user feedback,Twitter,software evolution,text mining,requirements elicitation
AI 理解论文
溯源树
样例
生成溯源树,研究论文发展脉络