| 本文已被：浏览 1987次 下载 3944次 |
码上扫一扫！ |
|
|
| 一种语义感知的细粒度App评论缺陷挖掘方法 |
|
王亚文1,2, 王俊杰1,2, 石琳1,2, 王青1,2,3
|
|
1.中国科学院软件研究所 互联网软件技术实验室, 北京 100190;2.中国科学院大学, 北京 100049;3.计算机科学国家重点实验室(中国科学院 软件研究所), 北京 100190
|
|
| 摘要: |
| 手机用户提交的App评论为开发者提供了一个了解用户满意度的沟通渠道. 许多用户通常使用“send a video”和“crash”等关键短语来描述有缺陷的功能(即用户操作)和App的异常行为(即异常行为), 而这些短语可能会与其他琐碎信息(如用户的抱怨)一起交杂在评论文本中. 掌握这些细粒度信息可以帮助开发者理解来自用户的功能需求或缺陷报告, 进而有利于提升App的质量. 现有的基于模式的目标短语提取方法只能对评论的高层主题/方面进行总结, 并且由于对评论的语义理解不足, 短语提取的性能较差. 提出了一种语义感知的细粒度App评论缺陷挖掘方法(Arab), 来提取用户操作和异常行为, 并挖掘两者之间的关联关系. 设计了一种新颖的用于提取细粒度目标短语的神经网络模型, 该模型将文本描述和评论属性相结合, 能更好地建模评论的语义. Arab还根据语义关系对提取的短语进行聚类, 并将用户操作和异常行为之间的关联关系进行了可视化. 使用6个App的3 426条评论进行评估实验, 实验结果证实了Arab在短语提取方面的有效性. 进一步使用Arab对15个热门App的301 415条评论进行了案例研究, 以探索其潜在的应用, 并验证其在大规模数据上的实用性. |
| 关键词: App评论 信息提取 深度学习 |
| DOI：10.13328/j.cnki.jos.006697 |
| 分类号:TP311 |
| 基金项目:国家重点研发计划(2018YFB1403400); 国家自然科学基金(62072442) |
|
| Semantic-aware and Fine-grained App Review Bug Mining Approach |
|
WANG Ya-Wen1,2, WANG Jun-Jie1,2, SHI Lin1,2, WANG Qing1,2,3
|
|
1.Laboratory for Internet Software Technologies, Institute of Software, Chinese Academy of Sciences, Beijing 100190, China;2.University of Chinese Academy of Sciences, Beijing 100049, China;3.State Key Laboratory of Computer Sciences (Institute of Software, Chinese Academy of Sciences), Beijing 100190, China
|
| Abstract: |
| App reviews are considered as a communication channel between users and developers to perceive user’s satisfaction. Users usually describe buggy features (i.e., user actions) and App abnormal behaviors (i.e., abnormal behaviors) in forms of key phrases (e.g., “send a video” and “crash”), which could be buried with other trivial information (e.g., complaints) in the review texts. A fine-grained view about this information could facilitate the developers’ understanding of feature requests or bug reports from users, and improve the quality of Apps. Existing pattern-based approaches to extract target phrases can only summarize the high-level topics/aspects of reviews, and suffer from low performance due to insufficient semantic understanding of reviews. This study proposes a semantic-aware and fine-grained App review bug mining approach (Arab) to extract user actions and abnormal behaviors, and mine the correlations between them. A novel neural network model is designed for extracting fine-grained target phrases, which combines textual descriptions and review attributes to better represent the semantics of reviews. Arab also clusters the extracted phrases based on their semantic relations and provides a visualization of correlations between User Actions and Abnormal Behaviors. 3,426 reviews from six Apps are used to carry out evaluation test, and the results confirm the effectiveness of Arab in phrase extraction. A case study is further conducted with Arab on 301,415 reviews of 15 popular Apps to explore its potential application and examine its usefulness on large-scale data. |
| Key words: App review information extraction deep learning |
