DocumentCode
3758055
Title
Automatically Mining Negative Code Examples from Software Developer Q & A Forums
Author
Ryan Serva;Zachary R. Senzer;Lori Pollock;K. Vijay-Shanker
Author_Institution
Comput. &
fYear
2015
Firstpage
115
Lastpage
122
Abstract
In addition to learning good practices and reusing code from mining code examples, programmers can be supported in their learning and code improvement processes through negative, or poorly written, code examples. While it is challenging to identify negative code examples automatically from within source code, we leverage a key insight that the natural language in questions that include code examples posted on forums can provide adequate clues. In this paper, we describe an automatic sentiment analysis-based technique for mining negative code examples from developer question and answer forums along with a technique to automatically mine negative sentiment indicators commonly used by developers, which are used to drive the sentiment-based technique.
Keywords
"Data mining","Software","Sentiment analysis","Computer languages","Programming","Education"
Publisher
ieee
Conference_Titel
Automated Software Engineering Workshop (ASEW), 2015 30th IEEE/ACM International Conference on
Type
conf
DOI
10.1109/ASEW.2015.10
Filename
7426648
Link To Document