Exploiting Natural Language Structures in Software Informal Documentation

Communication means, such as issue trackers, mailing lists, Q&A forums, and app reviews, are premier means of collaboration among developers, between developers end-users. Analyzing sources information is crucial to build recommenders for example suggesting experts, re-documenting source code, or transforming user feedback in maintenance evolution strategies developers. To ease this analysis, previous work we proposed Development Emails Content Analyzer (DECA), a tool based on Natural Language Parsing that classifies with high precision development emails' fragments according their purpose. However, DECA has be trained through manual tagging relevant patterns, which often effort-intensive, error-prone requires specific expertise natural language parsing. In paper, first show, an empirical study, the extent producing rules identifying patterns effort, depending nature complexity patterns. Then, propose approach, named Nlp-based softwarE dOcumentation aNalyzer (NEON), automatically mines rules, minimizing effort. We assess performances NEON analysis classification mobile discussions, issues. simplifies identification definition processes, allowing savings more than 70 percent time otherwise spent performing activities manually. Results also show NEON-generated close manually identified ones, achieving comparable recall.