Incremental Annotate-Generalize-Search Framework for Interactive Source Code Comprehension
| Autor: | Shun'ichi Tano, Eko Sakai, Ken Nakayama, Tomonori Hashiyama |
|---|---|
| Rok vydání: | 2017 |
| Předmět: |
Source code
business.industry Computer science media_common.quotation_subject Knowledge engineering 020207 software engineering 02 engineering and technology Semantics computer.software_genre Software 020204 information systems 0202 electrical engineering electronic engineering information engineering Artificial intelligence Tuple business Abstract syntax tree computer Classifier (UML) Generative grammar Natural language Natural language processing media_common |
| Zdroj: | COMPSAC (1) |
| Popis: | Understanding unfamiliar source code is inherently difficult for a software engineer, despite its importance. Thus, an experienced engineer prefers to guess the intended behavior, rather than to trace it line-by-line, by combining semantic chunks found in the source code. It is, however, still hard for a system to help in this activity, for lack of ways of both representing semantic chunks and of preparing a rich dictionary of chunks. In this paper, an integrated framework for annotating and searching source code is presented. Since the research is still in its early stage, this paper focuses on the framework itself, together with a brief description of our prototype implementation. In the framework, each engineer gathers (annotates) semantic chunks that have the same meaning and interactively generalizes them to get a search pattern. As a result, a dictionary of semantic chunks together with their search patterns is incrementally created through engineer collaboration. To realize this, two representations are used: a tuple of nodes of an abstract syntax tree (AST) for a semantic chunk and a classifier on generative attribute vectors for search patterns. |
| Databáze: | OpenAIRE |
| Externí odkaz: |
|