Code Recommendation Based on a Degree-of-Interest Model (bibtex)

by Naoya Murakami, Hidehiko Masuhara and Tomoyuki Aotani

Abstract:

Code recommendation systems predict and present what the user is likely to write next by using the user's editing context, namely textual and semantic information about the programs being edited in a programming editor. Most existing systems however use information merely around the cursor position—i.e., the class/method definition at the cursor position—as the editing context. By including the code related to the current method/class, like the callers and callees of the method, recommendation could become more appropriate. We propose to use the user's editing activity for identifying code relevant to the current method/class. Specifically, we use a modified degree-of-interest model in the Mylyn task management tool, and incorporated the model in our repository-based code recommendation system, Selene. This paper reports the design of the system and the results of our initial experiments.

Reference:

Code Recommendation Based on a Degree-of-Interest Model (Naoya Murakami, Hidehiko Masuhara and Tomoyuki Aotani), In Proeceedings of the Fourth International Workshop on Recommendation Systems in Software Engineering (RSSE 2014) (Reid Holmes, Werner Janjic, Walid Maalej, eds.), 2014.

Bibtex Entry:

@inproceedings{murakami2014rsse,
  location = {Hyderabad, India},
  organization = {{ACM}},
  isbn = {978-1-4503-2845-6/14/06},
  editor = {Reid Holmes and Werner Janjic and Walid Maalej},
  booktitle = {Proeceedings of the Fourth International Workshop on Recommendation Systems in Software Engineering (RSSE 2014)},
  date = {2014-06-03},
  year = 2014,
  month = jun,
  author = {Naoya Murakami and Hidehiko Masuhara and Tomoyuki Aotani},
  pdf = {rsse2014.pdf},
  title = {Code Recommendation Based on a Degree-of-Interest Model},
  annote = {submitted on February 1, 2014, accepted},
  doi = {10.1145/2593822.2593828},
  pages = {28--29},
  keywords = {Selene, Mylyn, Eclipse},
  abstract = {Code recommendation systems predict and present what the user is likely to write next by using the user's editing context, namely textual and semantic information about the programs being edited in a programming editor. Most existing systems however use information merely around the cursor position---i.e., the class/method definition at the cursor position---as the editing context. By including the code related to the current method/class, like the callers and callees of the method, recommendation could become more appropriate. We propose to use the user's editing activity for identifying code relevant to the current method/class. Specifically, we use a modified degree-of-interest model in the Mylyn task management tool, and incorporated the model in our repository-based code recommendation system, Selene. This paper reports the design of the system and the results of our initial experiments.}
}