A Collection of Semantically Related Words from the Paper:

SWordNet: Inferring Semantically Related Words from Software Context

Introduction

We make publicly available manually verified semantically related words (referred to as rPairs) in order for other people to:

• Check our results, in the spirit of any scientific work.
• Provide a software version of WordNet and share with the research community.

We have chosen to publish the correct rPairs by manual verification in our MSR'12 paper.

Publication

SWordNet: Inferring Semantically Related Words from Software Context. Jinqiu Yang and Lin Tan. Accepted to the Springer Empirical Software Engineering. (28 pages)

Inferring Semantically Related Words from Software Context. Jinqiu Yang and Lin Tan. In the proceedings of the Working Conference on Mining Software Repositories. June, 2012. Zurich, Switherland. (10 pages)

Data

In each rPair file, Column 1&2 are a pair of words, Column 3 shows the one of the five category (synonym, antonym, near antonym, related, identifier, please refer to our MSR'12 paper for more details). Column 4&+ shows the context where this rPair is discovered.

The stopword list used in the experiment is here.
rPair files can be found in the following links:

• the Linux kernel:  comment  code  comment-code

People

• Jinqiu Yang
• Lin Tan