Saved in:
| Main Author: | |
|---|---|
| Format: | Recurso digital |
| Language: | |
| Published: |
Zenodo
2026
|
| Online Access: | https://doi.org/10.5281/zenodo.19234919 |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
Table of Contents:
- <p>This repository contains the data and scripts used in the paper:</p> <p>"Beyond the Good First Issue: Progressive Recommendation of Second Tasks for Newcomers in OSS Communities"</p> <p> </p> <h2>Data-Preparation</h2> <p>The folder <code>Data-Preparation/</code> includes the following resources:</p> <ul> <li> <p><code>NewcomerList.csv</code> A list of identified newcomers based on the public GFI-Bot dataset [1].</p> </li> <li> <p><code>PRInfo/</code>, <code>PRComments/</code>, <code>issueInfo/</code>, <code>IssueComments/</code>, <code>commitInfo/</code> Activity data collected via the GitHub API for repositories involving these newcomers, including pull requests, issues, comments, and commits.</p> </li> <li> <p><code>get_issue_chain.py</code> Identifies issues resolved by newcomers and extracts resolver information.</p> <ul> <li> <p>Outputs issue resolution data to <code>issueGetResolver/</code></p> </li> <li> <p>Constructs newcomer task chains and stores them in <code>IssueChainGE2/</code></p> </li> </ul> </li> </ul> <div> </div> <h2>RQ1</h2> <p>The folder <code>RQ1/</code> contains scripts for feature extraction and statistical analysis:</p> <ul> <li> <p><code>cal_issue_feature.py</code> Extracts and enriches issue features used in the RQ1 analysis.</p> </li> <li> <p><code>similarity-compare.py</code> Computes feature similarity between issues and performs Wilcoxon tests for issue-level similarity analysis.</p> <ul> <li> <p>Output: <code>RQ1/Result/similarity_wilcoxon_results.csv</code></p> </li> </ul> </li> <li> <p><code>progression-compare.py</code> Evaluates issue-level progression patterns using Wilcoxon tests.</p> <ul> <li> <p>Output: <code>RQ1/Result/progression_wilcoxon_results.csv</code></p> </li> </ul> </li> <li> <p><code>network-compare.py</code> Gets social network metrics and performs Wilcoxon tests for social network analysis.</p> <ul> <li> <p>Output: <code>RQ1/Result/network_wilcoxon.csv</code></p> </li> </ul> </li> </ul> <div> </div> <h2>RQ2</h2> <p>The folder <code>RQ2/</code> contains scripts for dataset construction and model evaluation:</p> <ul> <li> <p><code>0-get_first_participate_time.py</code> Identifies the timestamp of a newcomer’s first participation in the second issue (recommendation time).</p> </li> <li> <p><code>1-get_candidate_list.py</code> Constructs candidate issue lists for each newcomer.</p> </li> <li> <p><code>2-get_candidate_issue_feature.py</code> Extracts features for each (first issue, candidate issue) pair, including:</p> <ul> <li> <p>Attributes of the first issue (FI)</p> </li> <li> <p>Attributes of candidate issues</p> </li> <li> <p>Similarity features between FI and candidate issues</p> </li> <li> <p>Output stored in <code>CandidateIssueInfo/</code></p> </li> </ul> </li> <li> <p><code>3-get_rec_dataset.py</code> Constructs the dataset for training and evaluating recommendation models.</p> </li> <li> <p><code>4-recommend.py</code> Trains learning-to-rank models and performs evaluation and ablation studies.</p> </li> </ul> <div> </div> <h2>Reference</h2> <p>[1] He H, Su H, Xiao W, et al. GFI-Bot: Automated Good First Issue Recommendation on GitHub Proceedings of the 30th ACM Joint European Software Engineering Conference and Symposium on the Foundations of Software Engineering (ESEC/FSE), 2022, pp. 1751–1755.</p>