Install the packages by using pip install -r requirements.txt

ref file contains the refs to the keywords suggested on the comments

main.py has an usage example, using the Pull Request file. The file has a list of PRs with all its info.

output_<project_name>.json contains the features

In that output we are extracting the following features:

• number: PR number

• title_n_words: PR title number of words
• description_n_words: PR description number of words
• comments_n_words: PR comments number of words
• comments_review_n_words PR review comments number of words

• n_commits Number of commits associated with the PR
• median_words_commits: Median of words in commits associated with the PR

• core_all: Number of keywords (in general) mentioned by core developers on comments and review comments
• contributor_all: Number of keywords (in general) mentioned by contributors developers on comments and review comments
• newcomer_all: Number of keywords (in general) mentioned by newcomers developers on comments and review comments
• core_maint: Number of maintainability keywords mentioned by core developers on comments and review comments
• contributor_maint: Number of maintainability keywords mentioned by contributor developers on comments and review comments
• newcomer_maint: Number of maintainability keywords mentioned by newcomer developers on comments and review comments
• core_sec: Number of security keywords mentioned by core developers on comments and review comments
• contributor_sec: Number of security keywords mentioned by contributor developers on comments and review comments
• newcomer_sec: Number of security keywords mentioned by newcomers developers on comments and review comments
• core_perf: Number of performance keywords mentioned by core developers on comments and review comments
• contributor_perf: Number of performance keywords mentioned by contributor developers on comments and review comments
• newcomer_perf: Number of performance keywords mentioned by newcomer developers on comments and review comments
• core_robu: Number of robustness keywords mentioned by core developers on comments and review comments
• contributor_robu: Number of robustness keywords mentioned by contributor developers on comments and review comments
• newcomer_robu: Number of robustness keywords mentioned by newcomer developers on comments and review comments

• title_all: Number of keywords (in general) in PR title
• description_all: Number of keywords (in general) in PR description
• comments_all: Number of keywords (in general) in PR comments
• review_comments_all: Number of keywords (in general) in PR review comments
• commits_all: Number of keywords (in general) in PR associated commits
• general_all: Number of keywords (in general) in all artifacts
• title_maint: Number of maintainability keywords in PR title
• description_maint: Number of maintainability keywords in PR description
• comments_maint: Number of maintainability keywords in PR comments
• review_comments_maint: Number of maintainability keywords in PR review comments
• commits_maint: Number of maintainability keywords in PR associated commits
• general_maint: Number of maintainability keywords in all artifacts
• title_sec: Number of security keywords in PR title
• description_sec: Number of security keywords in PR description
• comments_sec: Number of security keywords in PR comments
• review_comments_sec: Number of security keywords in PR review comments
• commits_sec: Number of security keywords in PR associated commits
• general_sec: Number of security keywords in all artifacts
• title_perf: Number of performance keywords in PR title
• description_perf: Number of performance keywords in PR description
• comments_perf: Number of performance keywords in PR comments
• review_comments_perf: Number of performance keywords in PR review comments
• commits_perf: Number of performance keywords in PR associated commits
• general_perf: Number of performance keywords in all artifacts
• title_robu: Number of robustness keywords in PR title
• description_robu: Number of robustness keywords in PR description
• comments_robu: Number of robustness keywords in PR robustness comments
• review_comments_robu: Number of robustness keywords in PR review comments
• commits_robu: Number of robustness keywords in PR associated commits
• general_robu: Number of robustness keywords in all artifacts