Replication package

RQ1, Replication: how repetitive and predictable is software?

1. Projects for each language
2. Frequency distribution for each language, n-gram with n = 1 to 10
3. Code to calculate cross-entropy. If you do not want to calculate the cross-entropy values yourself, please follow this link to get the values calculated by us.

RQ2, Artificial Repetition: how repetitive and predictable is code once we remove language specific tokens?

1. Frequency distribution for each language without separators, operators, keywords, all language specific
2. List of separators, operators, keywords for each language.
3. Code to calculate entropy

RQ3, API Usages: how repetitive and predictable are calls to the Java API?

1. Frequency distribution for each language with only Java API elements, n-grams with n = 1 to 10
2. List of Java API elements considered
3. Code to calculate entropy

RQ4, Code Graphs: how repetitive and predictable are graph representations of Java code?

1. List of tokens to ensure that n-grams and graphs contain the same set of elements (ie control flow, Java API).
2. Frequency distribution of Java graphs with nodes n = 2 to 4
3. Frequency distribution of Java tokens that are included in graphs with n-grams with n = 2 to 4
4. Graph database is available, but tools to extract graphs must be be obtained from Dr. Tien Nguyen
5. Code to calculate occurence frequency for graphs