Methods and systems for automated software natural language documentation

A method for automatically generating natural language documentation for source code files, the method comprising:

• parsing a source code file to extract a function and generate an abstract syntax tree;
• training a programming language neural network model on a plurality of key constructs of a programming language comprising at least two of syntax expression, functions, function signatures, programming language keywords and associated documentation;
• generating first natural language documentation for each of a plurality of program statements within the function using a programming language neural network model;
• generating second natural language documentation for the function as a whole by processing a code snippet of the function using a function documentation neural network model;
• consolidating the first natural language documentation and the second natural language documentation at a function level, a source file level, and a project level to create consolidated natural language documentation;
• and summarizing the consolidated natural language documentation using a customized summarization routine.

A computer program product for automatically generating natural language documentation for source code files, comprising a processor and memory storing instructions thereon, wherein the instructions when executed bythe processor cause the processor to:

• parse a source code file to identify an executable function within the source code file and a plurality of program statement lines of code within the executable function;
• train a programming language neural network model on a plurality of key constructs of a programming language comprising at least two of syntax expression, functions, function signatures, programming language keywords and associated documentation;
• generate first natural language documentation for each of the plurality of program statement lines of code using a programming language specification trained neural network model;
• aggregate the first natural language documentation for each of the plurality of program statement lines of code to generate an aggregated natural language documentation for the function;
• independent of the aggregated natural language documentation, generate a second natural language documentation for the function based on a code snippet of the function as a whole using a source code documentation trained neural network model;
• and
• combine the aggregated natural language documentation and the second natural language documentation to generate a combined natural language software documentation.