Weighted Abstract Syntax Trees for Program Comprehension in Java

In this paper, we examine the generation and applications of Weighted Abstract Syntax Trees (WAST) for Java based on usage statistics. We discuss several use cases for WASTs, and look into how they can aid in program comprehension. Program comprehension is the activity that developers take part in to understand the source code of a software system. A Weighted Abstract Syntax Tree's potential uses in program comprehension is showcased and an algorithm for generating them is described. The algorithm's results are demonstrated with open-source Java projects. We suggest that the algorithm can be used to analyse the characteristics of Java projects, which would aid in program comprehension.