Recommending random walks

Authors:

Abstract
We improve on previous recommender systems by taking advantage of the layered structure of software. We use a random-walk approach, mimicking the more focused behav- ior of a developer, who browses the caller-callee links in the callgraph of a large program, seeking routines that are likely to be related to a function of interest. Inspired by Klein- berg's work(10), we approximate the steady-state of an infi- nite random walk on a subset of a callgraph in order to rank the functions by their steady-state probabilities. Surpris- ingly, this purely structural approach works quite well. Our approach, like that of Robillard's"Suade"algorithm(15), and earlier data mining approaches (13) relies solely on the al- ways available current state of the code, rather than other sources such as comments, documentation or revision infor- mation. Using the Apache API documentation as an oracle, we perform a quantitative evaluation of our method, find- ing that our algorithm dramatically improves upon Suade in this setting. We also find that the performance of tra- ditional data mining approaches is complementary to ours; this leads naturally to an evidence-based combination of the two, which shows excellent performance on this task. Categories and Subject Descriptors: D.2.7 (Distribu- tion, Maintenance and Enhancement): Documentation

This publication has 20 references indexed in Scilit: