Expert Recommendation System for StackOverflow

Author/Creator

Author/Creator ORCID

Date

2017-01-01

Type of Work

Department

Computer Science and Electrical Engineering

Program

Computer Science

Citation of Original Publication

Rights

This item may be protected under Title 17 of the U.S. Copyright Law. It is made available by UMBC for non-commercial research and education. For permission to publish or reproduce, please see http://aok.lib.umbc.edu/specoll/repro.php or contact Special Collections at speccoll(at)umbc.edu
Distribution Rights granted to UMBC by the author.

Abstract

Identifying Subject Matter Experts (SMEs) is crucial to Community Question Answering (CQA) systems. The success of CQA systems heavily relies on the contribution of these experts who provide a significant number of high-quality, useful answers. SO is a community-based question answering platform for developers to ask technical questions. We propose a novel approach to find SMEs for StackOverflow (SO) in an unsupervised manner. Our technique uses the Latent Dirichlet Allocation (LDA) model and Latent Semantic Analysis (LSA) to automatically predict the skill-set needed to answer questions based on their content and find experts with the same skill-set. The effectiveness of this approach is demonstrated through comprehensive experiments on the SO dataset for Python, C++, Java and C# programming languages by considering SO threads of a configurable elapsed time window and predicting who will answer questions in the following month.