Hierarchical Expert Recommendation on Community Question Answering Platforms

Abstract

The community question answering (CQA) platforms, such as Stack Overflow (SO), have become the primary source of answers to most questions in various topics. CQA platforms offer an opportunity for sharing and acquiring knowledge at a low cost, where users, many of whom are experts in a specific topic, can potentially provide high-quality solutions to a given question. Our proposed recommendation system will recommend the best possible question to an expert who will probably answer this question. This project will be a content-based recommendation system that utilizes hierarchical attention-based neural networks to model users’ expertise and questions in these platforms. Our model contains two major components, i.e., Dense Encoder, a deep neural network to provide latent features for users and questions, and Hierarchical Encoder to learn dense question representations from the words and sentences. Also, we utilize attention layers in our model to select essential words and most important sentences. Our preliminary results show that our proposed recommendation system outperforms state-of-the-art baselines on a real-world dataset.