Studio Ousia's Quiz Bowl Question Answering System
This work addresses factoid question answering for quiz bowl competitions, demonstrating a system that outperforms both other AI systems and human experts, though it is incremental in combining existing and new methods.
The authors tackled the quiz bowl question answering task by combining novel neural network models with conventional information retrieval models, achieving the best performance in the HCQA competition and winning against top human experts by a wide margin.
In this chapter, we describe our question answering system, which was the winning system at the Human-Computer Question Answering (HCQA) Competition at the Thirty-first Annual Conference on Neural Information Processing Systems (NIPS). The competition requires participants to address a factoid question answering task referred to as quiz bowl. To address this task, we use two novel neural network models and combine these models with conventional information retrieval models using a supervised machine learning model. Our system achieved the best performance among the systems submitted in the competition and won a match against six top human quiz experts by a wide margin.