Skip to main content

Efficient Question Answering with Deep Neural Networks

Primary supervisor

Shirui Pan


Question Answering (QA) has attracted a large amount of attention in recent years. However, to precisely answer the natural language questions still remains a challenge, especially the complicated questions. This is mainly because a predicate could be expressed in different ways in natural language questions. Besides, the ambiguity of entity names and partial names makes the number of possible answers large. Another challenge for building a robust QA model is that a large amount of training data is required, which is expensive in many cases. 

This project aims to solve these limitations. Novel deep learning algorithms which exploit both labelled and unlabelled data to build robust QA models will be proposed. Different sources of data, including knowledge graphs, text data, images will be exploited to answer questions accurately. The project will also look into methods which will reduce the computation and memory cost of deep neural networks, enhancing the efficiency for QA answering.

Required knowledge

Machine Learning

Deep Learning


Project funding


Learn more about minimum entry requirements.