I'm generally interested in projects relating to Information Retrieval, Data/Text Mining and Recommendations.
Nevertheless, I am open to consider your own ideas as long as they are aligned to my research interests.
Below are some potential project ideas.
[1] AI-driven Conversational Social bot
Socialbots (such as Amazon Alexa) can converse coherently and engagingly with humans on a range of current events and popular topics such as entertainment, sports, politics, technology, and fashion. Such socialbots can be developed through exploiting deep learning and reinforcement learning. In this project you will design and implement a key component of such conversational agent (see Amazon Alexa Challenge for more details).
[2] Empathic Conversational Agent
Social bots currently can not respond with attached emotions. In this project, you aim to improve upon current conversational bot by making them more empathic. You will have the opportunity to learning deep learning and develop such bot.
[3] House Price Prediction
Accurately predicting the property price and visualize them in a dynamic environment is important for a number of applications.
In this project you will design and implement:
a. a crawler that seeks other useful data (such as crime, economical, etc.) that can complement the house property data. The additional data can be served as useful complementary features.
b. a property feature mining tool that can characterize many different characteristics of the property (textual and visual). You need to dig into especially use natural language processing and vision techniques to generate features from both textual and visual contents. Then you can use a machine (deep) learning algorithm to accurately predict the house price given many mined features.
[4] House Price Visualization
UK government has provided land registry data about all the house transactions throughout UK. This spatial-temporal data spans across different locations and over 20 years.
How to visualize those data and help others to explore its linkage to other economical data in a meaningful way is what this project aims to investigate.