FST Graduate Research Topic

Research Topic: Probabilistic Graphical Models for Text analysis
Suggested By: Dunwei (Grant) Wen
Last Update: September 9th, 2014

This research will focus on applying probabilistic graphical models, including topic models and Bayesian non-parametric models to natural language text processing and analysis. Machine learning and inference is the key to build and apply these models, which can be easily scaled in large text resources--a perfect fit for big data analysis. The specific task of text analysis can be chosen from a selection of related tasks, including topic recognition, document summarization, document classification, and semantic analysis, depending on the student's backgroudna d interest. Each text analysis task can be used for a range of real world applications. For instance, topic recognition may be applied to analyze conversations and documents to automatically recognize the included topics and sentiments. Topic recognition can be used in social networks, computer games and mobile learning environment for classifying and clustering discussions and users, and for facilitating some personalized services such as interest group recommendation. The graduate students who want to pursue the research need to be familiar with Artificial Intelligence, especially machine learning and natural language processing. This research topic is suitable for Project or Thesis.