Winter Term Teaching (First Semester)
Honours’ project topics (2016-2017)
Topic 1: Semantic sentiment analysis of Twitter
This project aims at building a system which is able to (1) automatically analyze the sentiment expressed in tweets data (e.g. ‘They had a great time at the party.’ will be recognized as positive sentiment); and (2) to extract the semantically hidden concepts from tweets, referred as semantic features. For example, the entities “iPad”, “iPod” and “Mac Book Pro” are all mapped to the semantic concept PRODUCT/APPLE. (To give you some flavor of what a sentiment analysis system can offer, take a look at http://www.tweetfeel.com/)
Topic 2: Exploring and analyzing real-world data with topic models
A topic model is a type of statistical model for discovering the abstract “topics” that occur in a collection of documents. Given a trained topic model and a new/unseen document, topic model is able to automatically infer whether the new document is talking about sports or politics, or a mixture of both, for instance. This project aims to implement a system, with topic model served as the core algorithm, for exploring and visualizing large set of text collections. (Example system: http://www.princeton.edu/~achaney/tmve/wiki100k/browse/topic-presence.html)
Note: The above examples are only indicative of the kind of work I am interested in. Final project allocation will be based on discussion with students.