Hints and tips:
  • You may need to look up some unfamiliar terms in the title or description to get more of a sense of what is involved.
  • If the project has "No" in the "Current Availability:" field, it is already taken or not being offered this academic year but may be available again in future years.
  • The supervisor name links to a contact details webpage so, if you are interested, you can arrange to discuss this project or even propose a related topic of your own.
Level: BSc, MSci, MSc
Title: Twitter opinion mining using sentiment analysis
Supervisor:

Dr Silvia Liverani

Research Area: Probability and Applications [Including Statistics]
Description:

Sentiment analysis refers to statistical techniques used to retrieve the attitude of a writer with respect to a certain topic from a written document. There are a number of applications of these methods. For example they can be used to assess the overall opinion of customers about a particular product or the views of people about a particular movie or about political candidates during an election campaign. In this project, the student will retrieve data from Twitter about a topic of interest (using the R package twitteR).

The project will then explore the use of a Naıve Bayes classifier to predict the polarity of each tweet with respect to that topic (i.e positive or negative) and the associated emotion (happiness, sadness, anger, etc).

See [1] for a description of the methodology involved and [2] for a general introduction to sentiment analysis.

  1. Naive Bayes Classifier, http://www.princeton.edu/˜achaney/tmve/wiki100k/docs/Naive_Bayes_classifier.html.
  2. D. Jurafsky, C. Manning, What is sentiment analysis– Stanford NLP:

Further Reading:
Key Modules:

MTH5120 Statistical Modelling

MTH6134 Statistical Modelling II

MTH5122 Statistical Methods

MTH6136 Statistical Theory

MTH6909/MTH709U/MTHM042/MTH776P Bayesian Statistics

MTH6931/MTH731U/MTHM731 Computational Statistics.

(have passed or be currently attending at least two of the above modules)
Other Information:

Programming skills are a prerequisite for this project, as well as willingness to learn to use the statistical software R. Prior knowledge of R is not a prerequisite.

Current Availability: Yes