Computer Laboratory

Natural Language and Information Processing Research Group

Masters projects in language processing

The following list is intended to give an idea of some of the varied research topics which have been studied by Masters students in the NLIP group. Some projects lead to peer-reviewed publications: some links to such publications are shown below. New projects will be suggested each year for the ACS MPhil, according to the interests of individual students and supervisors. Many projects have the potential to be expanded into PhD work. Prospective students should look at the list of funded research projects and the web pages of individual NLIP group members for further ideas.

  • Learning Low-Rank Tensors for Compositional Distributional Semantics

    Paper based on project:
    Low-rank tensors for verbs in compositional distributional semantics. Daniel Fried, Tamara Polajnar, and Stephen Clark Proceedings of the Short Papers of the 53rd Annual Meeting of the Association for Computational Linguistics (ACL), 2015 (pdf).

  • A summariser based on a model of human memory limitations

    Paper based on project:
    Fang, Yimai, and Simone Teufel. "A summariser based on human memory limitations and lexical competition." EACL 2014 (2014): 732. (pdf)

  • What if Tony Blair had converted to Islam? Explorations in creative concept combination
  • Topic models of social network data
  • Using distributional methods in parse ranking

    Paper based on project:
    Emerson, Guy, and Ann Copestake. "Leveraging a Semantically Annotated Corpus to Disambiguate Prepositional Phrase Attachment." IWCS 2015 (2015) (pdf)

  • Dependency Language Models for Sentence Completion

    Paper based on project:
    Dependency language models for sentence completion, Joseph Gubbins and Andreas Vlachos, Proceedings of 2013 Conference on Empirical Methods in Natural Language Processing (pdf).

  • Improving Click-Through Rates using Keyword Search Terms
  • Judging the Grammaticality of Paraphrases in Context
  • Improving the Performance of a Wide-Coverage Parser on Unbounded Dependency Recovery
  • Identification of meaning differences between language varieties
  • Efficient operations on semantic dependency structures
  • Finding syntactically irregular multiword expressions in corpora
  • Unsupervised graded word sense disambiguation
  • Citation and Reference Name Resolution
  • Named Entity Recognition and Parsing
  • Using Lexical Semantic Analysis to Investigate How Conceptual Knowledge is Represented in the Human Brain
  • An Adaptable Scientific Summariser
  • Hadoop cluster for statistical speech and language processing
  • Fingerprinting poetry: features for authorship attribution
  • Integrating simulation and experiment: A study in Japanese Second Language Acquisition
  • Argumentative Zoning for improving IR in Cancer Risk Assessment
  • Lexical chain-based segmentation of scientific text
  • Automatic recognition of evidence for the Structure-Dependent Generalisation Hypothesis in child-directed discourse

This page relates to: Postgraduate opportunities in the NLIP group