Computational Linguistics

The appointment of Stephen Clark in 2004 and the move in 2006 from Oxford's Linguistics department by Prof. Stephen Pulman established a leading group in computational linguistics. Their work encompasses and combines established knowledge-based approaches with statistical and machine learning methods. Their new group, currently comprising 15 members, is growing rapidly and has been particularly successful in attracting top-class doctoral students. The group has links with industry such as Sharp Laboratories of Europe, currently sponsoring an EPSRC CASE studentship, and Corpora Software, fully funding a DPhil student. The group also has interdisciplinary links with other Oxford departments, a notable example being a collaboration with the Engineering Department on robot navigation. It is a partner in the 12M Euro FP6 Companions project for research into learning methods for human-machine dialogues, and the FP7: Europa European Robotic Pedestrian Assistant project. Phil Blunsom joined the group in Oct 2009, replacing Stephen Clark who has moved to Cambridge.

Research

• Quantum Linguistics
• Variational Bayesian Inference
• Vector Space Modelling of Semantics
• Deep Learning

Selected Publications

View All

• Investigating the Role or Prior Disambiguation in Deep−Learning Compositional Models of Meaning

  Jianpeng Cheng‚ Dimitri Kartsaklis and Edward Grefenstette

  In Learning Semantics workshop‚ NIPS 2014. Montreal‚ Canada. December, 2014.

  Details about Investigating the Role or Prior Disambiguation in Deep−Learning Compositional Models of Meaning | BibTeX data for Investigating the Role or Prior Disambiguation in Deep−Learning Compositional Models of Meaning | Download (pdf) of Investigating the Role or Prior Disambiguation in Deep−Learning Compositional Models of Meaning

• Evaluating Neural Word Representations in Tensor−Based Compositional Settings

  Dmitrijs Milajevs‚ Dimitri Kartsaklis‚ Mehrnoosh Sadrzadeh and Matthew Purver

  In Proceedings of the 2014 Conference on Empirical Methods in Natural Language Processing (EMNLP). Doha‚ Qatar. October, 2014. Association for Computational Linguistics.

  Details about Evaluating Neural Word Representations in Tensor−Based Compositional Settings | BibTeX data for Evaluating Neural Word Representations in Tensor−Based Compositional Settings | Download  (pdf) of Evaluating Neural Word Representations in Tensor−Based Compositional Settings | Download (pdf) of Evaluating Neural Word Representations in Tensor−Based Compositional Settings

• Compositional Morphology for Word Representations and Language Modelling

  Jan A. Botha and Phil Blunsom

  In Proceedings of the 31st International Conference on Machine Learning (ICML). Beijing‚ China. June, 2014.

  *Award for best application paper*

  Details about Compositional Morphology for Word Representations and Language Modelling | BibTeX data for Compositional Morphology for Word Representations and Language Modelling | Download (pdf) of Compositional Morphology for Word Representations and Language Modelling