PhD student profiles - NLP research group

Ahmet Aker

The number of images tagged with location information on the web is growing rapidly. GPS equipped cameras and phones make it possible to take pictures that are already indexed with GPS coordinates, and online social sites like Flickr and Facebook ensure quick spread of these images.

However, GPS coordinates and/or minimal user written captions are the only descriptions for the majority of these images. This makes image indexing, organization and search a difficult task. Therefore methods which could automatically supplement the information available for image indexing and lead to improved image retrieval would be extremely useful.

I am working on a technique for automatic image captioning or enhancement of existing captions. My captioning technique takes only a set of place names pertaining to an image. For example an image showing Eiffel Tower needs to be tagged with the name “Eiffel Tower”. We have methods to obtain these place names using GPS information only. This place name is passed to my system that then generates a short description about it. This technique has an advantage over related work in automatic image captioning in that it is sufficient to have only the GPS information associated with the image to generate captions.

Before I started with my PhD in 2007 I did my Masters in Software Engineering at the University of Sheffield.

Click here to visit Ahmet's personal web page.

Leon Derczynski

After working with text processing for two years doing SEO in industry in the UK and then taking a break to teach English in China, I completed a Masters in Computer Science at the University of Sheffield. Commercial work made me interested in pursuing research and the department's excellent post-graduate programme catered for this, leading to my first publication as an undergraduate. I applied for a full time research Ph.D. within the department and received full funding.

My studies now relate to temporal information extraction - that is, developing systems and algorithms to comprehend the use of tense and time in language. This critical part of human communication lets us plan actions and understand previous events. Previous research in the NLP group at the University of Sheffield has been key to the development of temporal annotation standards and temporal information extraction, so the people we have here are exactly the right people to help with my work. We have a large and diversely populated research group, occupying a decent amount of space at the department, which we share with researchers in many fields. Some of these are very related to NLP - Speech and Hearing, Organisations Information and Knowlege, and Machine Learning - which leads to some interesting discussions around the department.

As well as benefiting from the knowledge within just a few metres of my desk, work in temporal information extraction is at the cutting edge of NLP and it is not hard to find interesting results or investigations that can lead to a useful publication. The global temporal IE community is friendly and having good academics in the department makes social connections available that might not be found through other universities.