Research Fellow in Natural Language Processing and Information Extraction (1288620FP-R)

You will join our team working on deep learning-based language models for information extraction and data rescue in the GloSAT (Global Surface Air Temperature) research project. The post will be in the Agents, Interaction and Complexity (AIC) group, which is part of the department of Electronics and Computer Science (ECS) at the University of Southampton. The AIC group leads a number of nationally important research centres including the Trustworthy Autonomous System Research Hub (tas.ac.uk) and the UKRI MINDS Centre for Doctoral Training (mindscdt.ai).

The Paris Climate Agreement defines an ambition to limit global temperature change to between 1.5°C and 2°C above pre-industrial levels. In its Fifth Assessment report (AR5) the Intergovernmental Panel on Climate Change (IPCC) used a baseline of 1850-1900 for its definition of 'pre-industrial'. However, it has been estimated that global temperatures may have already increased by 0.0-0.2°C by this time, but this is uncertain due to a lack of data. Differences between the global temperature datasets arise primarily from two structural uncertainties: the use of sea surface temperatures (SST) rather than air temperatures over the oceans, especially ice-covered regions, and differences in data coverage and interpolation strategies. The GloSAT project (www.glosat.org) addresses both of these uncertainties using a combination of data rescue from historical documents to improve climate measurement data records, and improved climate change models making use of these improved records.

An important aspect of data rescue is information extraction from typed and handwritten historical ship and land measurement station logbooks. Existing initiatives (www.weatherrescue.org) use crowdsourced volunteers to manually enter data tables for thousands of scanned images of measurement log book pages. This approach does not scale to the millions of scanned images of logbook pages waiting to be processed, but it does provide an excellent ground truth for an annotated training corpus.

Your role will be to explore deep learning-based natural language models for information extraction leading to advances in column classification, relation extraction and knowledge-base population of tabular corpora. Identified domain context such as measurement type will be used for abnormality detection to improve the quality of the final measurement records. Tools for automated document layout analysis and optical character recognition have been developed within GloSAT to process millions of scanned images of logbook pages and generate a corpus of spatially annotated textual table components suitable for language modelling.

You will work in the context of a research partnership between the University of Southampton and UK research institutions including the National Oceanography Centre (NOC) and UK Met Office. To be successful you will have a PhD* in natural language processing, artificial intelligence or a closely related subject and have experience of information extraction. In addition to research, an important part of the role is to interact with project partners, and to become familiar with the environmental science domain.

The appointment is full time fixed term until 30/09/2022.

The department of Electronics and Computer Science is the leading university department of its kind in the UK, with an international reputation for world-leading research across computer science, electronics, and electrical engineering. Research takes place in a multidisciplinary, collaborative environment and draws on our outstanding facilities. With over 550 researchers from many different subject backgrounds, the research culture in ECS is fast-changing and dynamic. Our internationally renowned teaching and research have been ranked among the highest in the UK.

Equality, diversity and Inclusion is central to the ethos in the School of Electronics and Computer Science. We particularly encourage women, Black, Asian and minority ethnic (BAME), LGBT and disabled applicants to apply for this position. We are committed to improving equality for women in science and have been successful in achieving an Athena SWAN bronze award in April 2020. We give full consideration to applicants that wish to work flexibly including part-time and due consideration will be given to applicants who have taken a career break. The University has a generous maternity policy*, onsite childcare facilities and other benefits such as the cycle to work scheme.

The University of Southampton is in the top 1% of world universities and in the top 10 of the UK’s research-intensive universities. The University of Southampton is committed to sustainability and being a globally responsible university and has recently been awarded the Platinum EcoAward.  Our vision is to embed the principles of sustainability into all aspects of our individual and collective work, integrating sustainable development into our business planning, policy-making, and professional activities.  This commits all of our staff and students to take responsibility for managing their activities to minimise harm to the environment, whether this through switching off non-essential electrical equipment or using the recycling facilities.

For informal enquiries before submitting your application, contact Dr Stuart E. Middleton (sem03@soton.ac.uk). 

Please include a covering letter and full CV in your application. You will also need to specify at least two references on the application form.

At the University of Southampton, we value diversity and equality.

*Applications for Research Fellow positions will be considered from candidates who are working towards or nearing completion of a relevant PhD qualification.  The title of Research Fellow will be applied upon successful completion of the PhD.  Prior to the qualification being awarded the title of Senior Research Assistant will be given.

Application Procedure: 

You should submit your completed online application form at https://jobs.soton.ac.uk. The application deadline will be midnight on the closing date stated above. If you need any assistance, please call Kate Pounds (Recruitment Team) on +44 (0) 23 8059 5456. Please quote reference 1288620FP-R on all correspondence.

Closing Date: 09 Dec 2020
Post Type: Education, Research & Enterprise