Log in
The Natural Language Toolkit (NLTK) is a widely-adopted Python library for natural language processing. NLTK is run as an open source project. Three project leaders, Steven Bird (Melbourne University), Edward Loper (BBN, Boston) and Ewan Klein (University of Edinburgh) provide the strategic direction of the NLTK project.
NLTK has been widely used in academia, commercial / non-profit organisations and public bodies, including Stanford University and the Educational Testing Service (ETS), which administers widely-recognised tests across more than 180 countries. NLTK has played an important role in making core natural language processing techniques easy to grasp, easy to integrate with other software tools, and easy to deploy.
The Unit's work has had a direct impact on the child protection system in England. In 2010, the Secretary of State for Education invited Eileen Munro to review the child protection system, giving her a wide remit enabling her to address systemic factors (such as the inspection framework, statutory guidance and performance management systems) as well as front line practice. All 15 recommendations of her final 2011 report [B] have been accepted and are being implemented. Munro's research has had significant reach: she has given evidence to two state government reviews of child protection in Australia; and in Queensland a charity is running a campaign to persuade the state government to learn from her work.
The Statistical Cybermetrics Research Group (SCRG) has developed social science sentiment analysis methods that estimate the strength of positive and negative sentiment in short informal social web text. These methods are encapsulated in the SentiStrength software, which is sold commercially, used commercially to develop socially useful computing applications (e.g., question answering systems, customer relations management systems), used to engage the public in science-related entertaining events, and used for data journalism to inform the public about specific news events. The research includes the development and evaluation of new sentiment analysis techniques that can detect informal expressions of sentiment in social web texts and that can detect the strength of positive and negative sentiment and not just its polarity. The research also includes the development of commercially viable software that includes the sentiment analysis methods.
The research has economic impact by enhancing the performance of commercial software systems, benefitting the owners of these systems (e.g., Yahoo!, Inbenta, Gemius, New Cities Foundation). The research also has economic impact by enhancing the customer relations of companies using sentiment-enhanced customer relations management systems, and with the traffic congestion detection system helping people to get to work on time. It has wide public services impact by helping people to find answers to their questions (via Yahoo! Answers). It has societal impact by supporting newsworthy analyses of social phenomena for the media. It has enhanced cultural life by driving spectacular lightshows during the London Olympics.
Since 2006 the University of East Anglia (UEA) has led a series of Government commissioned studies of all Serious Case Reviews of child death and serious injury in England. This work has provided the largest national database of analyses of child deaths and serious injury where abuse or neglect are known or suspected.
Since 2008, the findings have informed public understanding, practitioner thinking, multi-agency child protection practice, policy and law - in the UK, and internationally. Both key child protection policy and practice reviews commissioned by the UK Government 2008-13, the Laming report (2009) and the Munro Review of Child Protection (2011), drew on this research.
Language impairment is common in children and, without effective early intervention, presents a risk for educational, social and emotional disturbance. The team used their considerable experience in child language research to produce a novel assessment of child language development — The Reynell Developmental Language Scales 3 (RDLS3) — the first test of child language development to be standardised in the U.K. to assess both understanding and production of language in young children. Use of the test has improved language assessment and diagnosis for children with language impairments and, as such, has enabled health and education professionals to target therapy and monitor the effectiveness of interventions to improve child language development. The test has been widely used in the U.K. and is now also being used widely internationally. Recent research by the team has informed a revised version of the test — The New Reynell Developmental Language Scales (NRDLS). Sales figures of the RDLS3 and the commissioning of the NRDLS are testament to its outstanding reputation and enduring legacy worldwide.
Every year over 5,000 children and young people in the UK die. Previous research suggests that 20-25% of these deaths may be preventable, and in comparison to many other European countries, the UK has higher child mortality rates. Child Death Review processes, introduced in the USA in the 1970s have been proposed as a means of learning from child deaths and driving prevention initiatives. Prior to 2008, the UK had no national system for reviewing and learning from children's deaths.
From 2006 to 2007, a team from Warwick Medical School led by Dr Peter Sidebotham undertook government-funded research examining a number of Local Authorities across England who had set up pilot Child Death Overview Panels (CDOPs). The findings from this research were instrumental in developing national policy and procedures for child death reviews. The Warwick research emphasised the importance of a multi-agency approach to reviewing all child deaths, with a strong public health focus on learning lessons for prevention, and robust systems for notification and gathering information. This, together with other research by Warwick Medical School on fatal child maltreatment published between 2009 and 2011, has contributed to updated national policy and interagency practice to safeguard children.
Although it is too early in the process to demonstrate any impact on the ultimate goal of reducing preventable child deaths, CDOPs have now been established in every Local Authority in England, as well as an all-Wales panel, and current work in Scotland and Northern Ireland is considering how best to implement such reviews. These panels are reviewing all child deaths in England, resulting in local prevention initiatives, and national returns enabling a clearer picture of the nature of preventable child deaths.
The Scots Words and Place-names Project (SWAP) engaged the public through an embedded use of the internet and social media, a strategy designed both to collect data on the Scots language and to raise the profile of Scots in the wider community. In order to make the project accessible to younger generations, a successful schools competition was run using Glow, the Scottish schools intranet operated by Education Scotland. SWAP also involved two partner organisations from the cultural sector, Scottish Language Dictionaries and the Scottish Place-Name Society, in order to provide a bridge between academic and cultural bodies and the general public.
Clinical, observational and biographical research developed at UEL has produced and supported the novel application of a practice-near methodology adapted to evaluate social work practice and social problems. The benefits of this are described here in the context of two illustrative examples exemplifying the adaptation of the UEL methodology to address self-harm/suicide prevention and the safeguarding of children's rights in London's African communities. Those benefits accrue to practitioners, policy makers, community organisations and individual health and social care service users, and include: the delivery of training leading to positive changes in professional practice; and contributions to discussion, debate and policy and guideline formulation. The research has also been used to enhance public awareness of important social and cultural issues.
Research in Modern Languages linguistics at Exeter explores language use and variation, especially in spoken varieties of French and Italian. It has impacted on educational practices (Impact 1), helped encourage social cohesion (Impact 2), and enhanced public appreciation of language (Impact 3). The French-based research has informed language learning in H.E. and schools, and featured in online training resources and the mass media. It features significantly on University courses (UK and abroad). Aspects of the research have contributed to shaping educational and policy initiatives in Italy and France, aimed at immigrant communities or designed to improve social cohesion.
Research conducted by the Centre for the Study of Poverty and Social Justice (CSPSJ) led to a new way of assessing child poverty in developing countries. This novel method (termed the Bristol Approach) resulted in the United Nations General Assembly's adoption, for the first time, of an international definition of child poverty (2006). It also underpinned UNICEFs Global Study on Child Poverty and Disparities (2008-10), which was run in over 50 countries. In the last ten years, the CSPSJ's work has put child poverty at the centre of international social and public policy debates. Its researchers have advised governments and international agencies on devising anti-poverty strategies and programmes that specifically meet the needs of children, and have significantly influenced the way child poverty is studied around the world. The Centre has developed academic and professional training courses for organisations like UNICEF on the issues of children's rights and child-poverty. Our work has also spurred NGOs such as Save the Children to develop their own child-development indices, and so has had a direct and profound impact on the lives of poor children around the planet.