Choose color scheme

Characterizing human behaviour via Big Data

  • New team member for Dengue Fever research

    me_3 We are pleased to welcome the latest addition to the research team in the form of Abdullahi Shuaibu Adamu.

    Abdullahi is specializing in connectionist learning algorithms will be working on how both public and private sector data can help modelling and prediction of dengue outbreaks in Malaysia. This project has been established in partnership with University Malaya and UNMC, thanks to the award of a Newton Funds grant (British Council).

  • Presentation to the Red Cross, Washington USA

    Today we visited the Red Cross, Washington, to discuss the huge potential for early warning systems of Disaster Management, based not on post-event reporting, but on identifying behavioural pattern changes before the event occurs? An particularly promising example is the detection of impending Urban Flooding in developing nations through observation of changes in mobile money patterns uncovered by machine learning models.

    We’d like to extend our thanks to Aynur Kadihasanoglu for inviting us to speak. We aim in the next steps of this work to pilot a system in Dar es Salaam (see below an example of a physical flood model is visualized on the PARM system).

  • World_Bank2

    Policy Presentation: World Bank, Washington USA

    Thanks to the Learning, Leadership and Innovation Operations Manager Rob Baker and Ardash Desai, Unit Manager, for inviting us to Washington for a Brown Bag Lunch, to demonstrate the Neodem team’s research to the innovation sector of the World Bank. This included demonstrated the PARM.

    logo_WBGPersuant to this, we were also invited to the Global Facility for Disaster Risk and Reduction (GFDRR) by Disaster Risk Management Specialist, Keiko Saito, to demonstrate how the PARM can be used to simulate flood and other natural hazard events.

  • DTM

    Academic Presentation: AMA Winter Conference

    Today Gavin presented some of our work at the American Marketing Association Winter Marketing Academic Conference in Las Vegas, USA.

    The paper, entitled “Temporal Topic Modelling for Big Retail Data(Dr G Smith, Dr J. Goulding, M. Iliffe and Prof. A. Smith) This research details a novel method for segmenting/summarizing consumers based on the automatic identification of a number of interpret able underlying temporal purchasing patterns (consumer missions). More information (and photos of the event) after the link.

  • Partnership with Ipsos

    We are pleased to have started a collaboration with Ipsos in East Africa. Our research collaboration will involve the sharing of omnibus surveys for Tanzania that will provide valuable output futures for training demographic prediction models. For more details of this work see the project page.

  • Parm_Tanzania

    Policy Presentation: World Bank and local council, Dar Es Salaam, Tanzania.

    dar_councilToday, we were honoured to be able to present work being done on Neodemographics project to the Administrative Council of Dar es Salaam in Tanzania. In particular we demonstrated the huge potential to aid policy making and urban planning that analysis of digital footprint data can provide. Many thanks to Andre Bald of the World Bank for providing this opportunity.

    Many parts of East Africa are infrastructurally poor – and lack fundamental national statistics (due to the prohibitive cost of mass censuses and surveys). However, countries such as Tanzania are surprisingly data rich (80% of mobile phone usage occurs in developing countires). This opens up huge potential to bridge the “informational gap” by applying data scientists to datsets held by industry, in a privacy preserving fashion. Please see our urban prediction project for more details.

  • drone_map

    Computer Vision in collaboration with IBM Research, Kenya

    Following the Neodemographic’s project’s support for drone mapping in East Africa, we are now exploring the potential of using Machine Vision to analyse the resulting high-resolution maps. To achieve this we will be collaborating with Nottingham’s Machine Vision Lab (Michel Valstar) and IBM Research, Kenya (Meenal Pore) to predict demographics for the rapidly changing regions of Dar es Salaam via Deep Learning Computer Vision models.

  • photo_indoor

    EMG for Navigation Workshop

    Neodemographics team member Dr. Gavin Smith along with Horizon colleague Dr. James Pinchin, today presented at this workshop exploring the potential of using EMG for navigation and mobility. They took part in discussions on applying the machine learning techniques we use on the project to accelerate research into the use of EMG sensors for navigation when traditional forms (such as GPS) are unavailable or inaccurate.

    Talking to representatives from the Defence Science and Technology Laboratory (dstl), RED Scientific and various departments within the University of Nottingham Gavin and James detailed both the results and technical details behind previously published work on the potential of electromyography to aid personal navigation.

  • James2b

    RCUK Connected Nation – Digital Economy Research Showcase Event

    We are proud that Neodemographics was invited to showcase at the UK’s major EPSRC event “Connected Nation: Thriving in a Digital World”, with Dr. James Goulding presenting our data science research programme. Today’s invitation only event was targeted at businesses and policy stakeholders, featuring an interactive research exhibition as well as themed discussions around the challenges for securing a safe and thriving Connected Nation.

    As well as our presentation, we also exhibited a PARM system showing how we are working with a range of data providers in emerging economies and applying new machine learning techniques to generate otherwise unobtainable knowledge about populations and their dynamics. Photos after the link.

  • pollution

    Academic Presentation: British Thoracic Society

    What impact does the media have on behaviour? In conjunction with researchers from the East Midlands Academic Health Sciences Network and the University of Nottingham Division of Respiratory Medicine our work on the Potential impact of air pollution coverage in the media on respiratory disease admissions was presented at the British Thoracic Society Winter Meeting. A report on these work can be found here.