Inference, Dynamics and Interaction Group
The Inference, Dynamics and Interaction group brings together three fundamental research areas: modern inference techniques, dynamic systems and control theory and interaction design. These are applied in wide range of situations:
- Health, Wellness & Entertainment
- Systems Biology
- Mobile Interaction
- Cognitive Neuroscience/neuroimaging
- Social Interaction
The group's strength lies in the unusual combination of theoretical backgrounds from machine learning to HCI, and the focus on building innovative working systems which achieve performance previously thought impossible, using the latest algorithms, sensors and devices. The group's skills in combining software engineering and mathematical inference allows us to attack complex systems problems with large high-dimensional data spaces and so in real-time.
Visit the dedicated Inference, Dynamics and Interaction Group Website to find out more.
- European Network of Excellence called Social Signal Processing Network (SSPNet), coordinated by Alessandro Vinciarelli, together with with Maja Pantic (Imperial College). The network aims at establishing a European research community on modeling analysis and synthesis of social signals.
- TOBI: Tools for Brain-Computer Interaction, EC-funded project. Roderick Murray-Smith (Glasgow PI), John Williamson, project coordinator:Prof. José del R. Millán, 2008-2012.
- Multimodal, Negotiated Interaction in Mobile Scenarios, EPSRC funded project (£638k), Roderick Murray-Smith (PI), with Matt Jones (Swansea), Stephen Brewster, 2007-2010.
- EC-COST action IC0601 on Sonic Interaction Design.
- Social Interaction: A Cognitive-Neurosciences Approach, ESRC funded project (£3.7 million) , Simon Garrod (PI), 2008-2012.
- PASCAL network member, EC-funded network in Pattern Analysis, Statistical Modelling and Computational Learning.
Academic Staff: Prof. Roderick Murray-Smith, Dr. Maurizio Filippone, Dr. Simon Rogers, Dr. Alessandro Vinciarelli, Dr. John Williamson
Researchers: Mohammad Bin Md Noor, Dr. Andrew Crossan (TOBI), Rónán Daly, Andrew Ramsay (TOBI), Dari Trendafilov (Nokia/GU), Melissa Quek (TOBI), Dominik Gotojuch, Zac Mtema (joint supervision with Katie Hampson, Epidemiology), Lauren Norrie, Shimin Feng, Daryl Weir, Rebecca Mancy (jointly supervised with Pat Prosser), Edwin Thuma (jointly supervised with Iadh Ounis, IR), Mikhail Churakov (main supervisor is Rowland Kao, Vet School), Hugues Salamin, Joe Wandy
- Machine Learning
- Statistical Pattern Recognition
- Human Computer Interaction
- Mobile HCI
- Brain Computer Interaction
Send me bubbles: multimodal performance and social acceptability
Williamson, J.R.
Designing performative interactions in public spaces
Williamson, J.R.
Exploiting query logs and field-based models to address term mismatch in an HIV/AIDS FAQ retrieval system
Thuma, E., Rogers, S.
Bayesian approaches for mass spectrometry-based metabolomics
Rogers, S.
Statistical methods and models for bridging omics data levels
Rogers, S.
ODE parameter inference using adaptive gradient matching with Gaussian processes
Dondelinger, F., Filippone, M.
Evaluating bad query abandonment in an iterative SMS-Based FAQ retrieval system
Thuma, E., Rogers, S.
Investigating the disagreement between clinicians’ ratings of patients in ICUs
Rogers, S.
Discrete and continuous time simulations of spatial ecological processes predict different final population sizes and interspecific competition outcomes
Mancy, R.
Longitudinal analytics on web archive data: Its about time!
Weikum, G., Ntarmos, N., Spaniol, M., Triantafillou, P.
Focused and casual interactions: allowing users to vary their level of engagement
Pohl, H., and Murray-Smith, R.
Multimodal mobile interactions: usability studies in real world settings
Williamson, J.R.
A performative perspective on UX
Williamson, J.R.
Continuous auditory and tactile interaction design
Visell, Y., Murray-Smith, R.
Automatic role recognition in multiparty conversations: An approach based on turn organization, prosody, and conditional random fields
Salamin, H.
Bridging the gap between social animal and unsocial machine: A Survey of social signal processing
Vinciarelli, A.
Conversation analysis at work: detection of conflict in competitive discussions through semi-automatic turn-organization analysis
Pesarin, A., Cristani, M., Murino, V., and Vinciarelli, A.
Automatic personality perception: Prediction of trait attribution based on prosodic features
Mohammadi, G., and Vinciarelli, A.
Designing future BCIs: Beyond the bit rate
Quek, M.
Touching the Micron: Tactile Interactions with an Optical Tweezer
Lamont, S., Bowman, R., Williamson, J.
This Week’s EventsAll Upcoming EventsPast Events
This Week’s Events
There are no events scheduled for this week
Upcoming Events
Interdependence and Predictability of Human Mobility and Social Interactions
Inference, Dynamics and Interaction Group
Speaker: Mirco Musolesi
Date: 23 May, 2013
Time: 14:00 - 15:00
Location: Sir Alwyn Williams Building, 422 Seminar Room
The study of the interdependence of human movement and social ties of individuals is one of the most interesting research areas in computational social science. Previous studies have shown that human movement is predictable to a certain extent at different geographic scales. One of the open problems is how to improve the prediction exploiting additional available information. In particular, one of the key questions is how to characterise and exploit the correlation between movements of friends and acquaintances to increase the accuracy of the forecasting algorithms.
In this talk I will discuss the results of our analysis of the Nokia Mobile Data Challenge dataset showing that, by means of multivariate nonlinear predictors, it is possible to exploit mobility data of friends in order to improve user movement forecasting. This can be seen as a process of discovering correlation patterns in networks of linked social and geographic data. I will also show how mutual information can be used to quantify this correlation; I will demonstrate how to use this quantity to select individuals with correlated mobility patterns in order to improve movement prediction. Finally, I will show how the exploitation of data related to friends improves dramatically the prediction with respect to the case of information of people that do not have social ties with the user.
Past Events
IDI Seminar (29 November, 2012)
Speaker: Konstantinos Georgatzis
Visualisation of multivariate data sets is often done by mapping data onto a low-dimensional display with nonlinear dimensionality reduction (NLDR) methods. We have introduced a formalism where NLDR for visualisation is treated as an information retrieval task, and a novel NLDR method called the Neighbor Retrieval
Visualiser (NeRV) which outperforms previous methods. The remaining concern is that NeRV has quadratic computational complexity with respect to the number of data. We introduce an efficient learning algorithm for NeRV where relationships between data are approximated through mixture modeling, yielding efficient computation with near-linear computational complexity with respect to the number of data. The method is much faster to optimise as the number of data grows, and it maintains good visualisation performance.
Evaluating Bad Query Abandonment in an Iterative SMS-Based FAQ Retrieval System (14 February, 2013)
Speaker: Edwin Thuma
We investigate how many iterations users are willing to tolerate in an iterative Frequently Asked Question (FAQ) system that provides information on HIV/AIDS. This is part of work in progress that aims to develop an automated Frequently Asked Question system that can be used to provide answers on HIV/AIDS related queries to users in Botswana. Our system engages the user in the question answering process by following an iterative interaction approach in order to avoid giving inappropriate answers to the user. Our findings provide us with an indication of how long users are willing to engage with the system. We subsequently use this to develop a novel evaluation metric to use in future developments of the system. As an additional finding, we show that the previous search experience of the users has a significant effect on their future behaviour.
Pre-interaction Identification By Dynamic Grip Classification (28 February, 2013)
Speaker: Faizuddin Mohd Noor
We present a novel authentication method to identify users at they pick up a mobile device. We use a combination of back-of-device capacitive sensing and accelerometer measurements to perform classification, and obtain increased performance compared to previous accelerometer-only approaches. Our initial results suggest that users can be reliably identified during the pick-up movement before interaction commences.
Flexible models for high-dimensional probability distributions (04 April, 2013)
Speaker: Iain Murray
Statistical modelling often involves representing high-dimensional probability distributions. The textbook baseline methods, such as mixture models (non-parametric Bayesian or not), often don’t use data efficiently. Whereas the machine learning literature has proposed methods, such as Gaussian process density models and undirected neural network models, that are often too computationally expensive to use. Using a few case-studies, I will argue for increased use of flexible autoregressive models as a strong baseline for general use.
