Seminar by Dr Athen Ma, School of Electronic Engineering and Computer Science, Queen Mary University of London: "Rich-Cores in Networks"

PG10, Talbot Campus, BU, 23/07/2014 14:00

Dr. Athen Ma is an academic in the School of Electronic Engineering and Computer Science (EECS) at Queen Mary, University of London (QMUL). She began EPSRC funded PhD research on the modelling of power-law traffic in packet networks at QMUL in 2000, published a series of journals and completed in 3 years. Dr. Ma was firstly appointed as a Research Assistant in then the Department of Electronic Engineering at QMUL, working on the EU FP5 project ADAMANT (FP5 IST/2001-39117) in 2004, then as a South China Teaching Fellow in the same year and followed by as the Distance Studies Director in 2005. She was responsible for setting up the joint Information Systems Research Centre (ISRC) between Macao Polytechnic Institute (MPI) and QMUL in 2005, and helped the Centre to secure over £1 million research funding from the Macao government. She managed the overall operation of the centre, supervised research and was the line manager to 3 research staff during her time with the ISRC until 2011. Her active involvement in promoting research to the local community has led to a research exhibit at the Macao Science Centre. Her unique academic career path has allowed her to gain a diverse experience in research, working in interdisciplinary environments, collaborative research partnerships and project management. Dr. Ma became a Senior Lecturer at QMUL in 2012.

Abstract: A core comprises of a group of central and densely connected nodes which governs the overall behaviour of a network, and it is recognised as one of the key meso-scale structures in complex networks. Profiling this meso-scale structure currently relies on a limited number of methods which are often complex and parameter dependent or require a null model, and as a result, scalability issues are likely to arise when dealing with very large networks together with the need for subjective adjustment of parameters. The notion of a rich-club describes nodes which are essentially the hub of a network, as they play a dominating role in structural and functional properties. The definition of a rich-club naturally emphasises high degree nodes and divides a network into two subgroups. Here, we develop a method to characterise a rich core in networks by theoretically coupling the underlying principle of a rich-club with the escape time of a random walker. The method is fast, scalable to large networks and completely parameter free. In particular, we show that the evolution of the core in World Trade and C. elegans networks correspond to responses to historical events and key stages in the physical development respectively.

INFER Workshops on Data Science

Data as a utility and analytics as a service

Workshop chaired by Prof. Bogdan Gabrys, Data Science Institute, Bournemouth University

Executive Business Centre, Lansdowne Campus, Bournemouth University, Bournemouth, UK
9 June, 2014 

We are currently experiencing an incredible, explosive growth in digital content and information. According to IDC, there currently exists over 2.7 zetabytes of data. It is estimated that the digital universe in 2020 will be 50 times as big as in 2010 and that from now until 2020 it will double every two years. Research in traditionally qualitative disciplines is fundamentally changing due to the availability of such vast amounts of data. In fact, data-intensive computing has been named as the fourth paradigm of scientific discovery and is expected to be key in unifying the theoretical, experimental and simulation based approaches to science. The commercial world has also been transformed by a focus on BIG DATA with companies competing on analytics. Data has become a commodity and in recent years has been referred to as the ‘new oil’. We are entering a new era of predictive analytics and data intensive computing which has been recognised worldwide with various high profile reports highlighting the challenges and attempting to quantify its huge potential benefits.

This EPSRC IT as a Utility Network+ and EU INFER co-sponsored event organised as part of the Bournemouth University's Festival of Learning will explore the value of very quickly growing data and feasibility of providing data and predictive analytics as services in various industries, public sector and academic disciplines.

The workshop will feature five invited 30 minutes talks to set up the scene for:
i) looking at the growing value of data and treating it as a utility; and
ii) feasibility of providing data and predictive analytics as a service on a large scale and across many industries and disciplines.

The talks will be followed by breakout interactive/discussion sessions in mixed groups with potential linking of partners for various follow on activities (grant applications, proof of concept projects etc.).

The attendance is free and if you are interested to join us please register following this link: analytics-as-a-service/.


Confirmed invited speakers:

Prof. Nello Cristianini, Prof. of Artificial Intelligence, University of Bristol, UK
Prof. Detlef Nauck, Chief Research Scientist, BT's Research and Innovation Division, UK
Tom Quay, Director, We Are Base Ltd, UK
Prof. Trevor Martin, Prof. of Artificial Intelligence, University of Bristol, UK
Dr. Dymitr Ruta, Chief Researcher, EBTIC, Khalifa University, UAE



12.00 - 12.45 - Registration and buffet lunch.

12.45 - 13.00 – Welcome and introduction (Bogdan Gabrys, Bournemouth University, UK)

13.00 - 13.30 – Prof. Detlef Nauck (BT, UK)

13.30 – 14.00 –Prof. Nello Cristianini (Bristol University, UK)

14.00 – 14.30 – Tom Quay (We Are Base Ltd, UK)

14.30 – 15.00 – Coffee break

15.00 – 15.30 – Prof. Trevor Martin (Bristol University/BT, UK)

15.30 – 16.00 – Dr Dymitr Ruta (EBTIC, Khalifa University, UAE)

16.00 – 16.15 – Break

16.15 – 17.15 – Breakout discussion sessions: i) data as a utility; ii) analytics as a service.

17.15 – 18.00 - Summary, recommendations and follow on actions.


Data scientist: The sexiest job of the 21st century?

Workshop chaired by Prof. Bogdan Gabrys, Data Science Institute, Bournemouth University

Executive Business Centre, Lansdowne Campus, Bournemouth University, Bournemouth, UK
10 June, 2014 

UK Government has identified Data Science as the 'transforming and growth driving force across all sectors of economy' and named Big Data as one of the ‘eight great technologies’. With an unprecedented growth in digital content and data, as the digital universe in 2020 is estimated to be 50 times as big as in 2010, we have entered a new era of predictive analytics and data intensive computing. Data scientists are expected to play a key role in this data revolution and their job has even been referred to as "the sexiest job of the 21st century". This EU INFER sponsored one-day open workshop will combine talks by eminent speakers, a panel-audience discussion, exhibition of projects and provide a chance to meet data science experts from academia and industry.

Please register at: scientist-the-sexiest-job-of-the-21st-century/ and join us during this exciting event.


9.00 - 9.15 – Welcome and introduction

9.15 - 10.15 – Prof. Nello Cristianini (Bristol University, UK), ThinkBIG : The Impact of Big Data on Science and Society

10.15 – 10.30 – Break

10.30 – 11.30 – Prof. David van Dyk (Imperial College London, UK), Big Data and Complex Modeling Challenges in Astronomy and Solar Physics

11.30 – 14.30 – Lunch combined with networking, exhibitions, poster session and hands on experimenting.

14.30 – 15.45 – Panel discussion: Is Data Science “the transforming and growth driving force across all sectors of economy”? Is a Data Scientist the “sexiest job of the 21st century”? (Panelists to include the keynote speakers and a number of users and experts from academia as well as public and private sectors)

15.45 – 16.00 – Break

16.00 – 17.00 – Prof. Detlef Nauck (BT, UK), Predictive Analytics and Big Data

17.00 – 17.15 – Break

17.15 – 18.00 - Prof. Bogdan Gabrys (Bournemouth University, UK), Data Science at BU

Seminar by Dr Dimitris Pinotsis, Welcome Trust Centre for Neuroimaging, UCL "Electrophysiological Data and the Biophysical Modelling of Local Cortical Circuits"

P302 LT, Talbot Campus, BU, 12/03/2014 14:00

Dr Pintosis obtained his PhD in September 2006 from the Department of Applied Mathematics and Theoretical Physics (DAMTP) of the University of Cambridge. After an EPRSC Research Fellowship and  lectureship in Reading University he moved to UCL where he is working at the Welcome Trust Centre for Neuroimaging ; having secured funding from EPSRC and the Wellcome Trust.

Dr Pinotsis has a strong track record and a number of landmark publications in imaging neuroscience modelling; he is also the author of the most advanced versions of the state-of the art models for neuroimaging data, the dynamic causal models. I am familiar with Dimitris work and I very strongly encourage the attendance to researchers both in machine learning and in cognitive psychology.

The title of his exciting talk is “Electrophysiological Data and the Biophysical Modelling of Local Cortical Circuits”. “Dynamic Causal Modelling (DCM) is a general framework that allows for a formal (Bayesian) analysis of the properties of neuronal populations, based upon realistic biophysical models. In the past few years, a wide variety of such models has been implemented in the DCM framework. In this talk, I will first review some of these recent advances and then focus on models that allow one to infer spatial parameters of cortical infrastructures generating electrophysiological signals (like the extent of lateral connections and the intrinsic conduction speed of signal propagation on the cortex). I will try to highlight the links between different models and address how the experimental hypothesis or question asked might inform the choice of an appropriate model”.

Seminar by Dr Michael Head, University College London: "The funding of infectious disease research – data, databases and making it all mean something"

PG10, Talbot Campus, BU, 6/03/2014 15:00

Infectious diseases cause significant burden of disease both in the UK and globally. Research into these diseases is vital in order to further our understanding of them, and to aid the implementation of measures to prevent or treat infections. There has not previously been a systematic approach to analysing how funding monies are spent in this area of research. We created the Research Investments in Global Health (ResIn) study and obtained data from all the major public and charitable funding sources for infection-related research awarded to UK institutions for the period 1997-2010. We manually read each study and abstract (if provided) and assigned each study to a number of disease categories (e.g. HIV, tuberculosis, respiratory infections, antimicrobial resistance), as well as the type of science (e.g. laboratory studies, clinical trials) and several other areas.

We identified 6165 funded studies, with a total research investment of UK£2.6 billion. By disease, HIV received £461 million (17.7%), malaria £346 million (13.3%), tuberculosis £149 million (5.7%), influenza £80 million (3.1%), and hepatitis C £60 million (2.3%). We compared funding with disease burden (disability adjusted life years, DALYs, and mortality) to show where there may be low levels of investment relative to burden e.g. diarrhoeal infections (£254 million, 9.7%).  Further steps that we’d like to pursue include expansion from the UK to a global analysis that will allow more in-depth analysis of areas that should be prioritised in the future, and we are seeking funding to do that.

In the meantime, in order to make maximum use of our data, in collaboration with colleagues at Bournemouth University, we intend to create an online open-access database that will allow funders, policymakers and researchers to search and download the customised sections of the funding data, as well as presenting graphs and infographics as requested by the user.

Seminar by Dr Michael Bourne, the Head of Sports Science and Medicine for the England and Wales Cricket Board: "Sports Analytics"

P403, Talbot Campus, BU, 22/01/2014 15:00

Dr Michael Bourne is the Head of Sports Science and Medicine for the England and Wales Cricket Board. He has spent 15 years in the sports industry in analytical fields of sports science. In his career Michael has provided analytical sports science support to World and Olympic champions, Ashes winners and many others who are simply curious about performance. His PhD thesis focussed on how humans utilise biological motion as information for anticipation purposes in sport.

The title of his exciting talk will be “Sports Analytics”


Abstract: Evidence-based practice is a growing aspect of elite sports performance, supplementing the traditional experiential and intuitive approaches that have pervaded sport since its inception. Sports performance analysis specifically has been growing steadily as a discipline since the 80’s into one of the most widely used sports sciences in modern day professional and Olympic sports. Sports performance analysts have traditionally been sports scientists by training utilising video and statistical feedback to support athlete development. However in 2003 the publication of the popular book ‘Moneyball’ changed the landscape across the performance sports industry. The story of a low-budget baseball team who taps into the skills of a Yale economist to revolutionise how baseball players are valued showed the wider world what few in the traditional sports community had identified previously. Firstly, it opened sports coaches and executives eyes to a new view of their sport through the lens of big data. Second, it made bright young things with data analysis capabilities from a range of disciplines realise there was a more exciting world away from the business and financial sectors that they could apply their skills to. The world of sports analytics has never been the same since...

In this talk i journey through the current landscape of sports analytics in the UK, outlining some of the working practices that take place in my own sport of Cricket as well as reflecting on work undertaken in football, rugby and Olympic sports. The talk will give an insight into the differences between traditional and non-traditional sports analysts and will also seek to demonstrate how the skills of data scientists in particular, from big data analysis to data visualisation fit into the current demands of the sports performance industry.

Seminar by Mr Manuel Martinez-Salvador: "Artificial Intelligence for Automating Data Analysis"

PG22, Talbot Campus, BU, 27/11/2013 14:00

Our speaker is Mr Manuel Martinez-Salvador, advanced PhD student in Bournemouth University.

Abstract: The requirements for analysing big volumes of data have increased over the last few decades. The process of selecting, cleaning, modelling and interpreting data is called the Knowledge Discovery in Databases (KDD) process. The decision of how to approach each step in this process has often been made manually by experts. However, experts cannot be aware of all methods, nor is it feasible to try all of them. Researchers have proposed different approaches for automating, or at least advising, the different stages of the KDD process. This talk will outline the different types of Intelligent Discovery Assistants and point out some future directions.

Seminar by Professor Ben Azvine, the Global Head of Security Research and Innovation at BT: "Industrial applications of novel Intelligent Systems"

PG16, Talbot Campus, BU, 11/11/2013 15:00

Professor Azvine holds a BSc in Mechanical Engineering, an MSc in Control Engineering, a PhD in Intelligent Control Systems all from Manchester University and an MBA from Imperial College, London. Having held research fellowship and lectureship posts in several universities, he joined British Telecom Research in 1995 and set up a research programme to develop and exploit intelligent systems technologies within BT.   

Since then he has held senior, principal and chief research scientist as well as head of research centre posts at Adastral Park, the head quarter of BT R&D. Ben has edited several books and published more than 100 scientific articles. He is an inventor on 50 patents, has won two BCS gold medals, and the IET award for innovation in IT, holds visiting professorships at Universities of Bristol, Cranfield and Bournemouth in the UK.

Abstract: Intelligent systems play an important role in industry for managing customer relationship, providing business intelligence, helping organisations analyse their data and protecting organisations against Cyber-attacks. In this talk I'll present a number of case studies within BT where we have used intelligent system originating from within our research organisation and successfully downstreamed them into our operations.

I strongly encourage academics and PhD students not to miss the opportunity to attend to the seminar and to discuss potential collaborations.

Seminar by Dr Jérôme Kunegis: "Observing the Web: The Koblenz Network Collection"

PG 22, Talbot Campus, BU, 23/10/2013 14:00

Dr. Jérôme Kunegis is a postdoctoral researcher at the Institute for Web Science and Technologies (WeST) at the University of Koblenz–Landau, Germany, and currently a visiting postdoctoral research at the University of Cambridge's Computer Lab. He is the lead of the focus group on network analysis at the WeST Institute, as well as of the Koblenz Network collection project (KONECT), a large-scale effort to assemble network datasets for use in web science, network science and other disciplines. Dr. Kunegis has written over 30 peer-reviewed papers in the areas of web science, network analysis, data mining, machine learning and information retrieval. He graduated at the Berlin Institute of Technology (TU Berlin) in 2006, and received his PhD at the University of Koblenz–Landau in 2011 for his work on the spectral evolution model under the supervision of Prof. Steffen Staab.

Abstract: This talk presents the Koblenz Network Collection (KONECT), a project to collect network datasets in the areas of web science, network science and related areas, as well as provide tools for their analysis.  In the cited areas, a surprisingly large number of very heterogeneous data can be modeled as networks and consequently, a unified representation of networks can be used to gain insight into many kinds of problems. Due to the emergence of the World Wide Web in the last decades many such datasets are now openly available.  The KONECT project thus has the goal of collecting many diverse network datasets from the Web, and providing a way for their systematic study.  The main parts of KONECT are (1) a collection of over 160 network datasets, consisting of directed, undirected, unipartite, bipartite, weighted, unweighted, signed and temporal networks collected from the Web, (2) a Matlab toolbox for network analysis and (3) a website giving a compact overview the various computed statistics and plots.  In this talk, I describe KONECT's taxonomy of networks datasets, give an overview of the datasets included, review the supported statistics and plots, and briefly discuss KONECT's role in the area of web science and network science.

Seminar by Dr George Oikonomou: "Mobile Forensics"

PG22, Talbot Campus, BU, 9/10/2013 16:00

Dr Oikonomou expertise is in the development of advanced forensic tools, IPv6-enabled Wireless Sensor Networks and the Internet of Things

Abstract: This talk will focuses on the outcomes of recent research in the field of forensics for mobile devices, with emphasis on the Android operating system. We discuss a method for the acquisition of forensic-grade evidence from smartphones using open source tools, focusing specifically on Bluetooth and Wi-Fi facilities.

Forensic analysis reveals traces left in the inner structure of three mobile Android devices and also brings out security vulnerabilities. Subsequently, we investigate attacks against android's pattern lock mechanism, such as methods to recover the pattern using the oily residues left on screens when people move their fingers to reproduce the unlock code.

We present a pilot study on user habits when setting a pattern lock and on their perceptions regarding what constitutes a secure pattern. Based on the survey's results, we establish a scheme, which combines a behaviour-based attack and a physical attack on graphical lock screen methods, aiming to make it partially or fully retrievable. This work has been supported by the European Union's Prevention of and Fight against Crime Programme "Illegal Use of Internet" - ISEC 2010 Action Grants, grant ref. HOME/2010/ISEC/AG/INT-002.

INFER Workshop on Data Mining in Process Industry

Open Workshop on Data Mining in Process Industry

held as a part of the annual INFER project meeting

Marl, Germany
16 September, 2013

Following previous INFER annual meetings this workshop is organised around a theme of particular interest to Evonik. The workshop features presentation of the developed soft sensors as well as the current version of the INFER software platform.


12:30 – 13:00       Lunch

13:00 – 13:15       Introduction to the open workshop on “DataMining in Process Industry” by Monika Berendsen

13:15 – 13:45       “Gaussian Process-based Adaptive Soft Sensor for Large Scale Data” by Ali Abusnina

13:45 – 14:45       “An Adaptive Soft Sensor for Acrylic Acid Production – A Case Study” by Bogdan Gabrys

14:45 – 15:00       Break

15:00 – 16:30       “INFER Software Platform Live Demonstration” by Marcin Stoinski and Mateusz Wojciechowski

16:30 – 17:30       Discussion - "The role and need for INFER like software in Process Industry"

18:30                    Guided tour of „Maschinenhalle der Zeche Fürst Leopold“ in Dorsten

Syndicate content