Full Agenda – London, UK – 27-28 November, 2012

Day 1: Tuesday, 27th November, 2012

8:00am-9:00am

Registration & Coffee

[ Top of this page ] [ Agenda overview ]


9:00am-9:50am

Opening Session
Five ways to increase the impact of Predictive Analytics

This opening session will first and foremost serve as an introduction and 'Travel Guide' to Predictive Analytics World London 2012. Undoubtedly (in Gartners Hype Cycle terminology) Predictive Analytics has reached the plateau of productivity. On this conference alone, a distinct selection of managers and experts will explain their advances in the domain. No longer can this productivity be interpreted as a series of isolated success stories. As Predictive Analytics is maturing, the focus shifts from single successes to adoption of analytics as a recurrent business activity, wide-scale, multi-purpose usage, and also management of advanced analytics teams and competences.
The impact of Predictive Analytics on business, government policy and society is undoubtedly growing. Besides an introduction to the program, in this session, we also offer five concrete tips and tricks to further increase the impact of Predictive Analytics in business applications (including examples on customer acquisition and customer retention in utilities, telecommunications, financial services, etc).

Speaker: Geert Verstraeten, PhD, Program Chair Predictive Analytics World London 2012

[ Top of this page ] [ Agenda overview ]


9:50am-10:00am

Sponsored Session by Deloitte Sponsored Session by Deloitte

Speaker: Tom Scampion, Partner, Deloitte

[ Top of this page ] [ Agenda overview ]


10:00am-10:40am

Break / Exhibits

[ Top of this page ] [ Agenda overview ]


10:40am-11:25am

Special Featured Session: Government Analytics
Case Study: Danish ministry of Taxation
Implementing advanced analytics in Tax collection and compliance

In 2006, the Danish ministry of taxation initiated a large IT project where all types of arrears to the public sector – from unpaid train tickets to VAT – were to be collected through one automated system. The system is planned to be finished in 2013. To be able to segment the debtors a scoring model is needed. It shows the expected future payment behavior of the debtors. We have just finished the work and are able to present the model in detail, covering the scorecards, explanatory variables, the definition of Bad and the scores.

Speaker: Mads Nielsen, Special Adviser and Project Leader, Danish ministry of Taxation

[ Top of this page ] [ Agenda overview ]


11:25am-12:05pm

Case Study: MAN Diesel & Turbo
Challenges of Industrial B2B Analytics: Case Study of a Maritime Industry Manufacturer

Adoption of analytics varies among different industries. In certain industries (mostly of the B2C nature) such as insurance, personal finance, telco and FMCG, a set of analytical practices are often applied. However, looking at the B2B sector, examples are more scarce. Nevertheless, analytics can be applied to B2B with the same success as B2C. This talk will look into a set of organic differences between B2C and B2B businesses that could influence usage of analytics and will exemplify with a case of an analytical journey with one of the leading maritime equipment manufacturers of the world.

Speaker: Massih Mayeli, Business Analyst, MAN Diesel & Turbo

[ Top of this page ] [ Agenda overview ]


12:05pm-1:05pm

Lunch / Exhibits

[ Top of this page ] [ Agenda overview ]


1:05pm-1:55pm

Thought Leader
Keynote
Big “Time Series” Data – What Forecasting can learn from Predictive Analytics (and vice versa)

With more data being gathered at ever shorter time intervals, forecasting of “big” intraday time series data has become a reality for utilities companies, call center operators, TV advertisers etc. As time series datasets converge in size to those regularly faced by data miners, can forecasters learn from best practices, use algorithms and tools under development in predictive analytics? On the other hand, with consumer behaviour - and consequently data - changing more rapidly as well, can analytics learn from forecasters, who have traditionally looked to model and predict change over time?
Using the example of intraday forecasting of hourly UK electricity demand, we show that the state-of-the-art statistical forecasting methods are often equally “black” boxes, which can no longer be interpreted. As a result, forecasters are free to utilise data driven algorithms from predictive analytics, such as neural networks or decision trees, as well as novel concepts of time series clustering and time series classification, for a simpler and more accurate time series prediction than statistics can offer.

Speaker: Prof. Dr. Sven F. Crone, Deputy Director, Lancaster Research Centre for Forecasting

[ Top of this page ] [ Agenda overview ]


1:55pm-2:35pm

Case Study: Dŵr Cymru Welsh Water
One in a million. Using Applied Analytics to reduce rare pollution events in Wales.

How do you stop asset failures when there is only a one in a million chance they will fail? This was the problem facing Dŵr Cymru Welsh Water when they asked Trinity Horne to analyse pollution risk on their sewerage network. Trinity Horne used Applied Analytics, an approach with a proven record of reducing failures on distribution networks by using client expertise and far-reaching data collection. Using decision trees they identified high risk sewers in some surprising places. They also employed Bayesian updating to change probabilities in real-time as the high risk sewers were investigated.

Speaker: Alec Ross, Senior Statistician, Trinity Horne

Speaker: Donal O’Hehir, Analytics Practise Leader, Trinity Horne

[ Top of this page ] [ Agenda overview ]


2:35pm-3:05pm

Break / Exhibits

[ Top of this page ] [ Agenda overview ]


3:05pm-3:55pm

Special Featured Session: Managing Analytics
Managing Analytics: Developing and growing teams of Predictive Analysts

Building and developing an in-house predictive analytics department within a company is an amazing but challenging multidisciplinary project. Based on 12 years experience in different companies (in B2B and B2C; in Telco, Banking and other industries), the session will cover case studies and anecdotes from different predictive analytics journeys. This talk addresses tips and tricks to answer key questions in managing analytics: how to position and sell a new predictive analytics activity within a company; how to build team of predictive analytics experts; how to develop team members (competence model) and motivate them. What are the real keys to success?

Speaker: Martine George, PhD, Head of Marketing Analytics and Research, BNP Paribas Fortis

[ Top of this page ] [ Agenda overview ]


3:55pm-4:10pm

Sponsored Session by JMP Sponsored Session by JMP
Humpty Dumpty and Predictive Analytics

'When I use a word,' Humpty Dumpty said, in rather a scornful tone, 'it means just what I choose it to mean - neither more nor less.’ And so it is with the idea of ‘Statistical Discovery’. This brief overview shows why, as a practitioner of Predictive Analytics, you should care about Statistical Discovery. It also shows how you can find out more to start to make use of it.

Speaker: Ian Cox, Senior Director Marketing JMP, SAS

Sponsored Session by MuSigma Sponsored Session by Mu Sigma
Mu Sigma – A Brief Introduction

An elevator pitch on Mu Sigma

Speaker: Aditi Nitin, Regional Head, UK, Mu Sigma Inc.

[ Top of this page ] [ Agenda overview ]


4:10pm-5:00pm

Multiple Case Studies:
Commerzbank, Lloyds, Ursus Advisor
Optimization in Financial Services

Financial institutions are improving business performance by using optimization techniques in additional to behavioral and statistical modeling. Hear how retail banks are using prescriptive along with predictive analytics to get more customers, get them more profitably, price them appropriately, and retain them strategically.
We will cover the barriers to adoption and introduce an organizational readiness checklist. Attendees should leave ready to kick off the projects internally to implement optimization.

Speaker: Frank H Bria, Principal, Bria Analytics Group

[ Top of this page ] [ Agenda overview ]


5:00pm-6:00pm

Networking Reception

[ Top of this page ] [ Agenda overview ]


Day 2: Wednesday 28th November, 2012

8:00am-9:00am

Registration & Coffee

[ Top of this page ] [ Agenda overview ]


9:00am-10:10am

Thought Leader
Keynote
Becoming an Ace with a Robot as your Wingman

Humans and computers have strengths that are more complementary than alike – to the point where a sophisticated algorithm may be the best "2nd person" to put on a complex task. Yet, our and computer analytic weaknesses are surprisingly severe. To explore how to improve the man/machine partnership, we compare and contrast natural and artificial intelligence, with special attention to the growing realization of how challenging it is to think truly rationally.

Speaker: John Elder, PhD, CEO and Founder Elder Research, Inc.

[ Top of this page ] [ Agenda overview ]


10:10am-10:20am

Sponsored Session by SAP Sponsored Session by SAP
Predictive Analytics from SAP

With data volumes exploding, organizations need to extract the maximum possible value, whether from transactional data typically generated within the enterprise or from unstructured data created by external sources like social media. The introduction of in-memory technology has dramatically reduced the time and cost of data processing and makes it possible to perform predictive analysis against vast volumes of data in real time. An integrated set of analytics solutions from SAP has been designed from the ground up to enable you to fully exploit the new reality – from database to decision. Our solutions enable you to combine the power of predictive processing using the SAP® HANA platform with intuitive modeling and advanced data visualization provided by SAP Predictive Analysis software.

Speaker: Philip Mugglestone, Director, Global Solutions, SAP

[ Top of this page ] [ Agenda overview ]


10:20am-10:50am

Break / Exhibits

[ Top of this page ] [ Agenda overview ]


10:50am-11:35am

Special Featured Session: Sports Analytics:
Case Study: Leicester Tigers Rugby team
The Evolution of Analytics in Rugby Union

The use of analytics in sport has grown exponentially over the last 20 years, and rugby union is no exception. Since 1995 when the game turned professional alongside the introduction of a salary cap, it has become necessity for coaches and club owners alike to look after their playing assets. The ability to collect, manage, analyse and report useful data should be of the utmost importance for any business looking to positively influence performance. This presentation will demonstrate the growth of analytics in sport in general and subsequently focus on our use of predictive analytics with the aim of helping us understand and reduce players' injury rates at Leicester Tigers.

Speaker: Andy Shelton, Head of Sports Science Leicester Tigers Rugby team

[ Top of this page ] [ Agenda overview ]


11:35am-12:05pm

Measuring data mining myopia – the value analysts lose by relying on their favorite tools

In many applications, data mining is deployed with the concrete goal of finding objective patterns – serving either as an enhancement or a replacement of human intuition. However, a contradiction exists in the fact that many data miners themselves rely on intuition and affinity in their choice of algorithms used. Even the most advanced analysts are naturally biased towards their favorite methodologies – with unavoidable impact on final performance. In this session, based on a sample case study and past data mining experiences, we illustrate how problem-specific details (e.g., industry, sample size, number of independent variables, type of target variables, etc…) may help in the selection of the most appropriate methodology, which paves the way for automation of the design of a data-mining solution and ultimately leads to commercialization of data mining technology.

Speaker: Amjad Zaim, PhD, CEO, Cognitro Analytics

[ Top of this page ] [ Agenda overview ]


12:05pm-1:05pm

Lunch / Exhibits

[ Top of this page ] [ Agenda overview ]


1:05pm-1:35pm

Case Study: Deutsche Bank (Belgian Branch)
Conversion of savers to investors at Deutsche Bank Europe GmbH Belgian branch: Lessons learned

This talk goes into more detail on the results & lessons learned from the development of one of the first predictive models at Deutsche Bank Europe GmbH Belgian branch.
Both quantitative (modeling) aspects & results as well as qualitative (explaining results to management & internal clients) topics are discussed, keeping in mind that the focus in today’s business environment is on achieving workable results (that can easily be understood) in a short period of time.

Speaker: Matthias Meul, Customer Intelligence analyst Deutsche Bank Europe GmbH Belgian branch

[ Top of this page ] [ Agenda overview ]


1:35pm-2:15pm

Case Study: Medway Youth Trust
Predictive Analytics Helps Medway Trust to Reduce Youth Unemployment

Medway Youth Trust are working to identify education, training and employment opportunities for young people to reduce unemployment in communities. Applying predictive analytics software to its operations, Medway Youth Trust can now better identify and reach out to high-risk young people who are unemployed or need education and training to prepare them for future opportunities. Delivers a 250 percent improvement in accuracy of identification compared to manual search techniques. Provides a model that is 83.2 percent accurate against historical data – suggesting a high degree of validity for real-world predictions. Delivers a 250 percent improvement in accuracy of identification compared to manual search techniques.

Speaker: Graham Clewes, Chief Executive Medway Youth Trust

[ Top of this page ] [ Agenda overview ]


2:15pm-3:00pm

Panel Discussion
Management Panel
The War for Analytical Talent

In 2009, Google's chief economist Hal Varian envisioned that the most sexy job in the 2010s would be statisticians. Due to the increasing availability of data, the true bottleneck for analytical-driven business success would lie in the availability of analytical talent. This panel discussion will address the following questions:
 
• Has the war for analytical talent really started?
• How to attract and retain analytical experts?
• Best practices in managing analytical expertise
• (How) will automation change the role of the analytical expert?
• etc

Panelist: John Elder, PhD, CEO and Founder Elder Research, Inc.

Panelist: Martine George, PhD, Head of Marketing Analytics and Research, BNP Paribas Fortis

Panelist: Tom Khabaza, Data Miner and Founding Chairman of the Society of Data Miners

Panel moderator: Geert Verstraeten, PhD, Program Chair Predictive Analytics World London 2012

[ Top of this page ] [ Agenda overview ]


3:00pm-3:30pm

Break / Exhibits

[ Top of this page ] [ Agenda overview ]


3:30pm-4:10pm

Social Media Intelligence:
Text, Network Mining and Predictive Analytics combined

This paper combines text mining and network mining revealing new heterogeneous insights in social media that are not detectable using either technique alone. Sentiment analysis from online forum posts together with reference structures from the quotation network allows the detection of individual’s sentiment as well as their relative influence in the underlying discussion forum. Additionally, machine learning techniques are applied to identify relevant clusters.
Originally performed for Deutsche Telekom and The Economist, the techniques and methods presented here use public data to demonstrate the approach for social media. A conclusion is drawn about the relevance and practicalities of this new approach along with a recommendation for next steps.

Speaker: Phil Winters, Senior Managing Partner, CIAgenda

[ Top of this page ] [ Agenda overview ]


4:10pm-4:40pm

Case Study: MarketPsy Long-Short Fund LP
Predicting Investor Behavior: Trading the First Social Media Sentiment-Based Hedge Fund

MarketPsy Capital LLC utilized psychological and macro-eocnomic factors extracted from social and news media to manage the world's first social media-based hedge fund. From September 2, 2008 to the end of 2010 the MarketPsy Long-Short Fund LP outperformed the S&P500 by 27% with 1/3 of the market volatility by trading individual U.S. stocks in a market-neutral portfolio, but closed after losing 8% in 2010. In the spirit of a medical "Morbidity and Mortality Report" fund director Richard L. Peterson, M.D. will discuss the trials and tribulations of our predictive modeling and trading.

Speaker: Aleksander Fafula, PhD, Chief Data Scientist MarketPsy Capital LLC

[ Top of this page ] [ Agenda overview ]


4:40pm-5:00pm

Closing Remarks

Speaker: Geert Verstraeten, PhD, Program Chair Predictive Analytics World London 2012

[ Top of this page ] [ Agenda overview ]

Share |

Sponsors

deloitte SAP
JMP Mu Sigma
StatSoft Rapid-I KNIME.com AG
Ofvigo
ECMOD
AnalyticbridgeAnalyticBridge Customer Think

improvement & innovation Institute of Data Miners KDnuggets Top Seos Visibility Magazine TECHWORLD BAQMaR Information Age

Check out the official PAW Blog Partners
© 2014 Predictive Analytics World | Privacy

Program by: Python Predictions
Python Logo

Produced by Prediction Impact, Inc. and Rising Media, Ltd.

Predictive Analytics Company           Predictive Analytics Event Producer