× ATTENTION! You are currently viewing information about Predictive Analytics Worls 2015! Return to this year’s page..

Agenda

Predictive Analytics World London
etc.venues, 155 Bishopsgate, 28-29 October, 2015


See all Session Descriptions

Day 1 - Wednesday - 28 October, 2015

Session descriptions day 1

8.00 am

Registration

9.15 am
Room: London

Welcome & Opening

9.20 am
Room: London

Opening Featured Session

Analytics for TV Audiences – A Case Study on Combining Descriptive, Predictive and Prescriptive Analytics

Television impacts our daily lives. It provides news, entertainment and education. And, with minute-by-minute information on TV ratings and multi-billion-euro income streams from advertising, it is a real Big Data decision problem that requires sophisticated Analytics. TV viewers must be predicted for each 30 second timeslot, normally days in advance, in order to schedule the right advertisement. However, viewership patterns vary by demographic of target audiences, from young adults to elderly couples and the affluent to the unemployed, across geo-regions from rainy North to sunny South, with TV programme schedules and weather driving viewership patterns over time. Without analytical methods, comply decisions quickly become inefficient. We present a case-study from a leading private UK TV channel where we employed analytics to support decision making.

Attendees will learn:

  • How a time series approach helped to make sense of terabytes of data
  • How to explore viewer behaviour using time series clustering in Descriptive Analytics
  • How to forecast future viewers using k-nearest neighbours in Predictive Analytics
  • How to optimise advertising scheduling across changing variance in Prescriptive Analytics

Speaker:

Prof. Dr. Sven Crone, Director, Lancaster Research Centre for Forecasting

10.00 am
Room: London

Sponsored Session

How Analytics is Adapting in a Rapidly Changing World

Advanced analysis techniques that mine answers to our greatest challenges have been adapting due to both technological innovation and business leaders and not just statisticians becoming data driven.

Today we live in a world typified with an abundance of available data, but a scarcity of data scientists to unlock accurate and reliable results. Organisations that harness the power of IoT, big data and real-time analytics create new avenues to innovation and huge advantage.

Dell Software will be sharing key innovations in the field of advanced analytics that address:
- How to adapt to highly diverse data environments.
- How to make analytics a tool to be used by all members of your organisation.
- How to share models from data scientists globally that drive mankind’s discoveries, wealth and well-being.

Speaker:

Nuno Antonio, Pre and Post Sales Senior Manager, Dell

10.30 am

Coffee Break

10.55 am
Room: London

Keynote:

What Happened to Data Warehousing? Challenges & Opportunities in the Rapidly changing era of Predictive Analytics

With a fundamental change in the assumptions underpinning a structured data world dominated by relational databases, we are entering the age of BigData. The combination of economic drivers in enterprise computing, the need to leverage semi-structured and unstructured Data, and the emergence of the Internet of Things (IOT), a dramatic shift in the Data landscape is taking place. The advent of Hadoop and the Open Source stack in this space have accelerated the changes to a point of confusion. Today’s data analyst faces a bewildering environment of technologies and challenges involving semi-structured and unstructured data with access methodologies that have almost no relation to the past. This talk will cover issues and challenges in how to make the benefits of advanced analytics fit within the application environment. The requirement for Real-time data streaming and in situ data mining is stronger than ever. We demonstrate how many of the critical problems remain open with much opportunity for innovative solutions to play a huge enabling role. This opportunity makes Data Science and several related fields critical to almost all future analytical tasks. The talk will include 3 case studies to show the challenges and the great opportunities for BigData and Predictive Analytics.
Moderator:

Prof. Dr. Sven Crone, Director, Lancaster Research Centre for Forecasting

Speaker:

Usama Fayyad, Chief Data Officer, Barclays (@usamaf)

11.40 am

Session Change for Combo Pass Holders

11.45 am
Room: London

Analytics in the Supply Chain – Forecasting, Cross-Selling and Pattern Recognition at Shell

The Shell Lubricants Supply Chain (LSC) Analytics Team was formed with a global mandate to deliver value. Alex presents a series of case studies from the work of the LSC Analytics Team over the past two-years.

· Forecast Improvement: Forecast Accuracy is a Key Performance Indicator (KPI) for the supply chain. Accurately predicting future demand allow the supply chain to consistently supply customers without holding large amounts of stock to buffer unexpected demand. Improving the KPI was a significant change journey for Shell LSC.

· Customer Cross-Sell and Up-Sell: Working with a number of other Shell Analytics Teams the LSC Analytics Teams tacked the challenge of making appropriate cross-sell and up-sell recommendations based on customer sales – as well as how to cluster these customers.

· Pattern Recognition in Stock Holding: Tackling the issue of errors in stock management the team looked at computer pattern recognition within the time series. Detecting human errors as well as improving the time-to-opportunity ratio were key value drivers.

Moderator:

Prof. Dr. Sven Crone, Director, Lancaster Research Centre for Forecasting

Speaker:

Alex Hancock, Head of Treasury Analytics, Shell Oil Company

12.30 pm

Lunch Break

1.40 pm
Room: London

My Three Predictive Analytics Pet Peeves

Predictive Analytics (PA) has become increasingly mature as a technical discipline over the past decade in part because it stands on the shoulders of the related disciplines of data mining and machine learning. However, there are recurring themes that permeate discussion boards and conferences that have become my personal pet peeves. This talk examines three of them and why they matter to practitioners, including why we must have humility in how far data science and algorithms can take us, and the value of business objectives, measuring "success," and measuring "significance."
Moderator:

Prof. Dr. Sven Crone, Director, Lancaster Research Centre for Forecasting

Speaker:

Dean Abbott, Co-Founder and Chief Data Scientist, SmarterHQ (@deanabb)

Predictive Customer Interaction Beyond Attribution - Keep your friends Close but your Enemies Closer

The usual approach to improve targeted customer interaction is to analyze the converting part (e.g. a sale) of the past online traffic. Smart attribution is hugely important to evaluate the cost effectiveness of marketing campaigns and to optimize the media spending and budget mix. But attribution is per definition only focusing in the converting traffic and purely retrospective. Information of the major traffic share from none or better 'not yet' converting visitors is very much neglected. There are multiple communication/ interaction levels in the sales process and different marketing actions have different strength to capture certain level customers and transfer them from level to level. From transforming site visitor data into customer journeys of users to predicting the future behavior of these users and their next steps with time scales - this case study based on Dutch Online travel company BookIt.com will demonstrate how you can learn much more from the non-converts and use these insights in the predictions.
Moderator:

Prof. Dr. Sven Crone, Director, Lancaster Research Centre for Forecasting

Speaker:

Werner Vorstmann, Finance Director, Bookit.com (@WernerVorstman)

Alwin Haensel, Business Analytics & Founder, Haensel AMS

2.40 pm

Session Change for Combo Pass Holders

2.45 pm
Room: London

Keynote

Top Five Technical Tricks to Try when Trapped

There's no better source for tricks of the analytics trade than Dr. John Elder, the established industry leader renowned as an acclaimed training workshop instructor and author -- and well-known for his "Top 10 Data Mining Mistakes" and advanced methods like Target Shuffling. In this special plenary session, Dr. Elder, who is the CEO & Founder of Elder Research, North America's largest pure play consultancy in predictive analytics, will cover his Top Five methods for boosting your practice beyond barriers and gaining stronger results.
Moderator:

Prof. Dr. Sven Crone, Director, Lancaster Research Centre for Forecasting

Speaker:

Dr. John Elder, CEO & Founder, Elder Research, Inc. (@johnelder4)

3.30 pm

Coffee Break

3.55 pm
Room: London

Guerrilla Analytics: 7 Principles for Delivering Predictive Analytics in Dynamic Fast-Paced Projects

Delivering Data Science is difficult. Data changes, is updated and replaced. Business rules change as understanding evolves. Requirements change. In this environment, the team must evaluate many models and algorithms, each of which can be tuned in many ways. This can be overwhelming. Guerrilla Analytics is a methodology for managing agile Data Science teams. Its 7 principles promote team efficiency, flexibility and agility when building predictive models in dynamic environments. This talk will draw on almost 10 years of experience spanning data mining pre-sales, forensic analytics and academic research to illustrate Guerrilla Analytics in action. You will learn about the 7 principles with illustrated examples of their application on real projects. You will also learn about the key operational capabilities your teams need to increase agility with Guerrilla Analytics.
Moderator:

Prof. Dr. Sven Crone, Director, Lancaster Research Centre for Forecasting

Speaker:

Enda Ridge, Head of Algorithms, Sainsbury's (@enda_ridge)

4.40 pm

Session Change for Combo Pass Holders

4.45 pm
Room: London

Web Analytics to Optimize Keyword Selection and Pay-Per-Click Pricing for Online Retailing - A Collaboration between Summit and Argos

To automate and optimize daily PPC bid-price setting over portfolios which can contain over 400,000 products and keywords is a challenge. The solution has significant budget impact for big retailers such as Argos. In this session you will see a case study where Hedley will show how the methodology developed delivered a 37% improvement in ROI for Argos overall. You will see how a new methods allow tight control over budgets and cost-of-sale, how forecasts showing optimized digital spend and expected revenues for future periods are constructed, and how this allows Argos to optimize budgets over product categories in order to maximise expected profit.
Moderator:

Prof. Dr. Sven Crone, Director, Lancaster Research Centre for Forecasting

Speaker:

Hedley Aylott, CEO, Summit

Prof. David Wooff, Mathematical Sciences and Statistics and Mathematics Consultancy Unit, Durham University

Predictive Analytics for Personal Banking - Enhancing Credit Scoring with Transactional Customer Data and Genetic Algorithms

Lloyds Banking Group has over 20 million Personal Current Account (PCA) customers and vast amounts of data available for use in credit decisions. In the past there has been a reliance on ‘Expert Judgement’ to build characteristics which are used in modelling. As both the quantity and quality of data grows it is imperative to research new modelling techniques that effectively use all data to ensure the Bank makes the best decisions for our customers. Currently Lloyds uses a traditional logistic regression scorecard for PCA overdraft application decisions. These scorecards use characteristics derived from a number of data sources, however they are largely based on expert judgement and do not use the full extent of the data to build models, as a result the predictive power of the data is not fully exploited by these models.As both the quantity and quality of data has grown there are a number of sources of data that are underused, but easily accessible for making PCA credit decisions. One example is the Transactional data, this contains information on monetary transaction on accounts such as transfers, over the counter and point of sales. This data contains thousands of variables over a hierarchy that is completely unsuitable to feed raw into the traditional statistical methods.Here we present two approaches using evolutionary algorithms to mine these previously unwieldy quantities of data to find which characteristics add most incremental power to the model. One based on Genetic Algorithms creates a library of characteristics based on the raw variables before selecting a subset for use in logistic regression models. The second method uses Genetic Programming to evolve the characteristics from the raw data in a method that can be used both to create new characteristics and as a stand-alone predictive model.The predictive power of both is compared to the traditional scorecard to show the improvement both in terms of discriminatory power and business benefit.
Moderator:

Prof. Dr. Sven Crone, Director, Lancaster Research Centre for Forecasting

Speaker:

Richard Davis, Head of Analytics, Lloyds Banking Group

5.45 pm

Networking Reception in the Exhibit Hall

7.00 pm

End of First Conference Day

7.00 pm

Dinner with strangers

Dinner with strangers:
meet your fellow attendees.
See the registration desk for more information


Day 2 - Thursday - 29 October, 2015

Session descriptions day 2

8.30 am

Registration

9.30 am
Room: London

Developing Predictive Web Analytics at iProspect - Statistical Techniques or Machine Learning?

Instead of relying on the typical gut-feel, basic descriptive plots in Excel, or Google Analytics and Google Adwords descriptive analytics, iProspect Greece set out on a project to develop a rigorous predictive/prescriptive analytics solution utilizing traditional statistical techniques, including Regression analysis with automatic model selection based on the quality of fit and ANOVA (Analysis of variance with multiple comparisons and post-hoc hypothesis testing) as well as novel approaches including non-parametric, non-linear regression analysis using machine learning algorithms, all implemented in Python and R. The implemented analytics solution, although at its early stages, has already delivered a series of benefits, including an increase in the accuracy of KPI prediction between 40% and 80% for certain cases, huge reductions in the time required to analyse a campaign – from days to hours per analyst, an almost immediate response time to a client wanting to understand how their digital campaigns are performing and why. A pilot study for Allianz Direct achieved a 90% predictive accuracy on Google Adwords campaigns, driving conversions and subsequently finetuning their marketing strategy and budget, increasing ROI by 25%. In this session Michael Georgakopoulos will present how the implemented solution has since been applied across multiple major accounts in Insurance, FMCGs and Retail.
Moderator:

Prof. Dr. Sven Crone, Director, Lancaster Research Centre for Forecasting

Speaker:

Michael Georgakopoulos, Management Consultant & Analytic Solutions Architect

Costas Mantziaris, iProspect (@costasM)

Spotting Unusual Behavior in Internet Traffic Data: A Case Study of Predictive Analytics Usage at uSwitch

Successful Internet commerce sites have tens of thousands of unique daily visitors, creating large quantities of noisy traffic data. In the midst of this data are insights about customer behaviour and journeys that can help form commercial strategies, marketing decisions and data driven decision making, however deriving these insights can be a complex process. Sophisticated statistical predictive analytics techniques are required to indicate when there are subtle changes in the traffic data which can be acted upon. Using time series, textual analysis and statistical techniques, changes in customer profile traffic data can be isolated to give a daily or weekly detailed snapshot of the underlying customer behaviour prompting these changes.
Moderator:

Prof. Dr. Sven Crone, Director, Lancaster Research Centre for Forecasting

Speaker:

Abigail Lebrecht, Principal Analyst, uSwitch.com (@A_Lebrecht)

10.30 am

Coffee Break

10.55 am
Room: London

Keynote

Leveraging Big Data With Predictive Analytics

In a world of big data, organizations and consultants argue rightfully that more data is almost always better than less, and that successful organizations must find ways to collect and process big data to become more competitive. But is big data enough? If we collect enough data, do we naturally build better predictive models? Do we still need domain experts or should the data speak for us? If big data generates more hype than return on investment, what does it do beyond force us to spend ever increasing amounts of money to store let alone process the data. Fortunately, there are key ways big data can be leveraged using predictive analytics. More data means we can ask different questions from the data than we ever could with smaller data. It means we can be more sure that the patterns we are finding are stable and true. Mr. Abbott will describe approaches to leveraging big data with predictive analytics by selecting the right hardware infrastructure, software infrastructure, and business objectives. Examples will be provided based on recent advances in retail customer intelligence and cloud computing.
Moderator:

Prof. Dr. Sven Crone, Director, Lancaster Research Centre for Forecasting

Speaker:

Dean Abbott, Co-Founder and Chief Data Scientist, SmarterHQ (@deanabb)

11.40 am

Session Change for Combo Pass Holders

11.45 am
Room: London

How to Surf a Data Lake without Drowning? Leveraging the Data Landscape at GfK for Better Predictive Analytics

The data landscape of a global market research company like GfK is incredibly valuable and challenging at the same time. But the more data GfK collects or observes, and the more heterogeneous this data is, the more challenging it has become to keep transparency about data assets and to make the data easily accessible. so GfK decided to transfer the data into a global Data Lake – an integrated repository for all kinds of structured and unstructured data coupled with a sophisticated Big Data Analysis platform. In this presentation Ralph Wirth and Anna Machens will illustrate based on real client questions: (1) how integrated predictive modeling – i.e., the development of predictive models using information from different data sources as predictors – can lead to significantly improved model performance, (2) how exactly the challenge of efficiently leveraging a data basis which seems overwhelming at first sight can be approached, and (3) how the approach of developing predictive models changes significantly if Data Scientists have the possibility to utilize the flexibility of Data Lake architectures and Big Data tools. Real-life case studies that will be presented including predictive models for leading coffee brands, which make use of information from numerous different data sources, such as past sales of coffee (Consumer Panels), sales of coffee machines (from retail panels), social media, brand tracker and advertising tracker KPIs, as well as external data, such as weather, regional holiday density, or external shocks. Furthermore Ralph and Anna will give insights into the toolbox that they identified as most useful.
Moderator:

Prof. Dr. Sven Crone, Director, Lancaster Research Centre for Forecasting

Speaker:

Dr. Ralph Wirth, Head of Data Lab, GfK

Dr. Anna Machens, Data Scientist GfK Data Lab, GfK (@annamachens)

12.30 pm

Lunch Break

1.40 pm
Room: London

How Valuable is Your Hotel? A Case Study of Using Predictive Analytics at Expedia

The mission of Expedia is to revolutionize travel through the power of technology. The existing team based in Geneva serves a critical role in directing the organization’s efforts to improve hotel supply breadth and quality for the entire Expedia Inc. One of Expedia's goals is to acquire hotels that will produce. Sandro will show how Expedia uses Predictive Analytics algorithms to prioritize acquisitions. Contracted hotel data is leveraged and a model to predict the dollar value of hotels around the World is build. Regression problems are solved by using ensemble learning, through the Gradient Boosting Machine algorithm. Implemented in R, the prediction is based on data such as room capacity, star rating, latitude/longitude and TripAdvisor reviews. Sandro will describe the overall project, including data collection/processing, model tuning and current challenges.
Moderator:

Prof. Dr. Sven Crone, Director, Lancaster Research Centre for Forecasting

Speaker:

Sandro Saitta, Data Scientist, Expedia Inc. (@DataMiningBlog)

AFRAID: An Active Inference Approach for Fraud Detection Using Advanced Social Network Analysis

The Belgian Social Security Institution is a federal agency that registers and monitors every active company in Belgium, and is responsible for the collection of employer and employee tax contributions. The taxes are levied on employer level, making this process highly sensitive to fraud. One of the main analytical techniques used to spread fraud through a network, are the so-called collective inference (CI) techniques. Here, one assumes that the label of a node in the network depends on the label of the neighboring nodes. Although CI is proven very useful, it can be easily mislead. A wrongful estimate of one node’s label (e.g., a true legitimate node is classified as fraudulent) might further impact the network and misclassify all other nodes in the neighborhood. In this presentation, we will discuss how (1) active inference – a subdomain of active learning for data analytics – is able to improve CI algorithms by selecting a set of uncertain nodes in the network to be labeled by inspectors, and (2) given that fraud networks are highly adaptive and evolve over time, we temporally integrate an inspector’s decision, knowing that this decision does not necessarily hold in the near future.
Moderator:

Prof. Dr. Sven Crone, Director, Lancaster Research Centre for Forecasting

Speaker:

Veronique Van Vlasselaer, PhD researcher, KULeuven (@Veronique_VV)

2.40 pm

Session Change for Combo Pass Holders

2.45 pm
Room: London

Removing the Barriers to Rapid Analytics in R&D at GlaxoSmithKline – The Inside Story of an Analytics Partnership

The increasingly data-driven challenges of modern drug discovery require advanced computational and analytical techniques. However, the breadth and depth of skills required, and the unevenness in the demand for them, makes the overhead of full in-house provision impractical and uneconomic. This is the inside story of how GSK and Tessella, working together, are exploiting the flexibility and speed of a managed specialist analytics services model to meet this challenge. Examples illustrate the breadth of skills ranging from unsupervised machine learning and bayesian statistics through to signal processing and radar tracking that are needed to address novel analytics problems facing computational chemistry, biology, bioelectronics and advanced manufacturing. This ability to rapidly initiate multiple small and mid-size analytics initiatives is of immense practical importance in R&D, not a nice to have.
Moderator:

Prof. Dr. Sven Crone, Director, Lancaster Research Centre for Forecasting

Speaker:

Nick Clarke, Head of Analytics, Tessella (@Analytics_Lab)

Stephen Pickett, Computational Chemistry, GlaxoSmithKline

3.30 pm

Coffee Break

3.55 pm
Room: London

Featured Session

The Soul in the Machine: Anomaly Detection meets Image Mining in Predictive Maintenance

Anomaly detection has utility in many industries, and its potential to save costs and risk is huge when applied to maintenance prediction. But widespread sharing of techniques for predicting major episodes without a classic pattern of historical data has been limited by the lack of public data. In this paper, we use new, publicly available IOT data from motor sensors along with a variety of techniques to create predictive models. We also explore the effectiveness of Image Mining to further enhance our models. A roadmap for use on ANY type of sensor or IOT data that can be fast Fourier transformed is also discussed. As always, all examples are available and based on public data and employ a modern, open source (and free) platform so that everyone can benefit from our work.
Moderator:

Prof. Dr. Sven Crone, Director, Lancaster Research Centre for Forecasting

Speaker:

Phil Winters, Senior Managing Partner, CIAgenda (@CIAgenda)

4.40 pm

Session Change for Combo Pass Holders

4.45 pm
Room: London

Closing Note and Feedback

Prof. Dr. Sven Crone, Director, Lancaster Research Centre for Forecasting

5.30 pm

End of Conference

See the Predictive Analytics World London 2014 agenda here


Workshops – Friday, 30 October, 2015

Available on-demand

Online Workshop:

Predictive Analytics Applied
An Online Introduction

Instructor:

Eric Siegel, Ph.D. , Conference Founder
Predictive Analytics World