How to do practical Data Science?

Size: px
Start display at page:

Download "How to do practical Data Science?"

Transcription

1 How to do practical Data Science? Artur Suchwałko, QuantUp code::dive Conference 2018, , Wroclaw

2 Introduction

3 Introduction Several real-world Data Science projects + business results, methods, and findings: prediction of fraudulent insurance claims for an insurance company (with Machine Learning, generic) customer churn prediction for a bank (with Machine Learning, generic) prediction of pre-paid customers registration for a telecom (with Deep Learning, specific but important) Wrapping up and structuring these conclusions + adding recommendations regarding executing Data Sciences projects. (Traditional (and boring) businesses where AI/ML can give tons of money.)

4 Data frame Duration Amount Age Gender Car loan? Installment loan? Target (example: German credit dataset, 1994)

5 Detection of insurance frauds (ML)

6 Business case There was a car crash An insurance claim is reported It could be a fraud attempt We need to decide in a limited time: accept or decline the claim We have only some information Usually the claims are analysed by people

7 Goal & results Client: Insurance company Goal: Improving of detection of probable frauds for further investigation Probable / doubtful / suspicious claim: suspected to be a fraud but not proven to be one Finding and proving are two different things Result: Savings of order of 30% (comparing to past simple models)

8 Important business questions How to choose claims for investigation to: detect highest number of fraud attempts? detect highest amount of fraudulent claims? detect highest amount with limited resources and time for detection? be able to prove highest number / amount of fraud attempts? This requirement is translated into a suitable goal function for a model and should affect the optimization criterion.

9 How to build a model? Preparation of the predictors (can be complex because of aggregation of data from many sources) Having the target variable in the historical data Build a predictive model

10 Important Checking if modeling is possible (the process influences the historical data) Definition of new predictors Detection of false predictors Data enhancement: historical aggregates, textual, external Technicalities Using PCs

11 Claim case 1: Rules / human Description: A driver hit the rear side of a victim s car. The car was pushed to the crossroads area and there was a collision with a third car (Mercedes). The police was called. Rules: airbags inflated similar age of both drivers difference of cars age >=11 years historical loss coefficient >=5 Result: Not refused to pay because of fraud attempt: the description was consistent with the damages

12 Claim case 2: Model Description: I (victim) was driving a left lane. The second driver (a culprit) was driving a right lane (the same direction). He wanted to change the lane, haven t seen my car and hit my car. Its rear left side damaged my car s right front side. Analysis: no clear evidence only one year of cars age difference no age information for the second driver insurance policy was not new no claim history for drivers and cars Result: Refused to pay because of fraud attempt: no correlation between description and damages not possible to be a real claim

13 Example: Sample 119 top scored claims given for experts to verify if they are frauds 82 were confirmed as doubtful nearly 70% percent of positive verification

14 Pure analytics vs. business AUC: boosted trees: 0.85, log. regr.: 0.77 Figure 1: An example of simple / complex model difference in business / analytical terms

15 Inside Boosted trees with non-standard goal function Reduction of number of predictors

16 Findings & conclusions Really complex process of handling claims and detecting doubtful ones Uncertainty if the prediction is possible (sample representativeness) Great results An excellent ROI: 2 months including cost of third party software

17 Churn prediction (ML)

18 Business case Retaining existing customers is cheaper and easier than acquiring new ones Being able to predict churn means being able to prevent or reduce it It is predictable

19 Goal & results Client: Bank Goal: The best churn prediction for accounts (stated as) Result: Improvement of 10% (comparing to simpler models) Improving model actionability

20 Important The best churn prediction Reduction of churn? Activities Target Finding the optimal strategy vs. prediction of behavior assuming today s strategy (as in bad debt collection models) Explaining reasons of churn strategy Contract (termination) / no contract (reduced activity) Target definition Definition of predictors Money fransfers (+ descriptions) Stopping salary transfering: almost false predictor Avoid aggregates (especially long-term) Actionability Effect: prediction effectiveness (known) + action effectiveness (unknown) They are not splittable Theoretical effectiveness vs. actionability (improving models in a lab doesn t help)

21 Inside Boosted trees with suitable modifications and optimization Careful hyperparameter optimization Laptops, desktops

22 Findings & conclusions Definition of predictors and target is really hard if you want to have a good and actionable model We destroyed predictive power of a model by using correct predictors but this improved actionability

23 Prediction of pre-paid customers registration (DL)

24 Business case Pre-paid customers of a telecom Anonymous SIM card owners: no registration obligation Anti-terrorism law regulation (heavy historical usage for some of SIM cards!) Obligatory registration before Otherwise loosing phone number

25 Goal & results Client: Telecom Goal: Prediction of number of pre-paid clients who will register themselves after the registration obligation is implemented (February, 2017): Result: knowing budget for registration prediction of traffic in points of sales prediction of cash flow from pre-paids after the change Correct forecast (error of 3%) Well planned activities and savings Hard to give exact amounts or percentages

26 Important Registration obligation implemented for the first time: no historical data Proxy variable for registration: how to use the historical data? Just one number an expert method? Desktop PCs + GPUs

27 Inside I We tried a whole bunch of methods Just guess the percentage (not that stupid as it seems!) Vintage analysis (cohorts of clients) One-class classification (geometrical) One-class classification (SVM) Density estimation Anomaly detection methods (distance based) Deep Learning / representation learning (recurrent neural network) Deep learning turned out to be the best method

28 Inside II Assumptions possible to be validated after implementation of the registration obligation: How to define an active user? What does it mean that users are similar? Who does need to use a phone? Does does using need imply a registration? What is a threshold probability? A proxy measure for registration: making a pre-payment! 400 features x millions of clients

29 Results

30 Findings & conclusions Surprisingly good forecast There was always a magical parameter to set or a missing link Instead of guessing the parameter value we could directly guess the registration percentage because the parameter was directly tied to it in a complex and unknown way Guessing or not guessing the percentage is a matter of our comfort It doesn t influence forecasting quality Using wrong method could result in a really big error It was crucial to find a righ proxy variable

31 Recommendations

32 Data-driven decisions It is crucial to take data-driven decisions, regardless we use a model or not Business people often tell that: using AI requires only putting any data inside improving business decisions by models is impossible Don t trust both these groups!

33 Setting business goal + translating business into analytics Clearly state and work on the business goal Translate it really carefully into analytics not loosing much (usually you loose something) Usually custom goal function helps us a lot

34 Process & data The present process decides if the modeling is doable or not Is it possible to build a model if all the cases are currently analysed by people? Prepare a big set of potential predictors Need to understand both, process and data Carefully check the data quality Beware of false predictors (leaks from future)

35 Causing and measuring business influence Ensure that the model can be and will be used for decision making Actionability What are possible actions? Example: churn predictors Time to action for churn: not too soon Clear success criteria (not the best model possible!) Compare to no model

36 Project execution Resources people + skills (not just using the tools!) time (!) software hardware Management (CRISP?) Process of modeling Project risk management

37 Modeling tools Adequately complex What is an improvement by a better model? What are risks and costs associated? How to implement the solution? Use open source

38 There are no fully automatic tools! When you have the business goal translated into analytics then that s easy Mistakes in analytics itself (overfit, suboptimal model) are relatively easy to avoid The most painful mistakes are besides of it The automatic solutions simplify the analytics only

39 Implementation & keeping it working Planned implementation + planned tests / checks Cold running (without taking actions) Running on samples Initial validation + ongoing monitoring

40 Summary

41 Summary I (Deep Learning) Figure 3:

42 Summary II (practical modeling) Figure 4:

43 Contact

44 QuantUp We improve data-driven business decisions applying Data Science, Artificial Intelligence / Machine Learning We serve the following industries: banking, insurance, collection agencies, telecoms, retail, bio/med.

45 QuantUp We deliver end-to-end ML solutions Problem analysis, Studies on model impact on business processes, Solution definition & development, Maintenance services, Support in continuous model development, if required, Trainings & knowledge sharing

46 Contact Artur Suchwałko, Ph.D., CEO e: artur [at] quantup [dot] pl m: w: quantup.pl

The usage of Big Data mechanisms and Artificial Intelligence Methods in modern Omnichannel marketing and sales

The usage of Big Data mechanisms and Artificial Intelligence Methods in modern Omnichannel marketing and sales The usage of Big Data mechanisms and Artificial Intelligence Methods in modern Omnichannel marketing and sales Today's IT service providers offer a large set of tools supporting sales and marketing activities

More information

LENDING PEER-TO-PEER LENDING. Helping you to understand both borrowers and lenders

LENDING PEER-TO-PEER LENDING. Helping you to understand both borrowers and lenders PEER-TO-PEER LENDING Helping you to understand both borrowers and lenders WHEN YOU TRULY UNDERSTAND YOUR EXISTING CUSTOMERS, WE CAN HELP YOU FIND MORE OF THEM THE INSIGHT YOU NEED TO GROW YOUR BUSINESS

More information

US state auditor deploys machine learning to tackle healthcare fraud at scale

US state auditor deploys machine learning to tackle healthcare fraud at scale US state auditor deploys machine learning to tackle healthcare fraud at scale This e-book focuses on how a US state auditor leveraged machine learning to detect and prevent fraud in healthcare. The biggest

More information

Analytics for Banks. September 19, 2017

Analytics for Banks. September 19, 2017 Analytics for Banks September 19, 2017 Outline About AlgoAnalytics Problems we can solve for banks Our experience Technology Page 2 About AlgoAnalytics Analytics Consultancy Work at the intersection of

More information

Investigating the myths and realities of contactless payment

Investigating the myths and realities of contactless payment Investigating the myths and realities of contactless payment Questions & Answers There has been much recent coverage in the media (press, web, blogs.) about the perceived security vulnerabilities of contactless

More information

Nationwide Paym. Terms and Conditions. How will Nationwide use your personal information? Terms and Conditions. Building Society

Nationwide Paym. Terms and Conditions. How will Nationwide use your personal information? Terms and Conditions. Building Society Building Society Nationwide Paym Terms and Conditions How will Nationwide use your personal information? In signing up and agreeing to these terms and conditions, you give your express consent to the use

More information

ALTERNATIVE FINANCE PEER-TO-PEER LENDING. Helping you to understand both borrowers and lenders

ALTERNATIVE FINANCE PEER-TO-PEER LENDING. Helping you to understand both borrowers and lenders PEER-TO-PEER LENDING Helping you to understand both borrowers and lenders WHEN YOU TRULY UNDERSTAND YOUR EXISTING CUSTOMERS, WE CAN HELP YOU FIND MORE OF THEM THE INSIGHT YOU NEED TO GROW YOUR BUSINESS

More information

Fraud Presentation. IIA Regional Meeting. Gareth A.H. Edwards Head of Internal Audit Marston s PLC

Fraud Presentation. IIA Regional Meeting. Gareth A.H. Edwards Head of Internal Audit Marston s PLC Fraud Presentation IIA Regional Meeting Gareth A.H. Edwards Head of Internal Audit Marston s PLC Marston s PLC A bit about the Group FTSE250 ( 1,011.3m FY1617) 1600 pubs 6 breweries 14,500+ employees Contract

More information

Use of Data Mining and Machine. Use of Data Mining and Machine. Learning for Fraud Detection. Learning for Fraud Detection. Welcome!

Use of Data Mining and Machine. Use of Data Mining and Machine. Learning for Fraud Detection. Learning for Fraud Detection. Welcome! Use of Data Mining and Machine Use of Data Mining and Machine Learning for Fraud Detection CPE PIN Code: UKLK Learning for Fraud Detection Welcome! Ramazan Isik, CFE, CIA, CRISC, CRMA Chief Audit Executive

More information

Putting Big Data & Analytics to Work!

Putting Big Data & Analytics to Work! Putting Big Data & Analytics to Work! Prof. dr. Bart Baesens Department of Decision Sciences and Information Management, KU Leuven (Belgium) School of Management, University of Southampton (United Kingdom)

More information

Dear Valued Member, Sincerely, Jerry Jordan President & CEO CGR Credit Union

Dear Valued Member, Sincerely, Jerry Jordan President & CEO CGR Credit Union Dear Valued Member, To further support your financial needs into the future, we will convert our ATM/ debit card program from our current Mastercard ATM/debit card program to a new and improved VISA ATM/debit

More information

WHITE PAPER. Results Delivers Value

WHITE PAPER. Results Delivers Value WHITE PAPER SAS Title Results Delivers Value ii Contents Building on a foundation of analytics... 1 SAS Results: The right tools for the job... 1 Fill the analytical skills gap...2 Address limited IT support...2

More information

FINANCE + DEEP LEARNING SKYMIND / DEEPLEARNING4J 2015

FINANCE + DEEP LEARNING SKYMIND / DEEPLEARNING4J 2015 FINANCE + DEEP LEARNING SKYMIND / DEEPLEARNING4J 2015 WHICH OUTCOMES MATTER? BASIC APPLICATIONS Anomaly Detection for Compliance Rogue Traders, Fat Fingers Fraud, Money Laundering Detection Trading Strategies

More information

Reimburse Expense Reports with Total Confidence Through Real-time Reviews

Reimburse Expense Reports with Total Confidence Through Real-time Reviews Concur Detect Reimburse Expense Reports with Total Confidence Through Real-time Reviews Concur Detect Automatically Checks 100% of Your Expense Reports to Detect Fraud and Compliance Issues in Real-time

More information

G U I D E B O O K P R O G R A M : D A T A A N D A N A L Y T I C S D O C U M E N T N U M B E R : S 180 N O V E M BER TENSORFLOW ON AWS

G U I D E B O O K P R O G R A M : D A T A A N D A N A L Y T I C S D O C U M E N T N U M B E R : S 180 N O V E M BER TENSORFLOW ON AWS 2 0 1 8 G U I D E B O O K P R O G R A M : D A T A A N D A N A L Y T I C S D O C U M E N T N U M B E R : S 180 N O V E M BER 2 0 1 8 TENSORFLOW ON AWS A N A L Y S T Rebecca Wettemann Nucleus Research, Inc.

More information

Cash Flow if you re out of money, you re out of business.

Cash Flow if you re out of money, you re out of business. Cash Flow if you re out of money, you re out of business. Thanks for downloading this Cash Flow Cheat Sheet from SmartBusinessPlans.com.au Cash Flow is probably the most important aspect of keeping a business

More information

Our Membership background. General Offering Breakdown

Our Membership background. General Offering Breakdown Our Membership background In order to maintain low rates for our clients, we are providing our customers with an option to subscribe to our new membership program in lieu of a labor rate increase. We have

More information

Our Membership background. General Offering Breakdown

Our Membership background. General Offering Breakdown Our Membership background In order to maintain low rates for our clients, we are providing our customers with an option to subscribe to our new membership program in lieu of a labor rate increase. We have

More information

HOW TO USE AI IN BUSINESS

HOW TO USE AI IN BUSINESS HOW TO USE AI IN BUSINESS How to use AI in business 3 FOREWORDS 20 6 EXAMPLES OF USE CASES 4 INTRODUCTION TO AI 33 SUMMARY 10 MACHINE LEARNING IN PRACTICE 35 USEFUL LINKS 14 BEST PRACTISES FOREWORDS Artificial

More information

SAP Leonardo. Artur Gornik Cloud and Platform Technology SAP Middle & Eastern Europe

SAP Leonardo. Artur Gornik Cloud and Platform Technology SAP Middle & Eastern Europe SAP Leonardo Artur Gornik Cloud and Platform Technology SAP Middle & Eastern Europe Digitization Every company strives to become a connected business Connected Data Integration Collaboration Mobile Big

More information

Hidden Cost of Fraud How well are State & Local Governments detecting and managing fraud?

Hidden Cost of Fraud How well are State & Local Governments detecting and managing fraud? NASC Annual Conference Atlanta, GA March, 2019 Hidden Cost of Fraud How well are State & Local Governments detecting and managing fraud? Detect accidental expense fraud and compliance issues with Artificial

More information

Quick Start with AI for Businesses

Quick Start with AI for Businesses Quick Start with AI for Businesses ML Conference 2018, Dr. Ulrich Bodenhausen, AI Coach V1.0 2018-06-12 About me PhD in Machine Learning from KIT. Application of neural networks to speech and gesture recognition

More information

By: Aderatis Marketing

By: Aderatis Marketing By: Aderatis Marketing 01803 362 026 enquiries@aderatis.com Google AdWords for Small Businesses: Mistakes to Avoid Not getting much luck from your AdWords campaign and ready to admit defeat? Don t feel

More information

Analytics in India a few trends field notes from a layman. Hindol Basu

Analytics in India a few trends field notes from a layman. Hindol Basu 1 Analytics in India a few trends field notes from a layman Hindol Basu India presents its own unique challenges in leveraging the power of data 2 Focus on quantified impact leading to top and bottom line

More information

Customer Lifecycle Management How Infogix Helps Enterprises Manage Opportunity and Risk throughout the Customer Lifecycle

Customer Lifecycle Management How Infogix Helps Enterprises Manage Opportunity and Risk throughout the Customer Lifecycle Customer Lifecycle Management How Infogix Helps Enterprises Manage Opportunity and Risk throughout the Customer Lifecycle Analytics can be a sustained competitive differentiator for any industry. Embedding

More information

Lease is more. What you need to know about why you should lease your fleet, rather than buy.

Lease is more. What you need to know about why you should lease your fleet, rather than buy. Lease is more. What you need to know about why you should lease your fleet, rather than buy. LeasePlan. Lease is more. Welcome. Every business that s running a fleet wants to do the right thing. It s tempting

More information

Disclaimer Forward-Looking Statements Industry and Market Data Non-GAAP Financial Information

Disclaimer Forward-Looking Statements Industry and Market Data Non-GAAP Financial Information Disclaimer Forward-Looking Statements This presentation includes forward-looking statements within the meaning of the Securities Act of 1933, as amended (the Securities Act ), the Securities Exchange Act

More information

Knowledge Discovery Services and Applications

Knowledge Discovery Services and Applications Knowledge Discovery Services and Applications kdlabs AG www.kdlabs.com Dr. Jörg-Uwe Kietz Content Knowledge Discovery @ kdlabs Key features of Mining Mart for KD services and applications Clever processing

More information

Operations & Technology OPERATIONS & TECHNOLOGY CONFERENCE & EXHIBITION SEPTEMBER 11-13, 2017 THE WESTIN CHARLOTTE CHARLOTTE, NC

Operations & Technology OPERATIONS & TECHNOLOGY CONFERENCE & EXHIBITION SEPTEMBER 11-13, 2017 THE WESTIN CHARLOTTE CHARLOTTE, NC Operations & Technology OPERATIONS & TECHNOLOGY CONFERENCE & EXHIBITION SEPTEMBER 11-13, 2017 THE WESTIN CHARLOTTE CHARLOTTE, NC Conference and Exhibition The Future is Here: Do You Have the Tools Necessary?

More information

Important changes to our Individual Savings Accounts (ISAs) Terms and Conditions

Important changes to our Individual Savings Accounts (ISAs) Terms and Conditions Important changes to our Individual Savings Accounts (ISAs) Terms and Conditions Which take effect on 13 January 2018 What s in this brochure We re letting you know about some important changes we re making

More information

Optimization of infill drilling locations

Optimization of infill drilling locations Digital Technologies INVESTMENT OPPORTUNITY: Optimization of infill drilling locations Executive summary Today, decisions on drilling locations are made by using reservoir models and simulations, but with

More information

How to select the right advertising technology solution.

How to select the right advertising technology solution. How to select the right advertising technology solution. With the broad fragmentation of advertising buying sources and the escalating need to better leverage customer data, more organizations are looking

More information

The role of analytics, big data and machine learning as a driver for digital and artificial intelligence transformation

The role of analytics, big data and machine learning as a driver for digital and artificial intelligence transformation Dr. Mark Nasila Head of Advanced Analytics First National Bank South Africa The role of analytics, big data and machine learning as a driver for digital and artificial intelligence transformation Education

More information

Electronic Banking (E-Banking)

Electronic Banking (E-Banking) Electronic Banking (E-Banking) E-banking is defined as the automated delivery of new and traditional banking products and services directly to customers through electronic, interactive communication channels.

More information

in Fintech, Healthcare, and ecommerce

in Fintech, Healthcare, and ecommerce 10100100011011111111111110011 01001000101 1010 1010 1101011111011010101010101011 0110011100111111111101010000 Fraud Detection: How Machine Learning Systems Help Reveal Scams 11110000100101101010101010

More information

Enterprise Artificial Intelligence

Enterprise Artificial Intelligence Ron Bodkin VP and GM, Artificial Intelligence Enterprise Artificial Intelligence Applications and Approaches 1 2015 Teradata 2016 Teradata Evolution of Analytics Prescriptive Predictive Descriptive REPORTING

More information

Payment Processor Buying Guide. How to prepare for sending out an RFP

Payment Processor Buying Guide. How to prepare for sending out an RFP Payment Processor Buying Guide How to prepare for sending out an RFP Payment Processor Buying Guide This document is meant to provide potential customers a framework to evaluate payment processors. PRELIMINARY

More information

Machine Learning 101

Machine Learning 101 Machine Learning 101 Mike Alperin September, 2016 Copyright 2000-2016 TIBCO Software Inc. Agenda What is Machine Learning? Decision Tree Models Customer Analytics Examples Manufacturing Examples Fraud

More information

Financial Services: Maximize Revenue with Better Marketing Data. Marketing Data Solutions for the Financial Services Industry

Financial Services: Maximize Revenue with Better Marketing Data. Marketing Data Solutions for the Financial Services Industry Financial Services: Maximize Revenue with Better Marketing Data Marketing Data Solutions for the Financial Services Industry DataMentors, LLC April 2014 1 Financial Services: Maximize Revenue with Better

More information

No one noticed until $250,000 and 7 months later, a former employee called asking why his W-2 reflected $50,000 more than he was paid!

No one noticed until $250,000 and 7 months later, a former employee called asking why his W-2 reflected $50,000 more than he was paid! THE CONSTRUCTION PROCESS OF FRAUD PREVENTION: The Absolutes of the Company Anti-Fraud Program Presented for the Fort Worth Chapter of the TSCPA Presented by STEVE DAWSON, CPA, CFE Dawson Forensic Group

More information

5 best (and worst) uses for Net Promoter Score

5 best (and worst) uses for Net Promoter Score 5 best (and worst) uses for Net Promoter Score. Issue: 2016 InsightSofa.com is a member of ROUCEK Group s.r.o.. All rights reserved 2016 Without exaggeration, Net Promoter SCORE is the best tool for measurement

More information

MORE POWER TO PERSUADE.

MORE POWER TO PERSUADE. MORE POWER IN YOUR PORTFOLIO. MORE POWER TO PERSUADE. MORE POWER TO YOU. Answers to your Merchants objections about accepting American Express Cards. 2015. American Express. All Rights Reserved. 1672_NAR_AMX_Objections_Brochure_r14.indd

More information

Model Selection, Evaluation, Diagnosis

Model Selection, Evaluation, Diagnosis Model Selection, Evaluation, Diagnosis INFO-4604, Applied Machine Learning University of Colorado Boulder October 31 November 2, 2017 Prof. Michael Paul Today How do you estimate how well your classifier

More information

Fraudulent Behavior Forecast in Telecom Industry Based on Data Mining Technology

Fraudulent Behavior Forecast in Telecom Industry Based on Data Mining Technology Fraudulent Behavior Forecast in Telecom Industry Based on Data Mining Technology Sen Wu Naidong Kang Liu Yang School of Economics and Management University of Science and Technology Beijing ABSTRACT Outlier

More information

5 Key Components Needed for an Effective Data Analytics Program

5 Key Components Needed for an Effective Data Analytics Program 5 Key Components Needed for an Effective Data Analytics Program Jim Harris Vice President, Analytics & Reporting Genex December 14, 2017 1 What is Analytics and Why Should You Care? Big Data is everywhere,

More information

BEYOND TELEMATICS LYTX.COM

BEYOND TELEMATICS LYTX.COM BEYOND TELEMATICS LYTX.COM 1 INTRO 90% of motor-vehicle collisions are caused by human error* Many organizations rely on traditional telematics technology to gain insights into their drivers performance.

More information

STAR Network Overview

STAR Network Overview STAR Network Overview Presented by: Jeff Jakopec, Sr. Strategy Business Development September 26, 2017 What Differentiates STAR Network From the Rest STAR provides market leading fraud solutions that help

More information

CHAPTER 3 OBJECTIVE. received a lot of attention on the research front, in particular from payment card

CHAPTER 3 OBJECTIVE. received a lot of attention on the research front, in particular from payment card OBJECTIVE 87 CHAPTER 3 OBJECTIVE In recent years, topics such as fraud detection and fraud prevention have received a lot of attention on the research front, in particular from payment card issuers. The

More information

WHITE PAPER. Anti-money laundering and counter terrorist financing A functional and technical description of the approach and tool

WHITE PAPER. Anti-money laundering and counter terrorist financing A functional and technical description of the approach and tool WHITE PAPER BayesAML Anti-money laundering and counter terrorist financing A functional and technical description of the approach and tool WHITE PAPER TABLE OF CONTENT Introduction................................................................

More information

Commerzbank 4.0: simple digital efficient Winning in the German Retail Banking Market

Commerzbank 4.0: simple digital efficient Winning in the German Retail Banking Market : Winning in the German Retail Banking Market 10 May 2017 Commerzbank AG Michael Mandel Member of the Board of Managing Directors London Achieving profitable growth and gaining market share Customers (GER*;

More information

Speaker: Steve Dawson 7/13/2017

Speaker: Steve Dawson 7/13/2017 Speaker: Steve Dawson 7/13/2017 THE CONSTRUCTION PROCESS OF FRAUD PREVENTION: The Absolutes of the Organization s Anti-Fraud Program Presented for the 2017 TACA On the Road Area Training Presented by STEVE

More information

Speech Analytics: Discovering the ROI. Whitepaper

Speech Analytics: Discovering the ROI. Whitepaper Speech Analytics: Discovering the ROI Whitepaper Table of Contents Speech Analytics: Discovering the ROI... 1 How Do You Find the Return on Investment?... 1 Collect the Data from Speech Analytics... 1

More information

Performance Metrics: Good, Bad, or Indifferent?

Performance Metrics: Good, Bad, or Indifferent? Performance Metrics: Good, Bad, or Indifferent? AN ENTREPRENEUR S GUIDE TO KEY PERFORMANCE METRICS By Mark Collins, Trep Advisors Revised December 2013 http://www.trepmetrics.com 2013 MCX Enterprises LLC

More information

This document (including, without limitation, any product roadmap or statement of direction data) illustrates the planned testing, release and

This document (including, without limitation, any product roadmap or statement of direction data) illustrates the planned testing, release and Alexander Uborcev, David Sweenor & Angela Waner Scaling Data Science & Empowering the Masses with TIBCO Statistica DISCLAIMER During the course of this presentation, TIBCO or its representatives may make

More information

KEEPING YOUR MONEY SAFE IS YOUR BANK S NUMBER ONE PRIORITY

KEEPING YOUR MONEY SAFE IS YOUR BANK S NUMBER ONE PRIORITY WWW.KNOWFRAUD.CO.UK * Know Fraud is a trademark of the National Fraud Intelligence Bureau, which is part of the City of London Police 2 KEEPING YOUR MONEY SAFE IS YOUR BANK S NUMBER ONE PRIORITY THAT S

More information

Decisions, decisions. With so many choices available to us, how can we be sure we're making the right decision?

Decisions, decisions. With so many choices available to us, how can we be sure we're making the right decision? LESSON PLAN Consumer Awareness Combine decision-making methods with comparative shopping techniques, recognize common consumer scams, and provide the knowledge and skills necessary to effectively handle

More information

https://youtu.be/p8di1km6v0i is there something Artificial about Intelligence? the dangers of climbing a ladder André De Locht Sr Business Value Consultant, Europe Unified Governance and Integration andre.de.locht@be.ibm.com

More information

Who Are My Best Customers?

Who Are My Best Customers? Technical report Who Are My Best Customers? Using SPSS to get greater value from your customer database Table of contents Introduction..............................................................2 Exploring

More information

ProfytPro ICO CONTENTS

ProfytPro ICO CONTENTS 1 ProfytPro ICO CONTENTS ProfytPro ICO CONTENTS... 2 INTRODUCTION... 4 WHAT IS ProfytPro (PFTC) PLATFORM... 5 CORE OBJECTIVES... 6 ADVANTAGES OF BLOCKCHAIN... 7 WHY WE DO TOKEN SALE... 8 TOKEN FUNCTIONS

More information

Mobile Marketing and PPC. What SMBs Should Be Doing, NOW! A WordStream Guide

Mobile Marketing and PPC. What SMBs Should Be Doing, NOW! A WordStream Guide Mobile Marketing and PPC What SMBs Should Be Doing, NOW! A WordStream Guide Mobile Marketing and PPC What SMBs Should Be Doing, NOW! Did you know that just under half of all Internet searches are performed

More information

Finance and Accounting outsourcing

Finance and Accounting outsourcing Finance and Accounting outsourcing RISK SOLUTION SERVICES Financial Solutions RISK SOLUTION SERVICES Integrated strategies for customer optimization We help clients minimise risk by analysing and scoring

More information

Summary. Tel: Fax

Summary. Tel: Fax Summary The analytical report provides the economic crisis scenario forecast for the consumer goods sector. The report also lists anti-crisis measures that should be taken in the nearest future and considers

More information

Cognitive Data Governance

Cognitive Data Governance IBM Unified Governance & Integration White Paper Powered by Machine Learning to find and use governed data Jo Ramos Distinguished Engineer & Director IBM Analytics Rakesh Ranjan Program Director & Data

More information

Mobile Banking Impact: Quantifying the ROI and Customer Engagement Benefits. Understanding the Value of Engaging Consumers in the Mobile Channel

Mobile Banking Impact: Quantifying the ROI and Customer Engagement Benefits. Understanding the Value of Engaging Consumers in the Mobile Channel Mobile Banking Impact: Quantifying the ROI and Customer Engagement Benefits Understanding the Value of Engaging Consumers in the Mobile Channel It goes without saying that mobile is an important channel

More information

TAS CASHLESS 3.0 FOCUS ON. The absolute framework for electronic payment management. CASHLESS 3.0: the ultimate. payment experience

TAS CASHLESS 3.0 FOCUS ON. The absolute framework for electronic payment management. CASHLESS 3.0: the ultimate. payment experience TAS CASHLESS 3.0 The absolute framework for electronic payment management CASHLESS 3.0: the ultimate payment experience CASHLESS 3.0 is TAS innovative processing platform that enables financial institutions,

More information

7 Tips to Successful Check Cashing

7 Tips to Successful Check Cashing 7 Tips to Successful Check Cashing Now that you ve started cashing checks here is what you need to know to ensure success Use the following best practices to help make your check cashing business successful.

More information

Advanced Recognition

Advanced Recognition HARVEY SPENCER ASSOCIATES Advanced Recognition Expanding the Capabilities of Capture 2017 HSA Inc. Page 1 of 10 Introduction Advanced Recognition consists of a series of technologies including pattern

More information

Mobile Financial Services : Perspectives with Agent Networks

Mobile Financial Services : Perspectives with Agent Networks Mobile Financial Services : Perspectives with Agent Networks Discussion Leaders Jordan Weinstock, OpenRevolution Leo Tobias, Grameen Foundation Session Information Introductions Approaching the Business

More information

PIN & PAY CARD FREQUENTLY ASKED QUESTIONS (FAQS) (Revised 17 August 2017)

PIN & PAY CARD FREQUENTLY ASKED QUESTIONS (FAQS) (Revised 17 August 2017) PIN & PAY CARD FREQUENTLY ASKED QUESTIONS (FAQS) (Revised 17 August 2017) 1. What is PIN? A PIN, or Personal Identification Number, is a secret code that is either assigned to, or selected by the cardmembers

More information

DIGITAL FINANCIAL SERVICES BASIC TERMINOLOGY

DIGITAL FINANCIAL SERVICES BASIC TERMINOLOGY INCLUDING ACRONYMS FOR DFS TRANSACTIONS THIS GUIDELINE NOTE WAS DEVELOPED BY THE AFI DIGITAL FINANCIAL SERVICES (DFS) WORKING GROUP TO PROVIDE UNIVERSAL DEFINITIONS OF KEY DIGITAL FINANCIAL SERVICES TERMS.

More information

AI Strategies in Retail

AI Strategies in Retail AI TRANSFORMATION AI Strategies in Retail Executive Brief Executive Summary The retail industry is evolving rapidly with large volumes of data and increasing challenges from new technologies. Early adopters

More information

SME BAROMETER Sector Report Accounting. exact.com/be

SME BAROMETER Sector Report Accounting. exact.com/be SME BAROMETER 2018 Sector Report Accounting exact.com/be 2 SECTOR REPORT - ACCOUNTING SECTOR REPORT - ACCOUNTING 3 CONTENT Preface 4 The SME Barometer for SMEs 2018 5 Show me the money! 6 Changing the

More information

TRUCK BUDS DRIVER'S INTRODUCTION MANUAL

TRUCK BUDS DRIVER'S INTRODUCTION MANUAL TRUCK BUDS DRIVER'S INTRODUCTION MANUAL Welcome to the Truck Buds platform. We are a family owned and operated company, that truly grows new roots every time a new driver signs up. Truck Buds is owned

More information

INTELLIGENT FINANCIAL CRIME DETECTION GETTING AHEAD OF FINANCIAL CRIME WITH AI THE POWER OF AI

INTELLIGENT FINANCIAL CRIME DETECTION GETTING AHEAD OF FINANCIAL CRIME WITH AI THE POWER OF AI INTELLIGENT FINANCIAL CRIME DETECTION GETTING AHEAD OF FINANCIAL CRIME WITH AI THE POWER OF AI BUSINESS SITUATION CRIME-DETECTION AND COMPLIANCE CAPABILITIES ARE STRAINED Financial crime is a major threat

More information

Datamart Helps Grow Profits & Customer Satisfaction

Datamart Helps Grow Profits & Customer Satisfaction Case Study Datamart Helps Grow Profits & Customer Satisfaction RTS Labs helps accounting services provider improve operational efficiencies and customer retention through the development of a customer

More information

Predictive Risk Analytics

Predictive Risk Analytics Predictive Risk Analytics Data-Driven Risk Measurement & Mitigation for Competitive Market Advantage 1of 15 Contents I. Risk Types II. Predictive Analytics, AI, ML Driving Capital Solutions: Opportunities

More information

Chip and PIN Programme. Using chip and PIN

Chip and PIN Programme. Using chip and PIN Chip and PIN Programme Using chip and PIN Introduction This material is designed to train retail staff on using chip and PIN. This presentation is in PDF format, however if you would like to extract slides

More information

Save time and make cash flow. Optimise your accounting software and send invoices that can be paid in moments.

Save time and make cash flow. Optimise your accounting software and send invoices that can be paid in moments. Save time and make cash flow Optimise your accounting software and send invoices that can be paid in moments. 2 Imagine getting paid in moments There are many ways that businesses can see their money sooner,

More information

Mind Your Own Business

Mind Your Own Business Mind Your Own Business You may be asking by now, what is the point of all this financial analysis that has been presented in the previous seven articles? It isn t to make work for your accountant although

More information

HOW TO USE ARTIFICIAL INTELLIGENCE & ANALYTICS IN AUDIT. Babu Jayendran

HOW TO USE ARTIFICIAL INTELLIGENCE & ANALYTICS IN AUDIT. Babu Jayendran HOW TO USE ARTIFICIAL INTELLIGENCE & ANALYTICS IN AUDIT Babu Jayendran SPEAKER PROFILE Babu Jayendran has over 35 years experience in Information Systems Auditing and Controls. Consults on SAP s Governance,

More information

7 STEPS TO SUCCESSFUL RETENTION AUTOMATION YOUR GUIDE TO MAXIMIZING REVENUE FROM YOUR CUSTOMER DATA

7 STEPS TO SUCCESSFUL RETENTION AUTOMATION YOUR GUIDE TO MAXIMIZING REVENUE FROM YOUR CUSTOMER DATA 7 STEPS TO SUCCESSFUL RETENTION AUTOMATION YOUR GUIDE TO MAXIMIZING REVENUE FROM YOUR CUSTOMER DATA It costs a lot to acquire a new customer, but most will make a single purchase and then leave. A repeat

More information

7 STEPS TO SUCCESSFUL RETENTION AUTOMATION YOUR GUIDE TO MAXIMIZING REVENUE FROM YOUR CUSTOMER DATA

7 STEPS TO SUCCESSFUL RETENTION AUTOMATION YOUR GUIDE TO MAXIMIZING REVENUE FROM YOUR CUSTOMER DATA 7 STEPS TO SUCCESSFUL RETENTION AUTOMATION YOUR GUIDE TO MAXIMIZING REVENUE FROM YOUR CUSTOMER DATA It costs a lot to acquire a new customer, but most will make a single purchase and then leave. A repeat

More information

EY Digital Boardroom. Overview. EY Digital Boardroom 1

EY Digital Boardroom. Overview. EY Digital Boardroom 1 EY Digital Boardroom Overview EY Digital Boardroom 1 Dear EY community, Markus Heinen Partner, Advisory Services EY GSA The disruption of finance functions is not a reality of the distant future but actually

More information

Empowering Customer Analytics, Fraud Detection, and Threat Assessment. Whitepaper

Empowering Customer Analytics, Fraud Detection, and Threat Assessment. Whitepaper Empowering Customer Analytics, Fraud Detection, and Threat Assessment Whitepaper EMPOWERING CUSTOMER ANALYTICS, FRAUD DETECTION, AND THREAT ASSESSMENT October 2015 novetta.com 2015, Novetta Most organizations

More information

ROCKING ANALYTICS IN A DATA FLOODED

ROCKING ANALYTICS IN A DATA FLOODED ROCKING ANALYTICS IN A DATA FLOODED WORLD: CHALLENGES AND OPPORTUNITIES Prof. dr. Bart Baesens Department of Decision Sciences and Information Management, KU Leuven (Belgium) School of Management, University

More information

Predictive Analytics Using Support Vector Machine

Predictive Analytics Using Support Vector Machine International Journal for Modern Trends in Science and Technology Volume: 03, Special Issue No: 02, March 2017 ISSN: 2455-3778 http://www.ijmtst.com Predictive Analytics Using Support Vector Machine Ch.Sai

More information

the S p e e c h A n a l y t i c s a n d t h e C o n t a c t C e n t r e

the S p e e c h A n a l y t i c s a n d t h e C o n t a c t C e n t r e Speech Analytics In the Contact Centre Why Speech Analytics Matters Keeping up with customers is getting harder The Gap Service Design Speed of Innovation Low Effort Engagement C o n t a c t C e n t r

More information

STUDY ON THE PERFORMANCE IMPACT ON PAID SEARCH RESULTS DUE TO THE ELIMINATION OF ADS ON THE RIGHT HAND SIDE OF THE GOOGLE SERP

STUDY ON THE PERFORMANCE IMPACT ON PAID SEARCH RESULTS DUE TO THE ELIMINATION OF ADS ON THE RIGHT HAND SIDE OF THE GOOGLE SERP APRIL 28, 2016 STUDY ON THE PERFORMANCE IMPACT ON PAID SEARCH RESULTS DUE TO THE ELIMINATION OF ADS ON THE RIGHT HAND SIDE OF THE GOOGLE SERP A NETELIXIR UNIVERSITY WHITEPAPER Table of Contents 1. Executive

More information

ANZ EFTPOS card and ANZ Visa Debit card CONDITIONS OF USE

ANZ EFTPOS card and ANZ Visa Debit card CONDITIONS OF USE ANZ EFTPOS card and ANZ Visa Debit card CONDITIONS OF USE As part of our commitment to you, this document meets the WriteMark Plain English Standard. If you have any questions about these Conditions of

More information

Month End Close Valuation of AR and Revenue Analysis

Month End Close Valuation of AR and Revenue Analysis Month End Close Valuation of AR and Revenue Analysis HFMA Annual Accounting Update Great Lakes Chapter Douglas C Banks CPA, CIA, FHFMA Director Revenue Excellence Reimbursement Trinity Health January 25,

More information

Texting Is Not Just for Teenagers: The Power of Proactive SMS

Texting Is Not Just for Teenagers: The Power of Proactive SMS Texting Is Not Just for Teenagers: The Power of Proactive SMS John Tallarico VP, Product Management, Genesys Fred Mondragon VP, Business Development, Medallia Consumers prefer SMS for customer support

More information

Feature engineering is Your ticket to survival in Director Data Science

Feature engineering is Your ticket to survival in Director Data Science Feature engineering is Your ticket to survival in Analytics @FilipVitek, Director Data Science Who the hell is Filip Vitek? Mr. Filip Vítek 15 years building business Built strategies, Data Science, CRM

More information

First Data Personal Financial Manager (PFM) FAQ s

First Data Personal Financial Manager (PFM) FAQ s Q: What browsers are supported with PFM? First Data Personal Financial Manager (PFM) FAQ s A: IE9 and 10, and current versions of Firefox, Chrome, and Safari are supported. Older versions may work but

More information

Customer Lifetime Value Modelling

Customer Lifetime Value Modelling Rhino Risk Customer Lifetime Value Modelling Alan Lucas Rhino Risk Ltd. 1 Purpose of Talk To discuss the Customer Value paradigm, explaining how it can be used to optimize customer-level decisions Key

More information

ECONOMIC AND FINANCIAL LITERACY LEARNING FRAMEWORK TO SUPPORT THOSE DEVELOPING CURRICULUM RE FINANCIAL CAPABILITY AND RESPONSIBILITY

ECONOMIC AND FINANCIAL LITERACY LEARNING FRAMEWORK TO SUPPORT THOSE DEVELOPING CURRICULUM RE FINANCIAL CAPABILITY AND RESPONSIBILITY ECONOMIC AND FINANCIAL LITERACY LEARNING FRAMEWORK TO SUPPORT THOSE DEVELOPING CURRICULUM RE FINANCIAL CAPABILITY AND RESPONSIBILITY Developed by: Gary Rabbior, President, Canadian Foundation for Economic

More information

WebSphere ILOG BRMS. Stephane Marouani Sales and Operations Manager, IBM ILOG A/NZ. IBM Insight Forum 09. Make change work for you

WebSphere ILOG BRMS. Stephane Marouani Sales and Operations Manager, IBM ILOG A/NZ. IBM Insight Forum 09. Make change work for you WebSphere ILOG BRMS Stephane Marouani Sales and Operations Manager, IBM ILOG A/NZ IBM Insight Forum 09 Make change work for you What I will cover The Needs and Pains for Changes Understanding Business

More information

A BUYER S GUIDE TO CHOOSING A MOBILE MARKETING PLATFORM

A BUYER S GUIDE TO CHOOSING A MOBILE MARKETING PLATFORM A BUYER S GUIDE TO CHOOSING A MOBILE MARKETING PLATFORM A Buyer s Guide to Choosing a Mobile Marketing Platform Today, mobile users are demanding more from their app experiences, and the priority is clear:

More information

NGDATA. Luc Burgelman, CEO

NGDATA. Luc Burgelman, CEO NGDATA Luc Burgelman, CEO Introduction The Company NGDATA Overview Helping data-rich companies drive connected customer experiences Recognised by the leading US-based CRM Magazine as one of the 7 Rising

More information

The Customer Engagement Company. Delivering solutions that forge deeper, more meaningful and valuable customer relationships

The Customer Engagement Company. Delivering solutions that forge deeper, more meaningful and valuable customer relationships The Customer Engagement Company Delivering solutions that forge deeper, more meaningful and valuable customer relationships Customer Engagement for the Long Term Engagement is fundamentally changing. End-customers

More information

Using Artificial Intelligence to Automate Expense Report Audit, Fraud Detection, and Risk Compliance

Using Artificial Intelligence to Automate Expense Report Audit, Fraud Detection, and Risk Compliance Using Artificial Intelligence to Automate Expense Report Audit, Fraud Detection, and Risk Compliance August 2016 Artificial Intelligence for Business Manual expense audit leaves companies vulnerable to

More information