Strategic Business Analytics (engelstalige workshop) (nieuwe sessie)

Doelstelling

During the past decades, many companies have gathered masses of data throughout their various business activities.Examples are data capturing the marketing, financial, or even fraud behaviour of customers. Today it becomes extremely important to extract, understand and exploit analytical patterns of customer behaviour and strategic intelligence.

It is the purpose of this workshop to clarify how analytics and data mining can be successfully adopted for strategic business applications. The workshop aims at providing a sound mix of both theoretical and technical insights, as well as practical implementation details, illustrated by several real-life cases. The workshop will be highly interactively organised. Background material will also be provided.

In this workshop, you will learn how to (1) use analytics and data mining to optimize strategic business processes, (2) sample and preprocess data in an optimal way, (3) use predictive analytics for detecting associations, sequences and/or clusters in your data and (4) put analytics successfully to work taking into account various model requirements (interpretation, efficiency, economical cost, compliance,etc.)

Attendee profile: business analyst, (senior) data analyst, data miner, (senior) CRM analyst, marketing analyst, risk analyst, analytical model developer, etc.  

Prerequisites: the attendees should have a basic background in statistics.

Programma

Part 1: Introduction

Strategic Business Analytics Process Model - Job profiles involved - Basic Nomenclature - Data collection and sampling - Predictive versus descriptive analytics - Putting analytics to work - Strategic impact of analytics - Key application areas

Part 2: Data collection and sampling  

Types of data sources - Sampling - Master Data Management - Data Quality

Part 3: Predictive Analytics  

Target defintion (e.g. fraud, churn, default, CLV) - Logistic regression for predictive modeling - Decision trees for predictive modeling - Measuring performance of predictive analytical models (lift curves, ...) - Applications in CRM and Risk Management - Demo: Predictive analytics in SAS Enterprise Miner

Part 4: Descriptive Analytics 

Association analysis - Sequence analysis - Customer segmentation - Applications (Up/Down/ Cross Selling, Market Basket Analysis) - Demo: Clickstream analysis using SAS Web Analytics

Part 5: Social Network Analytics

Social network definition - Examples (Facebook, Twitter, etc.) - Social network metrics (closeness, betweenness, ...) - Mining communities - Network analytics - Relational logistic regression

Part 6: Putting analytics to work  

Model requirements - Model interpretation - Operational efficiency - Economical cost - Regulatory compliance - Model implementation - Model monitoring (backtesting) - Benchmarking

Conclusions