sitnl 2015 algorithm programming made easy (marcel de bruin, sander de wildt)

Post on 13-Apr-2017

363 Views

Category:

Business

3 Downloads

Preview:

Click to see full reader

TRANSCRIPT

Use this title slide only with an image

Algorithm programming made easy

Sander de Wildt, The Next ViewMarcel de Bruin, SAPNov 2015

© 2013 SAP AG or an SAP affiliate company. All rights reserved. 2

Predictive analytics

HANAAFL_PAL

Application function library –

predictive analytics libary

APL

Automated predictive insight

Predictive functions are everywhere

Automated(KXEN/SAP Infinite

Insight

Expert(SAP Predictive Analysis)

© 2013 SAP AG or an SAP affiliate company. All rights reserved. 3

Predictive analytics

HANAAFM

Application function library –

predictive analytics libary

APL

Automated predictive insight

Predictive functions are everywhere

Automated(KXEN/SAP Infinite

Insight

Expert(SAP Predictive Analysis)

AdvancedAnalytics Enterprise

Business Intelligence

Agile Visualization

Advanced Analytics

How analytics need to evolve to deliver collective insights

RawData

CleanedData

Standard Reports

Ad Hoc Reports &

OLAP

Agile Visualization

Predictive Modeling

Optimization

What happened?

Why did it happen?

What will happen?

What is the best that

could happen?

Use

r Eng

agem

ent

Maturity of Analytics Capabilities

Self Service BI

Generic Predictive Analysis

End-to-endEasy adoption

Fast implementation Business focused Enable storytelling

Col

lect

ive

Insi

ght

© 2013 SAP AG. All rights reserved. 6

RETAIL

Store Segmentation & Performance

Returner Segmentation& Targeting Product Launch

Demographic Profiling for Marketing

Markdown Optimization

Real-time PersonalizedOffers

Buyer Classification & Churn Prevention

Sales and Inventory Forecasting

Customer Sentiment Analysis

Customer Loyalty Analysis

Short-term Check-outPrediction

Market BasketAnalysis

Predictive Use Cases

© 2013 SAP AG. All rights reserved. 7

RETAIL

Store Segmentation & Performance

Returner Segmentation& Targeting Product Launch

Demographic Profiling for Marketing

Markdown Optimization

Real-time PersonalizedOffers

Buyer Classification & Churn Prevention

Sales and Inventory Forecasting

Customer Sentiment Analysis

Customer Loyalty Analysis

Short-term Check-outPrediction

Market BasketAnalysis

Predictive Use Cases

© 2013 SAP AG. All rights reserved. 8

APRIORI Algorithm

LIFT:

Confidence :How often when a customer bought A he also bought B. Milk & Bread = 2/3 because out of 3 transactions that contain Milk, only 2 contain the Milk & Bread

Support:How many times does a article combination

Occur. Milk & Bread = 2/5

Transaction product

1 Milk1 Bread2 Butter3 Beer4 Milk4 Bread4 Butter5 Milk5 Butter Support Milk & Bread

Support Milk X Support Bread = 1,6

© 2013 SAP AG. All rights reserved. 9

Predictive Use Cases

Predictive results incorporated into more applications and processes, for more users

Bridges the skills gap between human experts vs. analytical experts

Predict and act in real-time

Advanced AnalyticsConfidently anticipate what comes nextto drive better business outcomes

PREDICT

© 2014 SAP AG or an SAP affiliate company. All rights reserved.

Thank You

m.de.bruin@sap.comSolution engineer HANA/ BI

sander.de.wildt@thenextview.nlBig Data Consultant

top related