Session about types of analytics. Descriptive, diagnostic, predictive and prescriptive analytics.
Conference DATA ANALYSIS DEVELOPMENT 2016 by RZECZPOSPOLITA.
2. Łukasz Grala
Architect Data
Platform
& BI Solutions |
Microsoft MVP Data
Platform
Architekt rozwiązań platformy danych i BI
Projektant, trener i konsultant
Wykładowca na wyższych uczelniach
Autor artykułów, publikacji i webcastów
Związany naukowo z Wydziałem Informatyki Politechniki
Poznańskiej (architektura baz i hurtowni danych, uczenie
maszynowe, eksploracja danych)
Zawodowo architekt freelancer w TIDK, trener i ekspert w
SQLExpert, współpracownik technologiczny Microsoft
Członek PLSSUG, PTI
Prelegent na licznych konferencjach
lukasz@tidk.pl
3. Agenda
Types of Analytics and Business
Intelligence
Prescriptive analytics
Summary
lukasz@tidk.pl
6. Business
Intelligence
lukasz@tidk.pl
ETL
BI
Applications
Enterprise Data
Warehouse (EDW)
Normalized tables
(3NF)
Atomic data
User queryable
ETL
Presentation Area:
Dimensional (star
schema or OLAP
cube)
Atomic and summary
data
Organized by
business process
Uses conformed
dimensions
Enterprise DW Bus
Architecture
Source
Transactions
Back Room Front Room
Source: Hybrid Hub-and-Spoke and Kimball Architecture (R.Kimball, M.Ross „The DataWarehouse Toolkit” 3ed - 2013
12. BI Solutions
ETL Tool
(SSIS, etc) EDW
(SQL Server, Teradata, etc)
Extract
Original Data
Load
Transformed
Data
Transform
BI Tools
Ingest (EL)
Original Data
Scale-out
Storage &
Compute
(HDFS, Blob Storage,
etc)
Transform & Load
Data Marts
Data Lake(s)
Dashboards
Apps
Streaming data
lukasz@tidk.pl
18. Classes
Learning
Problems
lukasz@tidk.pl
Classification: Assign a category to each item (Chinese | French
| Indian | Italian | Japanese restaurant).
Regression: Predict a real value for each item
(stock/currency value, temperature).
Ranking: Order items according to some criterion
(web search results relevant to a user query).
Clustering: Partition items into homogeneous groups
(clustering twitter posts by topic).
Dimensionality reduction: Transform an initial representation of items
into a lower-dimensional representation while preserving some
properties (preprocessing of digital images).
23. Prescriptive
Analytics
lukasz@tidk.pl
e.g. business rules, regulatory requirments,
technolgoy requirments, HR policies
e.g. profit per item, per unit,
throughput per hour
Find best solutions (variable values) to meet objectives
e.g. maximise profit, minimise cost, minimise downtime