oracleadvancedanalyticsv2otn-2859525.pptx

Copyright © 2016 Oracle and/or its affiliates. All rights
reserved. |
Charlie Berger, MS Engineering, MBA
Sr. Director Product Management, Data Mining and Advanced
Analytics
charlie.berger@oracle.com www.twitter.com/CharlieDataMine
Oracle’s Advanced Analytics
Make BigData +AnalyticsSimple

reserved.
Safe Harbor
Statement
The following is intended to outline our general product direction. It is intended for
information purposes only, and may not be incorporated into any contract. It is not
a commitment to deliver any material, code, or functionality, and should not be
relied upon in making purchasing decisions. The development, release, and
timing of any features or functionality described for Oracle’s products remains at the
sole discretion of Oracle.
2

Data Analysis
platforms
requirements:
• Be extremely powerful
and handle large data
volumes
• Be easy to learn
• Be highly automated &
enable deployment
Data, data everywhere
Growth of Data Exponentially
Greater than Growth of Data
Analysts!
reserved.
http://www.delphianalytics.net/more-data-than-analysts-the-real-big-data-
problem/ http://uk.emc.com/collateral/analyst-reports/ar-the-economist-data-data-
everywhere.pdf

Analytics + Data Warehouse +
Hadoop
reserved.
• Platform Sprawl
– More Duplicated Data
– More Data Movement
Latency
– More Security challenges
– More Duplicated Storage
– More Duplicated Backups
– More Duplicated Systems
– More Space and Power

Visio
n
• Big Data +Analytic Platform for the Era of Big Data and
Cloud
– Make BigData +AnalyticsModel DiscoverySimple
• Any data size, on any computer infrastructure
• Any variety of data (structured, unstructured, transactional, geospatial),
in any combination
– Make BigData +AnalyticsModel DeploymentSimple
• As a service, as a platform, as an application
reserved.

 Scalable in-Database + Hadoop
data mining algorithms and R
integration
 Powerful predictive analytics and
deployment platform
 Drag and drop workflow, R and SQL
APIs
 Data analysts, data scientists &
developers
 Enables enterprise predictive
analytics applications
KeyFeatures
Fastest Way to Deliver Scalable Enterprise-wide Predictive
Analytics
reserved.

Data remains in Database &
Hadoop
 Model building and scoring occur in-
database
 Use R packages with data-parallel
invocations
Leverage investment in Oracle IT
 Eliminate data duplication
Eliminate separate analytical servers
Fastest way to deliver enterprise-
wide predictive analytics
 GUI for Predictive Analytics and code
generation
 R interface leveraging database as HPC
engine
KeyFeatures
avings
Model“Scoring”
EmbeddedDataPrep
Data Preparation
ModelBuilding
OracleAdvanced Analytics
Secs,Mins or Hours
TraditionalAnalytics
Hours,Daysor Weeks
DataExtraction
Data Prep&
T
ransformation
DataMining
Model Building
Data Mining
Model “Scoring”
Data Prep.&
T
ransformation
Data Import
reserved.

OBIEE
OracleDatabaseEnterprise Edition
Oracle’s AdvancedAnalytics
OracleAdvancedAnalytics- Database Option
SQLData Mining & Analytic Functions +RIntegration
for Scalable, Distributed, Parallel in-Database ML Executio
SQL Developer/
Oracle Data Miner
Applications
R Client
Multiple interfaces across platforms —SQL, R, GUI, Dashboards,
Apps
Users R programmers Data & Business Analysts Business
Analysts/Mgrs Domain End Users
Platfor
m
OracleDatabase
n 12c
Hadoop
ORAAH
Parallel,
distributed
algorithms
OracleCloud
reserved.

Oracle Advanced Analytics Database
Evolution
Analytical SQL in the Database
2002 2004 2005 2008 2011 2014
• 7 Data Mining
“Partners”
1998
• Oracle acquires
Thinking Machine
Corp’s dev. team +
“Darwin” data
mining software
1999
Miner “Classic”
wizards driven GUI
• SQL statistical
functions introduced
• New algorithms (EM,
PCA, SVD)
• Predictive Queries
• SQLDEV/Oracle Data
Miner 4.0 SQL script
generation and SQL
Query node (R integration)
scalable R algorithms
• Oracle Adv. Analytics
for Hadoop Connector
launched with
scalable BDA
algorithms
• Oracle Data Mining
10gR2 SQL - 7 new
SQL dm algorithms
• Oracle Data Mining and new Oracle Data
9.2i launched – 2
algorithms (NB
and AR) via Java
API
• ODM 11g & 11gR2 adds
AutoDataPrep (ADP), text
mining, perf. improvement•sOAA/ORE 1.3 + 1.4
• SQLDEV/Oracle Data Mineradds NN, Stepwise,
3.2 “work flow” GUI
launched
• Integration with “R” and
introduction/addition of
Oracle R Enterprise
• Product renamed “Oracle
Advanced Analytics (ODM +
ORE)
reserved.

YouCanThink of Oracle’s Advanced Analytics Like This…
– “Human-driven” queries
– Domain expertise
– Any “rules” must be definedand
managed
SQLQueries
– SELECT
– DISTINCT
– AGGREGATE
– WHERE
– AND OR
– GROUP BY
– ORDER BY
– RANK
Traditional SQL OracleAdvancedAnalytics- SQL &
– Automated knowledge discovery,
model building and deployment
– Domain expertise to assemblethe“right”
data to mine/analyze
Analytical SQL“Verbs”
– PREDICT
– DETECT
– CLUSTER
– CLASSIFY
– REGRESS
– PROFILE
– IDENTIFY FACTORS
– ASSOCIATE
+
reserved.

Multiple Data Sources/Types with Predictive
Modeling
Ease of Deployment through SQL Script
Generation
Conside
r:
• Demographics
• Past purchases
• Recent purchases
• Customer comments &
tweets
Unstructured
data
also mined
by
algorithms
Transaction
al
POS
data
Generates SQL
scripts
for deployment
Inline
predictive
model to
augment
input data
SQL Joins and arbitrary SQL
transforms & queries – power of
SQL
reserved.

UK National Health
Service
Combating Healthcare Fraud
Objectives
 Use new insight to help identify cost savings and meet
goals
 Identify and prevent healthcare fraud and benefit eligibility
errors to save costs
 Leverage existing data to transform business and
productivity
Solution
 Identified up to GBP100 million (US$156 million)
potentially
saved through benefit fraud and error reduction
 Used anomaly detection to uncover fraudulent activity
where some dentists split a single course of treatment
into multiple parts and presented claims for multiple
treatments
 Analyzed billions of records at one time to measure
longer- term patient journeys and to analyze drug
prescribing patterns to improve patient care
 “Oracle Advanced Analytics’ data mining capabilities andOracle
Exalytics’ performance really impressed us. The overall solution is
very fast, and our investment very quickly provided value. We can
now do so much more with our data, resulting in significant
savings for the NHS as a whole”
– Nina Monckton, Head of Information Services,
NHS Business ServicesAuthority
OracleExadataDatabase
Machine
OracleAdvanced
Analytics
OracleExalyticsIn-Memory
Machine
OracleEndecaInformation
Discovery
OracleBusinessIntelligence EE
reserved.

Turkcell
Combating Communications
Fraud
Objectives
 Prepaid card fraud—millions of dollars/year
 Extremely fast sifting through huge
data volumes; with fraud, time is
money
Solution
 Monitor 10 billion daily call-data records
 Leveraged SQL for the preparation—1 PB
 Due to the slow process of moving data,
Turkcell IT builds and deploys models in-DB
 Oracle Advanced Analytics on
Exadata for extreme speed. Analysts
can detect fraud patterns almost
immediately
 “Turkcell manages 100 terabytes of compressed data—or one
petabyte of uncompressed raw data—on Oracle Exadata. With
Oracle Data Mining, a component of the Oracle Advanced
Analytics Option, we can analyze large volumes of customer data
and call-data records easier and faster than with any other tool
and rapidly detect and combat fraudulent phone use.”
– Hasan Tonguç Yılmaz, Manager, Turkcell İletişim HizmetleriA.Ş.
OracleAdvanced Analytics
In-Database FraudModels
Exadata
reserved.

reserved.
Oracle Advanced Analytics—On Premise or
Cloud
100% Compatibility Enables Easy Coexistence and Migration
14
OracleCloud
CoExistenceand Migration
SameArchitecture
SameAnalytics
SameStandards
T
ransparently move workloadsand analytical methodologies
betwee
n On-premiseandpublic cloud
On-Premise

Predictive Analytics & Data
Mining
Typical Use Cases
• Targeting the right customer with the right offer
• How is a customer likely to respond to an offer?
• Finding the most profitable growth opportunities
• Finding and preventing customer churn
• Maximizing cross-business impact
• Security and suspicious activity detection
• Understanding sentiments in customer
conversations
• Reducing medical errors & improving quality of
health
• Understanding influencers in social networks
reserved.

Automatically sifting through large amounts of data
to create models that find previously hidden
patterns, discover valuable new insights and
make predictions
•Identify most important factor (Attribute Importance)
•Predict customer behavior (Classification)
•Predict or estimate a value (Regression)
•Find profiles of targeted people or items (Decision
Trees)
•Segment a population (Clustering)
•Find fraudulent or “rare events” (AnomalyDetection)
•Determine co-occurring items in a“baskets”
(Associations)
What is Data Mining & Predictive
Analytics?
A1 A2 A3 A4 A5 A6
A7
reserved.

Data Mining
Provides
Customer
Months
Better Information, Valuable Insights and
Predictions
LeaseChurners vs. Loyal Customers
Insight &
Prediction Segment #1
IF CUST_MO > 14 AND
INCOME <
$90K, THENPrediction =
Lease Churner
Confidence =
100% Support =
8/39
Segment #3
IF CUST_MO > 7 AND
INCOME <
$175K, THEN
Prediction = Lease Churner,
Confidence =
83% Support =
6/39
Source: Inspired from Data Mining Techniques: ForMarketing, Sales,and CustomerRelationship Management by Michael J. A. Berry, Gordon S.
Linoff Copyright © 2016 Oracle and/or its affiliates. All rights
reserved.

Oracle Advanced Analytics: Supervised
Learning
“Classification” Decision TreeAlgorithm
• Profiling
with
discovered
If… Then…
“rules”
• Prediction
probabilitie
s
>4
5
<4
5
Manua
l
Automati
c
<=3
5
>3
5
Ag
e
<25 >25 M
Buy Camry= 0 Buy Camry= 1 Camry= 0 Camry=
1 Camry= 0 Camry= 1
Ag
e
Car
Type
Mp
g
Gende
r
Fast
Accelerations
F >4 <=4
Simplemodel: Could include
unstructured data (e.g. voice of
customer comments),
transactions data (e.g.
geospatial & filed performance
data), etc.
IF (Age >45ANDCARTYPE=Automatic ANDMpg= >25)
THEN Probability(Buy Camry) =.77 and Support =25,500 cases
reserved.

Oracle’s AdvancedAnalytics
In-Database Data MiningAlgorithms*—SQL& & GUIAccess
• Decision Tree
• Logistic Regression (GLM)
• Naïve Bayes
• Support Vector Machine
(SVM)
• Random Forest
Regression
• Multiple Regression (GLM)
• Support Vector Machine
(SVM)
• Linear Model
• Generalized Linear Model
• Multi-Layer Neural Networks
• Stepwise Linear Regression
AttributeImportance
• Unsupervised pair-wise KL
div.
g
•Hierarchical k-Means
•Orthogonal Partitioning
Clusterin
•Expectation-Maximization
FeatureExtraction& Creation
•Nonnegative Matrix
Factorizatio
n
•Principal Component
Analysis
•Singular Value
Decomposition
• Apriori – Association Rules
Anomaly Detection
•1 Class Support Vector
Machine
Time Series
• Single & Double Exp.
Smoothing
Classification Clustering
Market BasketAnalysis
•Clustering
•Regression
•Anomaly Detection
•Feature Extraction
Predictive Queries
•Ability to run any R package
via
Embedded R
mode
OpenSourceRAlgorithms
* supports partitioned models, text
mining
A1 A2 A3 A4
A5
• Minimum Description Length
A6
A7
reserved.

• R language for interaction with the
database
• R-SQL Transparency Framework
overloads R functions for scalable in-
database execution
• Function overload for data
selection, manipulation and
transforms
• Interactive display of graphical results
and flow control as in standard R
• Submit user-defined R functions for
execution at database server under
control of Oracle Database
• 15+ Powerful data mining
algorithms (regression,
clustering, AR, DT, etc._
• Run Oracle Data Mining SQL data mining
functioning (ORE.odmSVM, ORE.odmDT,
etc.)
• Speak“R” but executes asproprietary in-
database SQL functions—machine
learning algorithms and statistical
functions
• Leverage database strengths: SQL
parallelism,
scale to large datasets, security
Other R
packag
es
Oracle R Enterprise (ORE)
packages
R->SQLTransparency“Push-Down”
• R Engine(s) spawned by Oracle
DB for database-managed
parallelism
• ore.groupApply high performance
scoring
• Efficient data transfer to
spawned R engines
• Emulate map-reduce style algorithms
and applications
• Enables production deployment and
automated execution of R scripts
OracleDatabase12c
R->
SQL
Result
s
In-DatabaseAdvAnalytical SQLFunctions
R
Engine
Other R
packag
es
Oracle R Enterprise packages
EmbeddedRPackageCallouts
R
Result
s
Oracle AdvancedAnalytics
How Oracle R Enterprise Compute Engines
Work
1 2 3
reserved.

reserved. |
Oracle Advanced
Analytics
Brief Demos

reserved.
22

reserved.
23

reserved.
24

reserved.
25

reserved.
26

reserved.
27

reserved.
28

reserved.
29

reserved.
30

reserved.
31

reserved.
32

reserved.
33

Oracle’sAdvancedAnalytics
ExampleCustomer References
reserved. |

Fiser
v
Risk Analytics in Electronic Payments
Objectives
 Prevent $200M in losses every year using
data to monitor, understand and anticipate
fraud
Solution
 We installed OAA analytics for model development
during 2014
 When choosing the tools for fraud management,
speed is a critical factor
 OAA provided a fast and flexible solution for
model building, visualization and integration with
production processes
 “When choosing the tools for fraud management, speed is a
critical factor. Oracle Advance Analytics provided a fast and
flexible solution for model building, visualization and integration
with production processes.”
– Miguel Barrera, Director of Risk Analytics, FiservInc.
– Julia Minkowski, Risk Analytics Manager, Fiserv Inc.
3months
to run &
deploy
Logistic
Regression
(usingSAS)
1month
to estimate and
deploy Trees
and GLM
reserved. |
1 weekto
estimate, 1
weektoinstall
rules
in online
application
1 dayto estimateand
deploy
Trees + GLM models
(using Oracle
Advanced
Analytics)
Oracle Advanced
Analytics

© 2014 Fiserv, Inc. or its affiliates.
Data Miner Survey 2016 by Rexer Analytics
While 6 out 10 data miners report the data is available for analysis
within days of capture, the time to deploy the models takes
substantially longer. For 60% of
the respondents the
deployment time will range between 3 weeks and 1year.
Everyone
forgets about
deployment –
but is most
important
component!
Ease of Deployment

reserved.
Market Basket & Advanced Analytics at Dunkin
Brands
Objectives
 Store development dashboards to identify opportunities
 8 M daily transactions, ~25M transaction detail lines
 20 TB data warehouse size, sales data about 10 TB
 Market basket analysis and customer loyalty &
segmentation
Solution
 Exadata Engineered Systesm
 Oracle Advanced Analytics Option
 Market Basket Analysis, Clustering,
Classification, Segmentation, Loyalty
Analysis
 “Exponential growth in combinations with each hierarchy.
2 years of pre-computed Market Baskets and associated
sales measures for reporting. Nightly compute within ETL
window data with 1 day latency.”
– Dunkin Brands, Mahesh Jagannath, Senior Manager,
Business Intelligence
(Excerpts from Dunkin Brands presentation at Oracle Open World2014)

Excerpts from Dunkin Brands presentation at Oracle Open World
2014
https://oracleus.activeevents.com/2014/connect/fileDownload/session/82C7DB4F27CC2C97F29D8592754E4216/CON6545_Jagannath-
OOW2014MarketBasketAndAdvancedAnalyticsAtDunkinBrands10012014Final.pptx
Clusterin
g
Algorith
m
Custome
r Data
Profiles

reserved. |
DX
Marketing
Cloud Based Predictive Analytics/Database Marketing
Objectives
Solution
 “Time to market has significantly improved from 4-6 weeks to less
than a week with the result the company can bring new clients on
board faster. This has helped boost revenues by 25% in the six
months since using Oracle’s DBCS..”
– DX Marketing
OracleCloud
Oracle Advanced
Analytics
 Cloud-based solution
 Increase revenue
 Reduce time-to-
market

Accelerates Complex Segmentation Queries from Weeks to
Minutes—Gains Competitive Advantage
Objectives
 World’s leading customer-science company
 Accelerate analytic capabilities to near real time using
Oracle Advanced Analytics and third-party tools,
enabling analysis of unstructured big data from
emerging sources, like smart phones
Solution
 Accelerated segmentation and customer-loyalty analysis
from
one week to just four hours—enabling the company
to deliver more timely information & finer-grained
analysis
 Generated more accurate business insights and
marketing recommendations with the ability to analyze
100% of data— including years of historical data—
instead of just a small sample
 “Improved analysts’ productivity and focus as they can
now run queries and complete analysis without having to
wait hours or days for a query to process”
 “Improved accuracy of marketing recommendations by
analyzing larger sample sizes and predicting the market’s
reception to new product ideas and strategies”
– dunnhumby Oracle Customer Snapshot )
Customer Snapshot: http://www.oracle.com/us/corporate/customers/customersearch/dunnhumby-1-exadata-ss-
2137635.html
Copyright © 2016 Oracle and/or its affiliates. All rights reserved.

reserved. |
Differentiators

Oracle Advanced Analytics
Differentiators
• Data remains in the Database and Hadoop
– Brings algorithms to the data, not move data to the algorithms
– Oracle’sadvanced analytics parallel implementations run on Hadoop and Databasedata
– Provides parallel implementations of commonly used algorithms with auto-data preparation
– Enables data analysts and scientists to work directly on transactional data, star schemas,
text
• Leverage investment in Oracle IT
– Leverageslatest Oracletechnology: Engineered Systems(Exadata,BDA),BigData SQL,“smart scans”
– Provides wide range of options: Cloud and on-premise – same Oracle code & products
– Database provides industry leadership in data security, encryption, audit, backup, and
recovery
– Diverse data source gateways, social media data, geo-coding, ETL – on-premise / on-cloud
reserved. 4

Oracle Advanced Analytics
Differentiators
• Fastest way to deliver enterprise-wide predictive analytics
applications
– Delivers big data + analytics platform for model build, deployment & applications
– GUI to build analytical workflows with code generation and scheduling for deployment
– Supports near real-time scoring for embedding in applications
– Explains model logic and transparency via Prediction_Details
• Tight integration with open source R
– Leverage CRAN R packages using data- and task-parallel database infrastructure
– Store and manage R scripts and R objects in-database and invoke R scripts from SQL
– Use production quality infrastructure without custom plumbing or extra complexity
reserved. 4

SQL Developer/Oracle Data Miner
4.0
New Features
 SQLScriptGeneration
– Deploy entire methodology as a SQL
script
– Immediate deployment of dataanalyst’s
methodologies
R
reserved.

Fraud Prediction Demo
Automated In-DB Analytical
Methodology
drop table CLAIMS_SET;
exec dbms_data_mining.drop_model('CLAIMSMODEL');
create table CLAIMS_SET (setting_name varchar2(30), setting_value varchar2(4000));
insert into CLAIMS_SET values ('ALGO_NAME','ALGO_SUPPORT_VECTOR_MACHINES');
insert into CLAIMS_SET values ('PREP_AUTO','ON');
commit;
POLICYNUMBER PERCENT_FRAUDRNK
------------ ------------- ----------
6532 64.78 1
2749 64.17 2
3440 63.22 3
654 63.1 4
12650 62.36 5
begin
dbms_data_mining.create_model('CLAIMSMODEL', 'CLASSIFICATION',
'CLAIMS', 'POLICYNUMBER', null, 'CLAIMS_SET');
end;
/ Automated Monthly “Application”! Just
add:
Create
View CLAIMS2_30
As
Select * from CLAIMS2
Where mydate > SYSDATE – 30
Time measure: set timing on;
-- Top 5 most suspicious fraud policy holder claims
select * from
(select POLICYNUMBER, round(prob_fraud*100,2) percent_fraud,
rank() over (order by prob_fraud desc) rnk from
(select POLICYNUMBER, prediction_probability(CLAIMSMODEL, '0' using *) prob_fraud
from CLAIMS
where PASTNUMBEROFCLAIMS in ('2to4', 'morethan4')))
where rnk <= 5
order by percent_fraud desc;
reserved.

Oracle Advanced
Analytics
More Details
• On-the-fly, single record apply with new data (e.g. from
call center)
Call
Center
We
b
Mobile
Get
AdviceBranch
Office
Social
Media
Emai
l
R
Select prediction_probability(CLAS_DT_1_2, 'Yes'
USING 7800 as bank_funds, 125 as checking_amount, 20 as
credit_balance, 55 as age, 'Married' as marital_status,
250 as MONEY_MONTLY_OVERDRAWN, 1 as house_ownership)
from dual;
Likelihood to respond:
reserved.

Data Mining When Lack
Examples
Customer
Months
Better Information, Valuable Insights and
Predictions
CellPhoneFraud vs. Loyal Customers
Source: Inspired from Data Mining Techniques: ForMarketing, Sales,and CustomerRelationship Management by Michael J. A. Berry, Gordon S.
Linoff
?
reserved.

Challenge: Finding
Anomalies
• Considering
multiple
attributes
• Taken alone,
may seem
“normal”
• Taken
collectively, a
record may
appear to be
anomalous
• Look for what
is “different”
X
1
X2
X
3
X
4
X1
X2
X3
X4
reserved.

Tax Noncomplaince Audit
Selection
reserved.
• Simple Oracle Data
Mining predictive model
– Uses Decision Tree for
classification of
Noncompliant tax
submissions (yes/no)
based on historical 2011
data

Oracle Advanced
Analytics
OAA/OracleREnterprise(R integration)
reserved. |

• Strengths
– Powerful & Extensible
– Graphical & Extensive
statistics
– Free—open source
• Challenges
– Memory constrained
– Single threaded
– Outer loop—slows down
process
– Not industrial strength
reserved.
Renvironment
R—Widely Popular
R is a statistics language similar to Base SAS or SPSS
statistics

R: Transparency through function
overloading
Invoke in-database aggregation function
> aggdata <-
aggregate(ONTIME_S$DEST,
by =
list(ONTIME_S$DEST),
FUN = length)
ONTIME
_S
In-db
Stats
OracleDatabase
Oracle
SQL
select DEST, count(*)
from
ONTIME_S
group by
DEST
Oracle Advanced
Analytics ORE
Client Packages
Transparency
Layer
+
+
> class(aggdata)
[1] "ore.frame"
attr(,"package")
[1] "OREbase"
> head(aggdata)
Group.
1
1 ABE
2 ABI
3 ABQ
4 ABY
5 ACK
6 ACT
x
237
34
135
7
10
3
33
DatabaseServer
reserved.

R: Transparency through function
overloading
Invoke in-database Data Mining model (Support Vector
Machine)
CUS
T
In-db
Minin
g
Model
OracleDatabase
Oracle PL/SQL
BEGIN
DBMS_DATA_MINING.CREATE_
MODEL(
model_name =>’SVM_MOD’,
mining_function =>
dbms_data_mining.classification
...
Oracle Advanced
Analytics
ORE Client
Packages
Transparency
Layer
> svm_mod <-
ore.odmSVM(BUY~INCOME+YRS_CUST+MARITAL_STATUS,data=CUST,
"classification", kernel="linear")
> summary(svm_mod)
Call:
ore.odmSVM(formula = BUY ~ INCOME + YRS_CUST + MARITAL_STATUS, data =
CUST,
type = "classification", kernel.function = "linear")
Settings:
prep.auto
value
on
Coefficients:
class
1 0
reserved.
variable value
INCOME
estimate
5.204561e-
05
active.learning al.enable 2 0 MARITAL_STA
TUS
M -4.531359e-05
complexity.factor 46.044899 3 0 MARITAL_STA
TUS
S 4.531359e-05
conv.tolerance 1e-04 4 0 YRS_CUST 1.264948e-04
kernel.function linear 5 0 (Intercept) 9.999269e-01
6 1 INCOME 2.032340e-05
7 1 MARITAL_STA
TUS
M 2.636552e-06
8 1 MARITAL_STA
TUS
S -2.636555e-06
9 1 YRS_CUST -1.588211e-04
10 1 (Intercept) -9.999324e-01
DatabaseServer

reserved.
53

reserved.
54

reserved.
55

reserved.
56

reserved.
57

reserved. |
Oracle Advanced Analyticsfor Hadoop
Predictivealgorithmsthat execute in a parallel/distributed manner onHadoopwith
data in HDFS

HadoopCluster
with Oracle R Advanced Analytics for Hadoop (ORAAH)
Oracle R Advanced Analytics for Hadoop
Using Hadoop and HIVE Integration, plus R Engine and Open-Source R
Packages
R Analytics
Oracle R Advanced
Analytics for Hadoop
RClient
SQL
Developer
Other SQL
Apps
SQLClient
HQ
L
Oracle Database
with Advanced Analytics
option
reserved.
R
R interface to HQL Basic
Statistics,
Data Prep, Joins and View
creation
Parallel, distributed algorithms:
MLP Neural Nets*, GLM*, LM,
PCA, k-Means, NMF, LMF
* Spark-Caching enabled
Use of Open-source R packages via
custom R Mappers / Reducers

Oracle R Advanced Analytics for Hadoop
AdvancedAnalyticsalgorithmsin a HadoopCluster:Map-ReduceandSparkbased
Generalized Linear
Model
Logistic
Regression
Regression
Linear Regression
Multi-Layer Neural Networks
Classification
Attribute Importance
Principal Components
Analysis
Clustering
Hierarchical k-Means
Feature Extraction
Nonnegative Matrix
Fact(NMF) Collaborative
Filtering (LMF)
StatisticalFunctions
reserved.
Correlation
Covariance
Cross Tabulation
Summary
statistics

Copyright © 2016 Oracle and/or its affiliates. All
rights r
eserved
.
ORAAH: Machine Learning in Spark against
HDFS data
InvokeORAAHcustomparallel distributed GLMModel usingSparkCaching
> Connects to Spark
> spark.connect("yarn-client",memory="24g")
> # Attaches the HDFS file for use withinR
> ont1bi <- hdfs.attach("/user/oracle/ontime_1bi")
> # Formula definition: Cancelled flights (0 or 1) based on otherattributes
> form_oraah_glm2 <- CANCELLED ~ DISTANCE + ORIGIN + DEST + F(YEAR) +
F(MONTH)+
+ F(DAYOFMONTH) +
F(DAYOFWEEK)
> system.time(m_spark_glm <- orch.glm2(formula=form_oraah_glm2, ont1bi))
ORCH GLM: processed 6 factor variables, 25.806sec
ORCH GLM: created model matrix, 100128 partitions, 32.871sec
ORC
H
GLM: iter 1, deviance 1.38433414089348300E+09, elapsed time 9.582 sec
ORC
H
ORC
H
ORC
H
ORC
H
ORC
H
user system elapsed
84.107 5.606 143.591
Oracle Distribution of R version 3.1.1 (--) -- "Sock it to
Me"
YARN:
Apache
Spark Job
4
2
Custom Spark Java
Algorithm
distributed in-Memory
Computation
Custom Spark Java
Algorithm
distributed in-Memory
Computation
/user/oracle/ontime_
s
3
Oracle R Advanced
Analytics
for Hadoop Client
Packages
Spark-Based
Machine
Learning
algorithms
module
64
1
5

reserved. |
Big Data SQL
PushdownSQLpredictsto storagelayers

reserved.
What gives Exadata extreme
performance?
68
Oracle Database
12c
SQ
L
Small data subset
quickly returned
Offload Query to
Exadata Storage
Servers
Hadoop &
NoSQL

reserved.
69
Introducing Oracle Big Data SQL
MassivelyParallel SQLQueryacrossOracle,Hadoopand NoSQL
Oracle Database
12c
Offload Query to
Exadata Storage
Servers
Hadoop &
NoSQL
Offload Query
to Data
Nodes
data
subse
t
SQL
SQL
Small data
subset quickly
returned

reserved.
Manage and Analyze All Data—SQL & Oracle Big
Data SQL
70
StructuredandUnstructured Data Reservoir
• JSONdata
• HDFS/ Hive
• NoSQL
• Spatial andGraph data
• ImageandVideo data
• SocialMedia
JSON
Oracle Database 12c
Oracle Big Data Appliance
SQL/ R
Data analyzedvia SQL/ R/ GUI
• RClients
• SQLClients
• OracleData Miner
Storebusiness-criticaldata in Oracle
• Customer data
• Transactionaldata
• Unstructureddocuments, comments
• SpatialandGraph data
• ImageandVideo data
• SocialMedia

More Data Variety—Better Predictive
Models
• Increasing sources of
relevant data can
boost model
accuracy
Naïve Guess
or Random
100%
0% Population
Size
Responde
rs
Model with 20
variables
Model with 75
variables
Model with 250
variables
Model with “Big Data” and
hundreds -- thousands of
input variables including:
• Demographic data
• PurchasePOStransactional
data
• “Unstructured data”, text &
comments
• Spatial location data
• Longterm vs.recent historical
behavior
• Webvisits
• Sensordata
• etc.
100%
reserved.

Wargaming Creates Complex Analytical Models in
Minutes, Ensures Superior Gaming Experience
Objectives
 Online game developer, publisher, and leader in free-to-
play massively multiplayer online game market
 Implement a scalable and flexible data warehouse
 Ensured optimal game play for more than 110 million
players
 Expand insight into game health, market health, and
business
S
o
h
l
u
e
t
a
i
l
o
t
h
n
 Ran predicative analytics model in just three minutes
instead
of more than six hours
 Enabled developers to adapt game, based on where
players experience most exciting play
 Expanded insight with increased intelligence on
customer acquisition and loyalty as well as
identifying potential monetization opportunities
Customer Snapshot: http://www.oracle.com/us/corporate/customers/customersearch/wargaming-1-bda-ss-
2408474.html
 “Oracle Big Data Appliance gives us unprecedented
insight into game, business, and market health. The
possibilities are endless with this highly extensible
solution that enables us to gather, analyze, and use data,
including social-media data, in ways that were simply not
possible before.”
– Craig Fryar, Head of Wargaming Business Intelligence

Oracle Big Data Appliance, Oracle Database
Appliance,
Oracle Advanced Analytics, Oracle Big Data
Connectors
reserved.

reserved. |
PredictiveApplications+OBIEE Integration

Enabling “Predictive” EnterpriseApplications
Oracle Applications Using Oracle Advanced Analytics—
Partial List
• OracleHCM Fusion
– Employee turnover and
performance
prediction and “What if?” analysis
• OracleCRMFusion
– Prediction of sales
opportunities, what to sell,
amount, timing, etc.
• OracleIndustryData Models
– CommunicationsData Model churn
prediction, segmentation, profiling,
etc.
– Retail Data Model loyalty and
market basket analysis
– Airline Data Model analysis frequent
flyers, loyalty, etc.
– Utilities Data Model customer
churn, cross-sell, loyalty, etc.
• OracleRetail CustomerAnalytics
– ”Shopping cart analysis” and next bestoffers
• OracleCustomer Support
– Predictive Incident Monitoring
(PIM)
• OracleSpendClassification
– Real-time and batch
flagging of noncompliance
and anomalies in expense
submissions
• OracleFinServAnalytic Applications
– Customer Insight, Enterprise Risk
Management,
Enterprise Performance, Financial
Crime and Compliance
• OracleAdaptiveAccessManager
– Real-time security and fraud analytics
reserved.

Integrated Business
Intelligence
Enhance Dashboards with Predictions and Data Mining
Insights
• In-database
predictive models
“mine” customer
data and predict
their behavior
• OBIEE’sintegrated
spatial mapping
shows location
• All OAA results and
predictions
available in
Database via
OBIEE Admin to
enhance
dashboards
Oracle Data Mining
results available to
Oracle BI EE
administrators
Oracle BI EE defines
results
for end user
presentation
reserved.

Pre-BuiltPredictive Models
• Fastest Way to Deliver Scalable
Enterprise-wide Predictive
Analytics
• OAA’sclustering and predictions
available in-DB for OBIEE
• Automatic Customer
Segmentation, Churn
Predictions, and Sentiment
Analysis
Oracle Communications Industry Data
Model
Example Predictive Analytics Application
reserved.

Oracle Communications Data
Model
Pre-Built Data Mining
Models
1.Churn Prediction
2.Customer Profiling
3.Customer Churn Factor
4.Cross-Sell Opportunity
5.Customer Life Time Value
6.Customer Sentiment
7.Customer Life Time Value
reserved.

Oracle Communications Industry Data
Model
Predictive Analytics
Applications
OCDMTelcoChurnEnhancedby
SNAAnalysis
• Integrated with OCDM, OBIEE,
and leverages Oracle Data
Mining with specialized SNA
code
• Identification of social
network communities from
CDR data
• Predictive scores for churn and
influence at a node level, as
well as potential revenue/value
at risk
• User interface targeted at
business users and flexible ad-
hoc reporting Copyright © 2016 Oracle and/or its affiliates. All rights
reserved.

Fusion HCM Predictive
Workforce
Predictive Analytics
Applications
FusionHumanCapitalManagement
PoweredbyOAA
• Oracle Advanced Analytics
factory- installed predictive
analytics
• Employees likely to leave and
predicted performance
• Top reasons, expected behavior
• Real-time "What if?" analysis
reserved.

Getting
starte
d
reserved. |

reserved.
OAA Links and
Resources
• OracleAdvancedAnalytics Overview:
– OAApresentation—Big Data Analytics in Oracle Database 12c With OracleAdvancedAnalytics & Big Data
SQL
– BigDataAnalytics with OracleAdvancedAnalytics: Making BigData andAnalytics Simplewhite paper on OTN
– Oracle Internal OAA Product Management Wiki and Workspace
• YouTube recordedOAAPresentations and Demos:
– OracleAdvancedAnalytics andData Mining at the YouTube Movies
(6 +OAA“live” Demoson ODM’r 4.0 New Features, Retail, Fraud, Loyalty, Overview, etc.)
• Getting Started:
– Link to Getting Started w/ ODM blog entry
– Link to New OAA/Oracle Data Mining 2-Day Instructor Led Oracle University course.
– Link to OAA/Oracle Data Mining 4.0 OraclebyExamples(free) Tutorialson OTN
– Take a Free Test Drive of Oracle Advanced Analytics (Oracle Data Miner GUI) on the Amazon Cloud
– Link to OAA/Oracle REnterprise(free) Tutorial Serieson OTN
• Additional Resources:
– Oracle Advanced Analytics Option on OTN page
– OAA/Oracle Data Mining onOTNpage, ODM Documentation & ODM Blog
– OAA/Oracle REnterprisepageonOTNpage, ORE Documentation & ORE Blog
– Oracle SQL based Basic Statistical functions on OTN
– BIWASummit’16, Jan26-28, 2016 – Oracle Big Data & Analytics User Conference @ Oracle HQ Conference
Center

Books on Oracle
Advanced
Analytics
L
Bookavailable onAmazon
PredictiveAnalyticsUsingOracleData
Miner: Developfor ODM in SQL&PL/SQ
Copyright © 2016 Oracle and/or its affiliates. All rights reserved. | Oracle Confidential – Internal/Restricted/Highly
Restricted
83
Bookavailable onAmazon
UsingRto Unlockthe Valueof BigData

• Hands-on-Labs
• Customerstories,told bythe customers
• EducationalsessionsbyPractitionersandDirect from Developers
• OracleKeynote presentations
• Presentations covering:AdvancedAnalytics, BigData, BusinessIntelligence,
Cloud,Data WarehousingandIntegration, SpatialandGraph, SQL
• Networking with productmanagementanddevelopmentprofessionals
reserved.

reserved.

oracleadvancedanalyticsv2otn-2859525.pptx

oracleadvancedanalyticsv2otn-2859525.pptx

Recomendados

Recomendados

Mais conteúdo relacionado

Semelhante a oracleadvancedanalyticsv2otn-2859525.pptx

Semelhante a oracleadvancedanalyticsv2otn-2859525.pptx (20)

Mais de AdityaDas899782

Mais de AdityaDas899782 (12)

Último

Último (20)

oracleadvancedanalyticsv2otn-2859525.pptx