Oracle Database House Party_Oracle Machine Learning to Pick a Good Inexpensive Bottle of Wine

Copyright © 2020, Oracle and/or its affiliates
Using Oracle’s Converged “AI”
Database to Pick a Good but
Inexpensive Bottle of Wine
#converged-ai-database
Brendan Tierney
Oracle ACE Director,
Author, etc.
Jeff Smith
PM, SQL
Developer, REST,
etc.,
Charlie Berger
PM, Oracle
Machine Learning
Starts at 9 am ET

2 Copyright © 2020, Oracle and/or its affiliates | Confidential: Internal/Restricted/Highly Restricted [Date]
Brendan Tierney
President, Oralytics
Brendan Tierney is an Oracle ACE
Director and Ground Breaker Ambassador.
He is based in Dublin Ireland, and has
been working with Oracle products since
Oracle v5 Database, Forms 2.3, RPT, etc.
Most of the past 27 years he has been
working in the area of big data, analytics
and machine learning, with a focus on
building solutions.
About your presenters:
Charlie Berger
Sr. Director Product
Management, Machine
Learning, AI and Cognitive Analytics
Charlie has been working in the field of
machine learning, statistical and data
analysis for over 30 years. He joined Oracle
in 1999 during the acquisition of Thinking
Machines Corporation. @CharlieDataMine
@brendantierney

Jeff Smith
Master Product Manager
Jeff is the Product Manager at Oracle
for SQL Developer, SQLcl, SQL
Developer Data Modeler, and REST
Data Services (ORDS). Jeff has
worked with database tools for
business analysts, developers, and the
DBAs for 20 years.
About your presenters:
@oraclesqldev
@oraclesqlcl @oracleREST

Goal
Share an simple, fun use case that highlights the “art of the
possible” in the Oracle “Converged” Database using popular tools
and SQL.
Copyright © 2020 Oracle and/or its affiliates.

Oracle Database Converged Features
Oracle Machine Learning

Oracle Mission Statement
“Our mission is to help people see data in
new ways, discover insights, unlock
endless possibilities”
© 2020 Oracle - Portland OUG Training Day 10/22/2020Copyright © 2020 Oracle and/or its affiliates.

Using Oracle’s Converged “AI” Database to
Pick a Good but Inexpensive Bottle of Wine
Share a simple and fun use case:
SQL Developer
Import data
Oracle Data Miner
Explore data
Identify key attributes
Build predictive ML models
Apply ML models for predictions
APEX
Further investigate insights and predictions
SQLDEV/ORDS
Deploy ML Model using Oracle REST API (optional)Copyright © 2020 Oracle and/or its affiliates.

* Coming soon
Oracle Machine Learning extends
Oracle Database(s) and enables users
to build “AI” applications and
analytics dashboards
OML delivers powerful in-database
machine learning algorithms,
automated ML functionality, and
integration with open source Python
and R.
OML Services*
Model Deployment and Management,
Cognitive Image and Text
OML4SQL
SQL API
OML4Py*
Python API
OML4R
R API
OML Notebooks
with Apache Zeppelin on
Autonomous Database
OML4Spark
R API on Big Data
Oracle Data Miner
Oracle SQL Developer extension
OML AutoML UI*
Code-free interface on Autonomous Database

CLASSIFICATION
Naïve Bayes
Logistic Regression (GLM)
Decision Tree
Random Forest
Neural Network
SupportVector Machine
Explicit Semantic Analysis
CLUSTERING
Hierarchical K-Means
Hierarchical O-Cluster
Expectation Maximization (EM)
ANOMALY DETECTION
One-Class SVM
TIME SERIES
Forecasting - Exponential Smoothing
Includes popular models
e.g. Holt-Winters with trends,
seasonality, irregularity, missing data
REGRESSION
Linear Model
Generalized Linear Model
SupportVector Machine (SVM)
Stepwise Linear regression
Neural Network
ATTRIBUTE IMPORTANCE
Minimum Description Length
Principal Comp Analysis (PCA)
Unsupervised Pair-wise KL Div
CUR decomposition for row & AI
ASSOCIATION RULES
A priori/ market basket
PREDICTIVE QUERIES
Predict, cluster, detect, features
SQL ANALYTICS
SQL Windows
SQL Patterns
SQL Aggregates
Oracle Machine Learning Algorithms
FEATURE EXTRACTION
Principal Comp Analysis (PCA)
Non-negative Matrix Factorization
Singular Value Decomposition (SVD)
Explicit Semantic Analysis (ESA)
TEXT MINING SUPPORT
Algorithms support text
Tokenization and theme extraction
Explicit Semantic Analysis (ESA) for
document similarity
STATISTICAL FUNCTIONS
Basic statistics: min, max,
median, stdev, t-test, F-test, Pearson’s,
Chi-Sq, ANOVA, etc.
R & PYTHON
Third-party R & Python Packages
through Embedded Execution
Spark MLlib algorithm integration
MODEL DEPLOYMENT &
MONITORING
SQL—1st Class Objects
Oracle RESTful API (ORDS)
OML Services
X1
X2
A1 A2 A3 A4 A5 A6 A7
Includes support for Partitioned Models,Transactional data and aggregations, Unstructured
data, Geo-spatial data, Graph data. etc, * Coming soon
* Coming soon

Determine Best Wine for Under $30
Build machine learning (ML) model in ADW to predict good, inexpensive wines using in-
Database OML algorithms
BEGIN
DBMS_DATA_MINING.CREATE_MODEL(
model_name => 'Wine_CLASS_MODEL',
mining_function => dbms_data_mining.classification,
data_table_name => ''Wine_TRAIN_DATA',
case_id_column_name => 'ID',
target_column_name => 'POINTS_BIN',
settings_table_name => 'Wine_build_settings');
END;
/
Apply ML model using SQL query to predict a likely good, inexpensive wine
SELECT PREDICTION_PROBABILITY(Wine_CLASS_MODEL, 'GT_90_POINTS'
USING 25 as PRICE, ‘MALBEC' as VARIETY, ‘SPAIN' as COUNTRY)
FROM dual;
Using Machine Learning SQL Model Build & Model Prediction
14

CRISP-DM Methodology
Six Major Steps
https://en.wikipedia.org/wiki/Cross-industry_standard_process_for_data_miningCopyright © 2020 Oracle and/or its affiliates.
DATA UNDERSTANDING
DATA PREPARATION
MODELING
EVALUATION
DEPLOYMENT
BUSINESS UNDERSTANDING

Six Major Steps
DATA UNDERSTANDING
Assemble the “right data”
Data profiling
• Data visualization
• Univariate statistics/group by
• Bi-variate statistics
DATA PREPARATION
Sampling/Stratified
Algorithm req’d transforms
• Auto Data Preparation
• MissingValues, Binning, Normalization,
etc.
• Unstructured data
• Aggregations
Domain specific transforms
• “Engineered Features”
Features Selection
MODELING
Algorithm settings/defaults
• Stratified sampling
• Feature selection
• Build model(s)
EVALUATION
Model evaluation
Model comparison
Model selection
DEPLOYMENT
In-DB ML model apply
• Real-time ML apply
• In-database, REST
Embed methodology
• Applications
• Dashboards
Well-defined
business problem

Six Major Steps
DATA UNDERSTANDING
Assemble the “right data”
Data profiling
• Data visualization
• Univariate statistics/group by
• Bi-variate statistics
DATA PREPARATION
Sampling/Stratified
Algorithm req’d transforms
• Auto Data Preparation
• MissingValues, Binning, Normalization,
etc.
• Unstructured data
• Aggregations
Domain specific transforms
• “Engineered Features”
Features Selection
MODELING
Algorithm settings/defaults
• Stratified sampling
• Feature selection
• Build model(s)
EVALUATION
Model evaluation
Model comparison
Model selection
DEPLOYMENT
In-DB ML model apply
• Real-time ML apply
• In-database, REST
Embed methodology
• Applications
• Dashboards
Well-defined
business problem
*Automated and/or system defaults

ML Model Build
ML Model Apply
Prediction Probabilities

Wait, there is more!

Key Features:
• Collaborative UI for data
scientist and analysts
• Packaged with Autonomous
Databases
• Quick start Example notebooks
• Easy access to shared notebooks,
templates, permissions, scheduler,
etc.
• OML4SQL
• OML4Py coming soon
• Supports deployment of OML
models
Machine Learning Notebooks included in Autonomous Databases

Oracle APEX

Deployment using APEX
Oracle APEX
to query and
distribute
insights and
predictions
Apply the Models to Predict “BestWines”
Predictions

Real-time Wine Recommendation App +
OpenDiningTable App
3

ML Model Deployment via ORDS REST API
Launch Development APEX

Launch RESTful Services

Helpful example templates
provided

RESTful API for calling
OML model to make predictions

Congratulations!
You are now an
“Oracle Data Scientist”!
Data Scientist

Relevant Resources/Links
Oracle Machine Learning Blog with 3 Hands-on Labs
https://blogs.oracle.com/machinelearning/
• Hands-0n Lab: How to Pick a Good Wine for $30< using Oracle
Autonomous Database, Oracle Machine Learning, APEX,
Oracle Analytics Cloud and REST Services
• Learn How to Use Oracle Data Miner UI in 45 Minutes
• Hands-on Lab: Learn to Use Oracle Machine Learning Notebooks
OML on Autonomous Database Workshop
https://oracle.github.io/learning-library/workshops/adwc4dev/?lab=introduction
Oracle Machine Learning Overview:
From Oracle Data Professional to Oracle Data Scientist!
https://www.youtube.com/watch?time_continue=1&v=
jFBMhOapGL8&feature=emb_logo

Relevant Resources/Links
Google: Oracle DatabaseVM
https://www.oracle.com/downloads/developer-vm/community-downloads.html
41

For More Information
Google: Oracle SQL Developer
Oracle.com/SQLDeveloper
42

Google: Oracle REST Services (ORDS)
https://www.oracle.com/database/technologies/appdev/rest.html
43

Google: Oracle Machine Learning
Oracle.com/MachineLearning
44

ThankYou
Charlie Berger, charlie.berger@oracle.com
@CharlieDataMine
Jeff Smith, jeff.d.smith@oracle.com, @oraclesqldev
@oraclesqlcl @oracleREST
Brendan Tierney, brendan.tierney@oralytics.com,
@brendantierney

Coming up at 10am ET…
Spatial Technologies -
@Home and Everywhere
Else on the Map
with
David Lapp &
Tim Vlamis
Break time!
Remember to take the Dev Gym
quizzes to receive your certificate
of completion.
Also, please complete the
feedback form for the previous
session by 10:15am ET
… links in the Zoom chat

Oracle Database House Party_Oracle Machine Learning to Pick a Good Inexpensive Bottle of Wine

More Related Content

What's hot

Similar to Oracle Database House Party_Oracle Machine Learning to Pick a Good Inexpensive Bottle of Wine

Recently uploaded

Oracle Database House Party_Oracle Machine Learning to Pick a Good Inexpensive Bottle of Wine

Editor's Notes