Top Banner
Machine Learning Smackdown @LynnLangit
21

Machine Learning Smackdown for Microsoft Excel

Jun 14, 2015

Download

Technology

Lynn Langit

Machine Learning on the Microsoft stack (Excel) for Oredev conference 2014
Welcome message from author
This document is posted to help you gain knowledge. Please leave a comment to let me know what you think about it! Share it to your friends and learn new things together.
Transcript
Page 1: Machine Learning Smackdown for Microsoft Excel

Machine Learning

Smackdown

@LynnLangit

Page 2: Machine Learning Smackdown for Microsoft Excel

Agenda

Definitions

On premise solutions

3rd party Excel

Machine Learning Add-

ins

Microsoft SQL Server Data Mining Add-

ins

R Studio

Cloud solutions

Predixion Software

Azure Machine Learning

Page 3: Machine Learning Smackdown for Microsoft Excel

Analytics Defined• Business Analytics - deterministic

Query Aggregation

• Predictive Analytics - probabilistic Machine Learning

Statistics

Unsupervised Data Mining

Supervised Data Mining

Other

Page 4: Machine Learning Smackdown for Microsoft Excel

Machine Learning Roles Defined

Data Scientist

Store

Clean

Aggregate

ML Engineer

Selects Libraries

Applies Algorithms

Creates Solutions

ML ResearcherCreates Algorithms

Page 5: Machine Learning Smackdown for Microsoft Excel

ML Developer Learning Path Defined

Learn a ML

language

Pick your IDE

Pick a problem space

Get Data

Process and

ITERATE

Visualize results

Page 6: Machine Learning Smackdown for Microsoft Excel

What is the R Language?

Page 7: Machine Learning Smackdown for Microsoft Excel
Page 8: Machine Learning Smackdown for Microsoft Excel

R Language Semantics

search() and ls() # lists packages and objects in scope

?mean # shows function definition

Vectors (numeric, logical, character), lists, NULLs

Data Frame, Matrix (same types), Factors (Categorical)

meanx <- mean(x) or meanx = mean(x) # assignment

x[1] <- 9 # extracts and/or changes pieces

print(x) or x # prints x

plot(x) # graphs x

Page 9: Machine Learning Smackdown for Microsoft Excel

3rd party Excel Machine Learning Add-ins

• XLMiner

• StatsMiner

• XLStat

• RExcel

Important: All of these tools assume expert statistical knowledge

Page 10: Machine Learning Smackdown for Microsoft Excel

Add-ins - XLMiner

Page 11: Machine Learning Smackdown for Microsoft Excel

Excel Machine Learning Add-ins

• Microsoft SQL Server Analysis Services Data Mining Add-ins

• Predixion Add-in for Excel

Important: All of these tools require a local SQL Server SSAS instance

Page 12: Machine Learning Smackdown for Microsoft Excel

Data Mining Add-ins For Excel

• Table Analysis Tools for Excel Use mining models with Excel data or external data

• Data Mining Client for Excel Create/test/explore/manage Mining Models

• Data Mining Templates for Visio Render/share mining models as Visio Drawings

Important: Use requires connection to SQL Server 2012 SSAS

Page 13: Machine Learning Smackdown for Microsoft Excel

Data Mining Add-ins for Excel

Page 14: Machine Learning Smackdown for Microsoft Excel

Data Mining Structures

• Containers • Cleansed source data

• One+ SSAS Algorithm(s)• Clustering

• Time Series Prediction

• Market-Basket Analysis

• Text Mining

• Neural Networks

• Models • Query

• Model processing

Page 15: Machine Learning Smackdown for Microsoft Excel

About Predixion SoftwareSuite of tools for predictive analytics

Insight Now

Use mining models with Excel data or external data

Insight Analytics

Create/test/explore/manage

Mining Models

Insight Workbench

Prepare data for model creation

Web-based Viewers and

Tools

HTML 5

Important: Runs as EITHER connected to SSAS on premise OR Connected to Predixion’s cloud-based servers

Page 16: Machine Learning Smackdown for Microsoft Excel

Predixion Software

Page 17: Machine Learning Smackdown for Microsoft Excel

17

Page 18: Machine Learning Smackdown for Microsoft Excel

Azure MLCloud-based SaaS service

Create ML Experiments using Datasets

Can publish results as Web Services

Page 19: Machine Learning Smackdown for Microsoft Excel

Azure ML

Page 20: Machine Learning Smackdown for Microsoft Excel

Understanding options…

Add-inServer Required

Complexity of install

OtherCost of Add-in

Cost of Solution

XLMiner none easy Assumes stats expertise $$ $$

RExcel none easy Assumes R expertise $ $

Data Mining Add-ins

SQL Server SSAS

medium Designed for single user 0 $$$

Predixion on premise

SQL Express easy Requires local R install 0 $$-$$$

Predixion on premise

SQL Server SSAS

medium Your data is stored locally

0 $$$$

Predixion cloud none easy Supports SSAS Data Mining AND R Language

0 $$-$$$

Azure Machine Learning

none easy Rich set of algorithms and supports R

n/a unknown

Page 21: Machine Learning Smackdown for Microsoft Excel

@LynnLangit