An Overview of Predictive Analytics - MachinePulse

1
Predictive Analytics - An overview

• Vijaykumar Adamapure
• MachinePulse.

2
Agenda

• Introduction to Big Data.
• What is Analytics?
• Overview of Predictive Analytics Techniques.
• Business Applications of Predictive Analytics.
• Predictive Analytics Tools in Market.

3
Gartner Hype Cycle
4
Things That Happen On Internet Every Sixty Seconds
5
Things That Happen Every Sixty Seconds
6
The 5 V's of Big Data

• Big data is high-volume, high-velocity and
  high-variety information assets that demand
  cost-effective, innovative forms of information
  processing for enhanced insight and decision
  making.

7
Survey on Big Data Adoption Stages
8
What is Analytics?
9
Data Analysis OSEMN Process

• OSEMN is an acronym that rhymes with awesome

Obtain Data
Scrub Data
Explore Data
Model Data
iNterpret Results
10
What is Predictive Analytics?

• Predictive analytics is the practice of
  extracting insights from the existing data set
  with the help data mining, statistical modeling
  and machine learning techniques and using it to
  predict unobserved/unknown events.
• Identifying cause-effect relationships across the
  variables from the historical data.
• Discovering hidden insights and patterns with the
  help of data mining techniques.
• Apply observed patterns to unknowns in the Past,
  Present or Future.

11
Predictive Analytics Process Cycle
12
Common Predictive Analytics Methods

• Regression
• Predicting output variable using its
  cause-effect relationship with input variables.
  OLS Regression, GLM, Random forests, ANN etc.
• Classification
• Predicting the item class. Decision Tree,
  Logistic Regression, ANN, SVM, Naïve Bayes
  classifier etc.
• Time Series Forecasting
• Predicting future time events given past
  history. AR, MA, ARIMA, Triple Exponential
  Smoothing, Holt-Winters etc.

13
Common Predictive Analytics Methods (Contd.)

• Association rule mining
• Mining items occurring together. Apriori
  Algorithm.
• Clustering
• Finding natural groups or clusters in the data.
  K-means, Hierarchical, Spectral, Density based EM
  algorithm Clustering etc.
• Text mining
• Model and structure the information content of
  textual sources. Sentiment Analysis, NLP

14
Evaluating Predictive Models

• Need to check predictive models out of sample
  performance.
• Model Assessment Hit Rate, Gini Coefficient, K-S
  Chart, Confusion Matrix, ROC Curve, Lift Chart,
  Gain Chart etc.

15
Business Applications of Predictive Analytics
Renewable Energy
Multi-channel sales
Finance
Smarter Healthcare
Factory Failures
Telecom
Traffic Control
Spam Filters
Manufacturing
Trading Analytics
Fraud and Risk
Retail Churn
16
Business Applications (Contd.)

• Supply Chain
• Simulate and optimize supply chain flows to
  reduce inventory.
• Customer Profiling
• Identify high valued customers and retain their
  loyalty.
• Pricing
• Identify the optimal price which will increase
  net profit.
• Human Resources
• Best Employees selection for particular tasks at
  optimal compensation. Employee churn retention.

17
Business Applications (Contd.)

• Renewable Energy
• Energy forecasting, electricity price
  forecasting, Predictive Maintenance, Operational
  cost minimization.
• Financial Services
• Approval of credit cards/ loan applications
  based on credit scoring models, Options pricing,
  Risk analysis etc.
• E-Commerce
• Identify cross-sell and upsell opportunities,
  increase transactions size, maximize campaign's
  response based CRM data.

18
Business Applications (Contd.)

• Product Quality Control
• Detect product quality issues in advance and
  prevent them.
• Revenue Performance
• Identify key drivers of revenue generation and
  optimization of revenue.
• Fraud and Crime Detection
• Detect fraud , criminal activity, insurance
  claims, tax evasion and credit card frauds.
• HealthCare
• Identify prevalence of particular disease to a
  patient based health conditions.

19
Predictive Analytics Tools in Market
20
Thank you!Visit http//www.machinepulse.comEma
il sales_at_machinepulse.com