The document outlines an agenda for an analytics induction covering foundations of analytics including definitions and history, the lifecycle of analytics projects including strategy, management, and visualization, tools and techniques such as linear regression and decision trees, and provides a case study example. Key aspects of analytics projects are described including identifying opportunities, acquiring and cleaning data, modeling, evaluation, and deployment.

1. Analytics
Induction
Ankit Rathi

2. Breaking
the ice…

4. AGENDA
A. Foundation
B. Lifecycle
C. Tools & Techniques
D. Case Study
And some activities in between…

5. Let me start with
a story…

6. But, why my story?

7. Story I (2008-09)

8. 8
Story II (2012-13)

9. A. Foundation

10. Why Analytics
is required?

12. Analytics
Growth

14. Analytics in
Industry

16. What is
Analytics?

17. Analytics is the discovery, interpretation,
and communication of meaningful patterns
in data. Especially valuable in areas rich
with recorded information, analytics relies
on the simultaneous application of
statistics, computer programming and
operations research to quantify
performance.

18. History of
Analytics

20. Types of
Analytics

22. Activity
Time - I

23. Identify use-cases in my business to apply:
• Descriptive Analytics
• Diagnostic Analytics
• Predictive Analytics
• Prescriptive Analytics
• Cognitive Analytics
AR Pizza Chain

24. B. Lifecycle

25. Types of
Analytics
Projects

26. Strategy
Management
Consulting
Visualization

27. Analytics
Strategy

29. Analytics
Consulting

31. Analytics
Management

32. Appropriate management methodologies
Unavailability of past metrics
The value of a PoC
Achieving repeatability
Agile analytics
Management communication

33. Analytics
Virtualization

34. with Advanced
Technologies
with Internet 0f
Things
with Blockchain
with Artificial
Intelligence

35. Analytics
Visualization

36. When to use what?

37. Pie Chart
Bar Chart
Scatterplot Violin Plots
Histograms
Box-Whisker Plots
Line Chart Bubble Chart
Heat Map Tree Map

38. Pie
Chart

39. Bar
Chart

40. Histograms

41. Box & Whisker

42. Scatterplots

43. Line Chart

44. Bubble
Chart

45. Violin
Plots

46. Analytics
Integration

47. 47
Dashboards
Analytics Workbench
Data Engineers
Data Scientists
Analytics Architect
Structured Data
Semi/Unstructured Data
batch
Streaming
Ingestion &
Staging Layer
Transformation Layer Consumption Layer
APIs
Business
Sqoop
Storm
Flume
Kafka
Hadoop
Spark
UI
Tableau
PowerBI
MapReduce
SparkSQL
MongoDB
Cassandra
SparkR
PySpark

48. How to do
Analytics
Projects?

50. Identifying
Acquiring
Cleaning
Framing
Understanding
Modeling
Evaluating
Presenting
Deploying
Monitoring

51. Identifying Analytics Opportunity

52. Framing the Problem

53. Determine Your Goals and Objectives
Build Your Proof of Concept Team
Obtain Sample Data Set
Select the predictive model
Analyse POC Results
Stakeholders Buy-in
Conducting the
POC

54. Identify Data Sources
Filter Data Attributes
Build Ingestion Pipeline
Acquiring
the
Data

55. Identify Data
Types
Describe
Data
Explore
Data
Verify Data
Quality
Understanding the Data

56. Fill Missing
Data
Fill Missing
Data
Treat
Outliers
Standardize
Data
Normalize
Data
Cleaning
the
Data

57. Feature Engineering Select the Model Apply the Model
Building the Model

58. Identify Evaluation
Metrices
Prepare Cross-validation
Sets
Evaluate the
Model
Evaluating
the Model

59. Engage the
Audience
Be Creative
Know the
Audience
Craft a
Data Story
Presenting
the Insights

60. Model
Scoring
Performance
Monitoring
Deploying the Model

61. Data
Architecture

63. DB, ETL,
DWH & BI

65. Case
Studies

66. Associative Mining
in Retail Banking
Anomaly Detection
in AML
Automation in
BPM
Flight Prediction

67. Activity
Time - II

68. Build a high-level conceptual architecture for
my business to apply analytics:
• Operational System
• ETL Pipelines
• Data Warehouse/Data Lake
• Business Intelligence/Data Science
AR Pizza Chain

69. C. Tools &
Techniques

70. Techniques

72. Linear Regression
Logistic Regression
Decision Trees
Naïve Bayes
Associative Mining
Anomaly Detection

73. Linear Regression

74. Logistic Regression

75. Decision Trees

76. Naïve Bayes

77. Associative Mining

78. Clustering

79. Anomaly Detection

80. Text Analytics

81. Tools

82. Business Layer

83. Data Analytics
Tools

85. Data Ingestion
Tools

87. Data
Transformation
Tools

89. Data
Presentation
Tools

91. D. Case Study

92. • Pick an existing business process
• Identify analytics opportunity
• Hypothesize transformed
business process
• Present the business case