Course Agenda

CT-583 •
•
Introduction: What is Data Science?
Statistical Inference
• Exploratory Data Analysis and the Data Science Process
Tools and Techniques for Data Science • Basic Machine Learning Algorithms
• One More Machine Learning Algorithm and Usage in Applications
The power of data lies in understanding the story it tells • Feature Generation and Feature Selection (Extracting Meaning From
Data)
• Recommendation Systems: Building a User-Facing Data Product
• Mining Social-Network Graphs
• Data Visualization
• Data Science and Ethical Issues
1 2

1 2

Tools & Techniques

• Tool
• Specific item to perform a task/activity.
• Example: software OR ……

• Technique
Tools and Techniques • Broader method or approach applied to perform tasks
• Example: …….

3 4

3 4

Tools & Techniques

Tools Techniques
• These are instruments or • A technique is a specific
devices used to perform a task approach to efficiently
• Example: ruler, thermometer, or accomplish a task in a manner
microscope or software
programs etc
• Techniques involve the
application of systematic
procedures, approaches, or
Data Science
strategies.
• Example: statistical analysis,
content analysis etc

5 6

5 6

1
 Data

Wisdom
Knowledge

DATA
Information

Data

7 8

7 8

DIKW Model
Data, Information, Knowledge, Knowledge (DIKW)

9 10

9 10

11 12

11 12

2
 DIKW Model

Data Information Knowledge Wisdom

13 14

13 14

DIKW Model OR DIKW Pyramid

Daily transactions

15 16

15 16

Highly Summarized
Further summarized information Executive Data/information
Management

• Summarize the Information

• Analyze information that
helps in Decision Making
Senior Management

• Manage the Workforce

• Data Summarized to form
Information
Supervisor

• Daily Transactions
• Routine Task
• Generate Data
Workforce Detailed Data /
Information

17 18

17 18

3
 Use the knowledge from the
analyzed information to take Executive External Focused
decisions Management

Use Information to take

Types of Information and
Information Systems
Senior Management
decisions / analyze Data to
identify problems/new trends

Using Data Generate Supervisor

Information

Generating Data Workforce More internal focused

Human Marketing Finance Operations

19 20
Resource

19 20

Strategic Decision based on

Highly summarized information

Strategic HIGHLY SUMMARIZED

More External Information System Users
INFORMATION
Executive Support Systems
(ESS) Level
• First users are clerical users on TPSs.
Decision Support Systems
Tactical SUMMARIZED
INFORMATION
(DSS) Level • Managerial Levels
• Strategic planning level
DETAILED • Management control level
Management Information SystemsOperational Level INFORMATION
• Operational control level
(MIS)

DATA /
Transaction Processing
Systems (TPS)
Workforce Level Highly detailed More Internal
information
Generate data through
Daily transactions
21 22

21 22

Management Levels Managers

Can Influence Both the Source and Presentation Form of Information Can Be Found on All Business Areas of the Firm

23 24

23 24

4
 Management Level May Influence the Relative Emphasis on the
Management Function Levels of Managerial Decision Making

25 26

25 26

Information Quality Attributes of Information Quality

• Information products made more valuable by their attributes,

characteristics, or qualities
• Information that is outdated, inaccurate, or
hard to understand has much less value

• Information has three dimensions

• Time
• Content
• Form

27
28

27 28

Decision Structure Decision Support Systems

• Structured (operational level) Management Information
Systems
Decision Support
Systems
• The procedures to follow when decision Decision Provide information about the Provide information and
is needed can be specified in advance support performance of the organization techniques to analyze
provided specific problems

• Unstructured (strategic level) Information form

and frequency
Periodic, exception, demand,
and push reports and
Interactive inquiries and
responses
• It is not possible to specify in advance responses
most of the decision procedures to follow Information Prespecified, fixed format Ad hoc, flexible, and
format adaptable format

• Semi-structured (tactical level) Information

processing
Information produced by
extraction and manipulation of
Information produced by
analytical modeling of
• Decision procedures can be pre-specified, methodology business data business data
but not enough to lead to the correct decision

29 30

29 30

5
 What is Data Science
• Study of “Data” OR Science of “Data”

• It is the study of data to extract meaningful insights for business.

• Data science combines math and statistics, specialized programming,

Data Science advanced analytics, artificial intelligence (AI) and machine
learning with specific subject matter expertise to analyze large
amounts of data.

• It helps to uncover actionable insights hidden in an organization’s data.

• These insights can be used to guide decision making and strategic planning.

31 32

31 32

Data Scientist Data Cycle

• Responsible for deriving sensible outcomes from large data sets.
Communicate Capture
• enabling a business to make the right decision

Analyze Maintain

Process

33 34

33 34

Data Cycle Data Science Cycle

1.Capture: 4. Analyze:
• This stage involves gathering raw structured and unstructured data.
• This stage involves performing the various analyses on the data.
• Example; Data Acquisition, Data Entry, Signal Reception, Data Extraction.
• Example; Exploratory/Confirmatory, Predictive Analysis, Regression,
Text Mining, Qualitative Analysis. Here is the real meat of the lifecycle.
2.Maintain:
• This stage covers taking the raw data and putting it in a form that can be used.
• Example; Data Warehousing, Data Cleansing, Data Staging, Data Processing, Data Architecture. 5.Communicate:
• In this final step, analysts prepare the analyses in easily readable forms
3.Process: such as charts, graphs, and reports.
• Data scientists take the prepared data and examine its patterns, ranges, and biases to determine • Example; Data Reporting, Data Visualization, Business Intelligence,
how useful it will be in predictive analysis
Decision Making.
• Example; Data Mining, Clustering/Classification, Data Modeling, Data Summarization.

35 36

35 36

6
 Data Science Process Roles
• Data Collection Data Data
Exploratory
Model Model
Data
• Data Cleaning Collection Cleaning
Analysis
Building Deployment

• Exploratory Data Analysis

• Model Building Data Engineers
• Model Deployment
Data Analyst Machine Learning
Engineers

DATA SCIENTISTS

37 38

37 38

Data Science Cycle Data Science Cycle

• Problem identification; • Data preparation;
• Understanding the Business – Clarify the problem and goals. • Preparing the Data – Cleaning and formatting for analysis.
• Defining the Problem – Specify the precise goals the data science
project aims to accomplish. • Model Building
• Building Models – Applying analytics like machine learning to get
insights.
• Getting the Data;
• Identify data requirement • Model testing / evaluation
• Identify source of data • Evaluating Models – Testing to pick the best performers
• Gather relevant data from different sources.
• Model deployment
• Deploying the Solution – Launching it for business use.
39 40

39 40

Questions

41

41