SUMMARY: INTRODUCTION TO DATA ANALYTICS

SESSION OVERVIEW-

By the end of this session, students will be able to:

1. Understand data analytics and its use in the real world.
2. Understand career scope and opportunities in the data domain.
3. Distinguish between different types of data and data analytics.

KEY TOPICS AND EXAMPLES:

1. Understand data analytics and its use in the real world.

● What is data analytics?
Data analytics refers to analyzing, interpreting and deriving insights from raw data to help
decision making, solving problems and optimize outcomes.

● Now with the help of real time examples, explanations on data analytics will help the
students to have a better understanding of data analytics.
○ Example-
1. e-commerce product recommendation.
We'll look at an example here of how data analytics are used by e-commerce
platforms to provide consumers with customized product recommendations. We will
go over the fundamental ideas, methods that underpin recommendation systems.
Imagine you are a user browsing for a new laptop. Based on the past purchases and
present browsing history, the platform recommends different laptops with the same
specifications, price range, etc. These recommendations are created with the help of
past data fed to an algorithm.
2. Uber finding the nearest driver.
3. Google maps showing the shortest route, amount of traffic and estimation of time
interval between source and destination.

4. Risk detection:

1
 In earlier times, amid financial challenges, organizations turned to data analytics as a
tool for addressing their predicaments. With access to extensive customer data,
including transaction histories and demographic details, they employed systematic
data analysis techniques. By carefully dissecting this data and scrutinizing patterns,
such as recent spending behaviors and customer profiles, they could discern
indicators suggestive of potential defaults or fraudulent activities. Through this
methodical approach, organizations successfully mitigated risks and enhanced their
fraud detection capabilities.

5. How Data analytics helped the world to fight COVID:

COVID-19 has had a huge impact on the world. But it could have been even more
catastrophic if not for the use of big data analytics in the healthcare industry.
If you were in India at the time, you must have downloaded the Aarogya Setu app on
your phone to detect any threat of infection. The app uses Bluetooth and data
analytics for contact tracing.

2. Understand career scope and opportunities in the data domain.

● Explanation on the vast array of career opportunities that the data domain offers with the
requirement of skills w.r.t each opportunity.
○ Data analyst- data collection, data cleaning and analyzing data.
○ Data science
○ Business Intelligence developer- develop dashboards using power BI, tableau, etc.

● How is data analytics important in today’s world?

○ Procter and gamble (FMCG):
■ Consumer based companies like Procter and Gamble (P&G) heavily rely on
data analysis to gain a competitive advantage.
■ Data analysis allows P&G to gather and analyze vast amounts of data from
various sources, including sales data, customer feedback, and market
trends.
■ Data analysis helps P&G to find emerging trends and customer needs
which helps companies to improve their products.
■ Data analysis enables P&G to optimize its marketing and advertising
campaigns for maximum effectiveness.
■ Through data analysis, P&G can gather competitive intelligence by
monitoring competitor activities, pricing strategies, and market share trends.
○ Healthcare:
■ Data analytics allows healthcare providers to analyze patient data,
such as electronic health records (EHRs), medical imaging, and genetic
information, to identify patterns, trends, and correlations that can
improve diagnosis, treatment, and patient care.

2
 ■ Enables healthcare organizations to optimize clinical workflows,
resource allocation, and operational efficiency.
○ Manufacturing:
■ To implement predictive maintenance programs, which involve
analyzing equipment sensor data to predict when machines are likely to
fail.
■ To optimize their supply chain operations by analyzing data from
suppliers, logistics providers, and internal production processes.
■ To optimize production processes, improve efficiency, and reduce
waste.
● How the course will help students build a career in data analytics.
○ Foundational knowledge
○ Technical skills - excel, SQL, Power BI, Python
○ Soft skills- Improving problem solving skills, solution designing, deriving insights
and communication.
○ Hands-on experience- Real time projects and case studies.

3.A. Introduction to data and types of data:

● What is data?
○ Collection of facts.
○ Types of data based on organization:
■ Unstructured data
■ Semi-structured data
■ Structured data

● How to convert data to information?

UNSTRUCTURED DATA- Data which do not have predefined structure or organization.
Example- Audio, text, video, photographs, etc.

EXAMPLE:
1. CUSTOMER REVIEWS:
Online review platforms like Yelp, TripAdvisor, and Amazon host millions of
customer reviews containing valuable insights about products, services, and
businesses. By applying text mining and sentiment analysis techniques to analyze
customer reviews, businesses can identify recurring themes, sentiments, and patterns,
enabling them to improve product quality, address customer concerns, and enhance
customer satisfaction.
2. SOCIAL MEDIA POSTS:
Social media platforms generate vast amounts of unstructured data in the form of text,
images, and videos. Data analytics techniques such as sentiment analysis can be
applied to analyze social media posts and understand public sentiment towards
products, brands, or events. Natural language processing (NLP) algorithms can
extract topics, trends, and opinions from unstructured text data, enabling

3
 organizations to make data-driven decisions based on customer feedback and market
trends.

STRUCTURED DATA:

INFORMATION:

SEMI-STRUCTURED DATA:
- Semi-structured data does not reside in relational database (but structured data does)
- Has organizational properties making it easier to analyze
- Example - emails (which have structured headers and metadata but unstructured body
content).
STRUCTURED DATA:
● When we receive semi-structured data, then we convert it into structured data and after
data processing, we finally end up getting information from the data.
● In the above hypothetical example of google maps, we have data of certain areas, let’s say A1,
A2 and A3. Variables involved in the data are personal ID, longitude, latitude, source and

4
 destination. When we convert the raw data into structured data, basically we have converted
the data into tabular form which is considered to be highly structured and can be stored in a
relational database.
● After converting the data into structured format which is meaningful and understandable, now
we can derive insights. Here, each source and destination suggests a particular area and if
replaced with the area number then it will be more organized.

- Importance of understanding type of data

STRUCTURED DATA SEMI-STRUCTURED UNSTRUCTURED DATA

DATA

Highly organized data Partially organized data Unorganized data

Easily machine-readable Moderately machine readable Challenging for machines

Guarantees highest Only allows anonymous Only textual query is

performances with structure queries executable

Relational database, HTML, JSON files, XML Images, audios, videos

spreadsheets

3.B. Introduction to types of data analytics:

● Types of data analytics:
○ Descriptive (WHAT):
—>Analysis of the past performances to analyze the trend.
—>Helps in understanding any underlying issues of the company.
—>Utilized by almost every company to create dashboards and reports to figure out
existing trends from historical data. E.g.Sales trends, etc.

1. EXAMPLE- (Hypothetical)
A company has been following a downward trend for the past 6 years which
has become a little concerning for the company as well as the stakeholders of
the company. The stakeholders are worried and want to understand the reason
behind the downward trend and want to discuss what can be done to increase
the revenue.

5
 2. EXAMPLE-

○ Diagnostic (WHY): (Root cause analysis)

—> It tells us the reason why a particular trend is visible.
■ Now we know that the company is facing a downward trend for a couple of
years.
■ The next step is to understand the reasons why the company is facing a
downward trend.
● Better products in the market (New competitor)
● Less differentiation.
Other examples:
Starbucks case:
■ Starbucks was facing customer retention and store performance issues
■ It employed diagnostic analysis by using
1. customer feedback analysis (semi structured data)
2. sales data analysis at store level (structured data)
3. operations analysis (staff schedule, inventory, and order fulfillment
efficiency)
■ Based on above analysis, it-
● offered new products (based on customer survey & sales)
● improved operations and redesigned store layout
● personalized marketing (based on customer level sales analysis)

○ Predictive (Forecasting future predictions):

—> Tell us what is likely to happen in the future.
—> It is based on historical data and helps us make assumptions about the future
trend.

6
 ■ If we forecast the revenue of the company for the upcoming 3 years, then we
notice that there is a downward trend in future revenue.

Example 2- How amazon uses predictive analysis to increase revenue.

● Personalized Recommendations
● Enhanced Customer Experience
● Optimized Inventory Management
Example 3- Finance and banking
● Assess credit risk
● Detect fraudulent transactions
● Personalize customer experiences.
● customer transaction data.
● Credit scores.
● Market trends to make informed decisions on lending

○ Prescriptive:
—>Based on the current data, it helps in future decision making.
—>Providing suggestions based on the analysis performed.
● Improvement in the product.
● Marketing of the product.