Data analytics

Introduction to data analytics

 Data analytics is the systematic process of collecting, cleaning, transforming, and
analysing data to extract meaningful insights that support decision-making.
 it combines statistics, computer science, and domain knowledge to identify
patterns, trends, and relationships in data.
 organizations use data analytics to improve efficiency, reduce risk, and gain
competitive advantage.
 the process typically involves data acquisition, preprocessing, analysis,
visualization, and interpretation.
 The growth of big data, analytics has become essential across industries.
Example: an e-commerce company analyses customer purchase data to understand
buying behaviour and improve product recommendations.
Data analysis

 Data analysis refers to examining, cleaning, and modelling data to discover useful
information and support conclusions.
 It focuses on answering specific questions using existing data. Data analysis
involves techniques such as data filtering, aggregation, statistical calculations, and
visualization.
 The goal is to interpret past or current data to understand what happened and
why.
 It is a subset of data analytics and is often task-oriented. Data analysis is widely
used in research, business reporting, and operational monitoring.
Example: a college analysis student marks to identify subjects where students
perform poorly.
Types of data analytics

Descriptive data analytics

Answers “what happened?”

 Descriptive analytics summarizes historical data to understand what has happened

in the past.
 It uses techniques such as averages, percentages, totals, charts, and dashboards.
 This type of analytics converts raw data into meaningful summaries that are easy
to understand.
 It is the foundation of all other analytics types and is widely used in routine
reporting.
 Example: a company generates monthly sales reports showing total revenue, top
products, and regional performance using bar charts and tables.
Diagnostic data analytics
 Diagnostic analytics focuses on identifying the reasons behind past outcomes. It
answers the question “why did it happen?
 ” By drilling down into data, comparing variables, and finding correlations.
Techniques such as root cause analysis, data slicing, and variance analysis are
commonly used.
 This helps organizations understand problems and inefficiencies.
Example: a business analysis sales data to find that a decline in revenue occurred
due to increased product prices and reduced customer footfall.
Predictive data analytics
 Answers “what might happen?”
 Predictive analytics uses historical data, statistical models, and machine learning
algorithms to forecast future outcomes.
 It identifies trends and probabilities rather than certainties.
 This type of analytics helps organizations plan ahead and manage risks.
 Common techniques include regression, time-series analysis, and classification
models.
Example: a bank uses predictive analytics to estimate the likelihood of loan default
based on customer income, credit history, and repayment behaviour.
Prescriptive data analytics
 Answers “what should we do?”
 Prescriptive analytics goes a step further by recommending actions to achieve
desired outcomes.
 It combines predictive models with optimization and simulation techniques.
 This analytics type answers the question “what should be done?” It is highly
valuable for strategic decision-making and automation.
Example: a logistics company uses prescriptive analytics to suggest optimal
delivery routes that minimize fuel cost and delivery time.
Importance and usage of data analytics
Data analytics is used in many fields like banking, farming, shopping, government and
more. It helps in many ways:
Data analytics importance
 helps in decision making: it gives clear facts and patterns from data which help
people make smarter choices.
 helps in problem solving: it points out what's going wrong and why making
it easier to fix problems.
 helps identify opportunities: it shows trends and new chances for growth that
might not be obvious.
 improved efficiency: it helps reduce waste, saves time and makes work smoother
by finding better ways to do things.
Real-world applications of data analytics across domains
Data analytics is used in many areas of daily life to make better decisions using data.
1. Healthcare
In healthcare, data analytics is used to improve patient care and treatment. Hospitals
analyze patient records, test reports, and medical history to predict diseases and reduce
medical errors. It also helps in managing hospital resources efficiently.
Example: doctors use patient data to predict diabetes or heart disease at an early stage.
2. Education
In education, data analytics helps schools and colleges understand student performance.
Attendance, marks, and test results are analyzed to improve teaching methods and
support weak students.
Example: colleges analyze exam results to identify subjects where students need extra
coaching.
3. Banking and finance
Banks use data analytics to detect fraud, manage risk, and improve customer services. It
helps in loan approval and credit scoring by analyzing customer data.
Example: banks identify suspicious transactions to prevent online fraud.
4. Retail and e-commerce
Retailers analyze customer purchase data to understand buying behavior. This helps in
managing inventory, offering discounts, and increasing sales.
Example: online shopping websites recommend products based on past purchases.
5. Manufacturing
In manufacturing, data analytics improves product quality and reduces machine failure. It
helps monitor machines and predict maintenance needs.
Example: factories use sensor data to detect machine faults before breakdown.
6. Transportation and logistics
Data analytics is used to optimize routes, reduce fuel cost, and improve delivery time.
Example: delivery companies choose the fastest routes using traffic data.

Role of python in data-driven decision making

1. Data collection
 Python is used to collect data from multiple sources such as csv files, excel sheets,
databases, and web API’s.
 It helps organizations gather structured and unstructured data efficiently, which is
the first step in data-driven decision making.

2. Data cleaning and preparation

 Raw data often contains missing values, duplicates, and errors. Python plays a
major role in cleaning and preparing data using pandas
  Clean data ensures that analysis results are reliable and meaningful for decision
making.
3. Data analysis and exploration
 Python is used to analyze data by calculating statistics, identifying patterns, and
discovering trends.
 Exploratory data analysis (eda) helps organizations understand business
performance and customer behavior, supporting informed decisions.
4. Predictive analysis and modeling
 Using machine learning libraries like scikit-learn, python builds predictive models
to forecast future outcomes.
 These predictions help organizations plan strategies, manage risks, and improve
performance.
5. Decision support and reporting
 Python generates reports and dashboards that summarize insights clearly.
 Decision-makers use these insights to choose the best actions based on data rather
than assumptions.
Data-Driven Decision Making (DDDM)
Definition:
Data-Driven Decision Making is the process of making business or operational
decisions based on data analysis rather than intuition, personal experience, or
guesswork.
Key Concept:
 Decisions are guided by insights extracted from structured or unstructured data.
 Python is widely used because of its data handling, analysis, and visualization
capabilities.
Why Python for Data-Driven Decision Making (DDDM):
 Easy to Learn & Use – Simple, readable syntax.
 Powerful Data Handling – pandas and numpy handle large datasets efficiently.
 Advanced Analytics – Supports statistics, machine learning, and predictive
modeling (scikit-learn, tensorflow).
 Data Visualization – Clear insights with matplotlib, seaborn, plotly.
 Automation & Integration – Automate reports and connect with databases/APIs.
 Open-Source & Community Support – Free, with extensive libraries and tutorials.
 Python combines analysis, visualization, and automation in one ecosystem,
making it ideal for DDDM.
 Difference b/w data analytics and data analysis
Basis Data analytics Data analysis
A broad process of using data to make A process of examining data
Meaning
decisions to find answers
Wide (includes analysis, prediction, and Narrow (mainly focuses on
Scope
decision-making) analyzing data)
Focus Past, present, and future Mostly past and present
What happened, why, what will happen,
Main question What happened and why
and what should be done
Techniques Descriptive, diagnostic, predictive,
Descriptive and diagnostic
used prescriptive
Output Insights, predictions, and action plans Reports and summaries
Complexity More complex Less complex
Tools used Python, r, sql, ml tools, bi tools Excel, sql, basic python
Goal Support strategic decisions Understand data results
Predicting future sales and suggesting
Example Calculating last year’s sales
strategy
difference between data mining and data analytics
Basis Data mining Data analytics
Data mining is the process of Data analytics is the process of
discovering hidden patterns, analyzing data to draw conclusions,
Meaning
relationships, and useful information make decisions, and predict future
from large datasets. outcomes.
Focuses on interpreting data to
Focuses on finding new and unknown
Focus answer questions and support
patterns in data.
decisions.
It is a broader concept that includes
Scope It is a subset of data analytics. data mining, analysis, and
visualization.
Uses clustering, classification, Uses descriptive, diagnostic,
Techniques
association rules, and anomaly predictive, and prescriptive
used
detection. analytics.
Works on small, medium, and large
Data size Mostly works on very large datasets.
datasets.
Produces insights, reports, and
Output Produces patterns, rules, and models.
actionable decisions.
To discover hidden knowledge from To improve decision making using
Goal
data. data.
Basis Data mining Data analytics
Finding buying patterns such as
Analyzing sales data to decide
Example customers who buy bread also buy
which product should be promoted.
butter.

Process of data analytics

1. Define the problem
 This is the starting point of data analytics. In this step, the objective of analysis is
clearly identified.
 The problem statement should explain what needs to be solved, why it is
important, and what outcome is expected.
 A clearly defined problem ensures that the analysis stays focused and relevant.
Example: a company wants to know why its sales have decreased in the last three
months.
2. Data collection
 After defining the problem, relevant data is collected from different sources such as
databases, surveys, sensors, or websites.
 The quality of data directly affects the accuracy of results.
Example: collecting monthly sales, customer feedback, and pricing data.
3. Data cleaning
 The collected data may contain missing values, errors, or duplicate records.
 Data cleaning removes or corrects these issues to improve data quality.
Example: removing duplicate customer records and correcting wrong entries.
4. Data preparation
 In this step, data is organized and transformed into a suitable format for analysis.
 It includes sorting, filtering, and converting data types.
Example: converting date fields into a standard format.
5. Data analysis
Data analysis involves applying statistical and analytical techniques to identify patterns
and trends related to the problem.
Example: analyzing sales trends to find low-performing products.
6. Data visualization
Analyzed data is presented using charts, graphs, and dashboards for easy understanding.
Example: showing sales performance using bar and line charts.
7. Interpretation and insights
The results are interpreted to generate meaningful insights that answer the problem
statement.
Example: identifying that sales dropped due to high product prices.
8. Decision making and action
Based on insights, appropriate actions or decisions are taken to solve the problem.
Example: reducing prices or offering discounts.
9. Monitoring and evaluation
After implementation, outcomes are monitored to measure effectiveness and make
improvements.
Example: tracking sales after price reduction.

Methods of data analytics

There are two types of methods in data analytics which are mentioned below:
1. Qualitative data analytics
Qualitative data analysis doesn’t use statistics and derives data from the words,
pictures and symbols. Some common qualitative methods are:

 Narrative analytics is used for working with data acquired from diaries,
interviews and so on.
 Content analytics is used for analytics of verbal data and behaviour.
 Grounded theory is used to explain some given event by studying.
2. Quantitative data analysis
Quantitative data analytics is used to collect data and then process it into the numerical
data.
Some of the quantitative methods are mentioned below:
 Hypothesis testing assesses the given hypothesis of the data set.
 Sample size determination is the method of taking a small sample from a large
group of people and then analysing it.
 Average or mean of a subject is dividing the sum total numbers in the list by the
number of items present in that list

Types of variable data

  In data analytics and statistics, a variable refers to any characteristic, number, or
quantity that can be measured or counted.
 Variables represent the data that researchers collect and analyze to draw
conclusions. Depending on how the data behaves and what it represents,
Variables are generally classified into two main types: qualitative (categorical)
variables and quantitative (numerical) variables. .
 Qualitative variables, also known as categorical variables, describe non-numeric
Characteristics or qualities of data.
 They are used to categorize or label attributes rather than measure them
numerically.
 These variables answer questions like “what type?” Or “which category?” And
cannot be meaningfully added, subtracted, or averaged. Qualitative data is
further divided into two subtypes — nominal and ordinal variables.
 Nominal variables represent categories without any logical order or ranking.
Examples-include gender (male, female, other), nationality (indian, american,
french), or blood type (a, b, ab, o).
 Ordinal variables, on the other hand, represent categories that can be arranged in
a Meaningful order, but the intervals between categories are not equal. Examples
include Education level (primary, secondary, higher) or customer satisfaction (poor,
fair, good, Excellent).
1. Quantitative variables,
 also called numerical variables, represent data that can be Measured and expressed
in numbers.
 These variables describe quantities and can be subjected to mathematical
operations such as addition, subtraction, multiplication, and division.
 Quantitative data is further classified into discrete and continuous variables.
 Discrete variables
take only whole, countable values — they cannot be divided into
Smaller parts
. For example, the number of students in a class, number of cars in a parking
Lot, or number of books on a shelf are all discrete variables because you cannot have half
A student or half a car.
2. Continuous variables,
 In contrast, can take any value within a given range, including Fractions and
decimals.
 They represent measurements rather than counts. Examples Include height, weight,
temperature, or time, as these can be measured to any level of Precision.

 In data analytics and statistics, the levels of data measurement describe how data
can be categorized, ordered, and measured.
 These levels determine the type of analysis and mathematical operations that can
be performed on a dataset.
 There are four main levels of measurement — nominal, ordinal, interval, and ratio
— each providing a different degree of information about the data.

1. Nominal level
 is the most basic form of data measurement.
 It involves classifying Data into distinct categories that have no logical order or
ranking.
 These categories are Used only for labeling purposes, and the numbers assigned to
them have no quantitative
 Meaning. Nominal data is qualitative in nature and is mainly used for classification
or Identification purposes rather than calculation.
Nominal example:
Gender (male, female, other) or blood type (a, b, ab, o). These categories are simply
Names without any inherent order.
2. Ordinal level
 of measurement classifies data into categories that can be ranked
 Or ordered, but the exact differences between the ranks are not equal or known.
This
  Means ordinal data shows the relative position of items, such as first, second, or
third, but
 Does not specify how much better one item is compared to another. It provides
more
 Information than nominal data but still lacks precise measurement.
Ordinal example:
Customer satisfaction levels such as “poor,” “average,” “good,” and “excellent.” The
Data shows ranking, but the exact difference between “good” and “excellent” is not
Measurable.
3. Interval level
 of measurement involves data that can be ordered and where the
 Difference between values is meaningful and equal. However, there is no true zero
point
 In an interval scale, meaning zero does not indicate a complete absence of the
variable
 Being measured. While differences can be compared, ratios are not meaningful
because
 There is no absolute zero. This level of data is often used in fields that measure
Temperature, time, or standardized test scores.
Interval example:
Temperature measured in celsius or fahrenheit. The difference between 20°c and 30°c
Is the same as between 30°c and 40°c, but 0°c does not mean there is no temperature.
4. Ratio level
 of measurement is the highest and most informative level.
 It has All the characteristics of the interval scale — data is ordered and differences
between Values are equal — but it also includes a true zero point.
 A true zero represents a Complete absence of the variable, allowing for meaningful
comparisons using ratios.
 Ratio data supports all mathematical operations, including addition, subtraction,
Multiplication, and division, and is often used for measurable quantities such as
height, Weight, age, or income.
Ratio example:
Weight or height. A person weighing 60 kilograms is twice as heavy as someone
Weighing 30 kilograms, and 0 kilograms means no weight