Unit 1 : Introduction to Data Analytics

The Internet of Things (IoT) is a big change in technology that connects everyday
devices—like smartwatches, home appliances, cars, and machines—to the
internet. These devices can talk to each other and share data.

IoT data analytics is the process of looking at all the data these smart devices
collect. By studying this data, we can find useful patterns, make better decisions,
and improve how things work
For example, data from a smart thermostat can help save energy, or data from
machines in a factory can help prevent breakdowns.
In short, IoT connects devices, and IoT data analytics helps us understand and use
the data they create.

three main steps in analyzing IoT data

1. Data Collection
This is the first step. Devices like sensors, smart gadgets, and machines
in the IoT system collect all kinds of information—like temperature,
location, humidity, and more. These devices are always gathering data
and then sending it to a central system or the cloud, where it can be
stored and worked on later.
Example: A smart home thermostat collects temperature data all day
and sends it to an app you can see on your phone.

2. Data Processing
Once the data is collected, it needs to be cleaned and organized. That
means removing any errors, filling in missing details, and turning messy
or unstructured data into a format that computers can understand.
Then, smart tools like machine learning are used to find patterns or
important information.
Example: If a machine in a factory sends data every second, processing
helps remove duplicate data and find the times when it wasn’t working
properly.

3. Data Interpretation and Insights

Now that the data is clean and organized, we can study it to discover
useful insights—like spotting trends, detecting problems early, or
predicting what might happen next. These insights help businesses and
organizations make smarter decisions, save time, reduce costs, and
work more efficiently.
Patterns (what usually happens)
Trends (how things are changing over time)
Anomalies (anything unusual or wrong)
Example: If a sensor on a truck shows it's overheating often, the
company can fix it before it breaks down—saving money and avoiding
delays.
Important Elements of Data Analytics for IoT
Element What it Does Example

Devices Collect data Smartwatch tracking steps

Data Collection Transfers data Sends it to the cloud

Storage Keeps data safe Stored in Google Cloud

Processing Cleans data Removes duplicates, fixes gaps

Analytics Finds patterns Predicts a machine failure

Visualization Displays results Graphs, dashboards

Insights Supports action Sends alert for repair or change

Uses for Internet of Things Data Analytics

IoT data analytics has many applications in many sectors, demonstrating
the revolutionary power of using insights from the massive amounts of
data connected devices create. The following are some important
applications:
1. Smart Cities
Sensors collect info about traffic, pollution, and public transport.
Helps city planners:
• Fix traffic jams
• Improve bus and train schedules
• Control traffic lights better
2. Finance and Insurance
Data from devices helps banks and insurance companies:
• Check if someone is risky to lend money to
• Find fake transactions quickly
• Offer plans based on real habits (like safe driving)

3. Healthcare
Wearables and remote devices send health data.
Doctors can:
• Watch patient health in real-time
• Give personalized treatments
• Spot problems early before they get worse

4. Industrial IoT (Factories)

Machines have sensors that check how they work.
Helps factories:
• Know when machines might break
• Fix machines before they stop
• Make production faster and better
5. Energy Management
Smart meters track how much energy homes and businesses use.
Helps people:
• Save energy by turning off unused devices
• Reduce electricity bills
• Help companies plan energy use better

6. Agriculture
Sensors check soil, weather, and crop health.
Helps farmers:
• Water plants only when needed
• Use fertilizer carefully
• Grow more food with less waste

7. Transportation and Logistics

Sensors track trucks, ships, and packages.
Helps companies:
• Find fastest routes
• Keep vehicles in good shape
• Let customers know where their delivery is

8. Retail
Stores use IoT to track products and customers.
Helps stores:
• Know when to restock shelves
• Give customers special offers they like
• Make shopping easier and faster

9. Building Automation
Sensors monitor temperature, light, and security in buildings.
Helps buildings:
• Save energy by adjusting lights and heating
• Keep people safe with smart alarms
• Make rooms comfortable automatically

10. Environmental Monitoring

Sensors watch air, water, and animals.
Helps scientists:
• Detect pollution early
• Protect endangered animals
 Challenges of IoT Data Analytics

1. Too Much Data

IoT devices create a huge amount of data very quickly, which is
hard to store and analyze.

2. Different Types of Data

Data comes in many formats from different devices, making it
difficult to combine and understand.

3. Bad or Incomplete Data

Sensors can send wrong or missing data due to errors or
damage.

4. Security and Privacy Risks

IoT data can be sensitive, and devices can be hacked, causing
privacy and safety issues.

5. Hard to Grow
As more devices are added, it's hard to keep systems running
fast and smoothly.

6. Need for Fast Decisions

Some IoT systems (like in health or cars) need data to be
analyzed instantly.

7. Storage Problems
Storing endless data costs money and needs smart ways to keep
only useful information.

8. Who Owns the Data?

It's often unclear who has rights to the data—the user or the
company.

9. Old Systems
Many businesses use outdated systems that can’t handle
modern IoT data.

10. No Common Rules

There are no universal standards for IoT data, making it hard for
systems to work together.

Types of Data Analytics

There are four major types of data analytics:
1. Predictive (forecasting)
2. Descriptive (business intelligence and data
mining)
3. Prescriptive (optimization and simulation)
4. Diagnostic analytics

1. Descriptive Analytics — What happened?

• This type looks at all the data collected from the past.

• It organizes and summarizes the data to explain what happened before.

• Imagine looking at a photo album of last year’s events — it shows you the story clearly.
 • It uses numbers like totals, averages, and percentages to tell the story.

provide historic reviews like:

• Data Queries

• Reports

• Descriptive Statistics

• Data dashboard

• Example: A store checks how many products it sold last month and which were the best
sellers.

• Why it’s useful: It helps you understand your past performance and see important
patterns.

2. Diagnostic Analytics — Why did it happen?

• This type digs deeper to find the reasons behind what happened.

• It asks questions and looks for connections between different data points.

• Think of it like a detective trying to find out why something went wrong or right.

• It uses tools like looking for patterns, comparing data, or finding relationships.

• Data discovery

• Data mining

• Correlations

• Example: A store finds out sales dropped because the delivery trucks were late and
shelves were empty.

• Why it’s useful: It helps fix problems and avoid making the same mistakes again.

3. Predictive Analytics — What will happen next?

• This type uses past and current data to guess what might happen in the future.

• It uses smart computer programs (called machine learning) to find patterns and make
predictions.
 • Imagine a weather forecast predicting rain based on past weather data.

• It can tell you the chance of something happening or how likely an event is.

Basic Cornerstones of Predictive Analytics

• Predictive modeling

• Decision Analysis and optimization

• Transaction profiling

• Example: A store predicts which products will be popular next month so they can stock
up.

• Why it’s useful: It helps you prepare and make better plans for the future.

4. Prescriptive Analytics — What should we do?

• This type suggests the best actions to take based on predictions and data.

• It helps you decide what is the smartest or most effective choice.

• Think of it like a GPS that tells you the fastest route to avoid traffic.

• It uses tools like simulations and optimization models to find the best solution.

• Example: A store plans the best delivery schedule to save money and keep shelves
stocked.

• Why it’s useful: It helps you make the best decisions to achieve your goals efficiently.

Key Roles of Data Analytics:

1. Data Mining

What happens:

• We collect data from many sources like websites, apps, sensors, machines,
or even people.
 • The data might be messy or in different formats, so we clean it and make it
all the same.

Why it matters:

• Having clean, organized data helps us analyze it better and get good results.

Example:
Imagine collecting customer feedback from emails, surveys, and social media, then
putting all answers into one list.

2. Data Management

What happens:

• We save the collected data in databases or storage systems.

• We use tools like SQL to organize, search, and manage the data easily.

Why it matters:

• Proper storage keeps data safe and makes it easy to find when needed.

Example:
Like saving all your photos in folders on your computer so you can find them later.

3. Statistical Analysis

What happens:

• We study the data to find important trends, patterns, or changes over time.
• We use software tools like Python or R to run calculations and create
models.

Why it matters:

• It helps us understand the data deeply and predict what might happen next.
Example:
Analyzing past sales data to predict which products will be popular next month.

4. Data Presentation

What happens:

• We turn the analysis results into easy-to-understand charts, graphs, or

reports.
• We explain the insights clearly so everyone can understand and use them.

Why it matters:

• Good presentation helps decision-makers understand the data and take

action.

Example:
Showing a bar chart to your team to explain how sales have grown over time.

Super Short Summary:

➢ Get the data

➢ Store the data
➢ Study the data
➢ Share the results

steps in data analysis

1. Define Data Requirements

➢ Decide what kind of data you need and how to group it.
➢ For example, you might want to organize data by age, gender, income, or
location.
➢ Data can be numbers (like age or salary) or categories (like male/female).

2. Data Collection

➢ Gather data from different sources.

➢ This can be from computers, websites, cameras, sensors, or even people.
3. Data Organization

➢ Put all the collected data into a neat, structured format.

➢ Use spreadsheets (like Excel) or special software to arrange the data so it’s
easy to work with.

4. Data Cleaning

➢ Check the data for mistakes, missing pieces, or repeated info.

➢ Fix or remove wrong or incomplete data to make sure the information is
correct and reliable.
➢ Clean data helps make better analysis later.

Uses of Data Analytics

1. Improve Business Decisions
Helps companies make smart choices by understanding sales, customers, and
market trends.
2. Predict Future Trends
Uses past data to guess what will happen next, like predicting weather or
stock prices.
3. Increase Efficiency
Finds ways to save time and money by improving processes and reducing
waste.
4. Understand Customers
Shows what customers like and want, so businesses can offer better products
or services.
5. Detect Problems Early
Finds issues before they become big, like spotting faulty machines or fraud.
6. Personalize Experiences
Tailors recommendations and services to individual users, like Netflix or
Amazon suggestions.
7. Healthcare Improvements
Analyzes patient data to improve treatments and predict disease outbreaks.

Here’s a simple and clear explanation of the future scope of data analytics:
 Future Scope of Data Analytics
1. More Use of Artificial Intelligence (AI)
Data analytics will work more with AI and machine learning to make
smarter predictions and automate decisions.
2. Growth of Big Data
As more devices and systems create data, analytics will handle even larger
amounts of information.
3. Real-Time Analytics
Faster data processing will allow companies to get instant insights and react
quickly.
4. Edge Analytics
Data will be analyzed closer to where it is collected (like on devices
themselves) to save time and reduce delays.
5. Better Personalization
Businesses will use analytics to offer even more personalized products,
services, and experiences.
6. Improved Healthcare
Data analytics will help in early disease detection, personalized treatments,
and managing healthcare better.
7. Enhanced Security
Analytics will help detect and prevent cyberattacks more effectively.
8. More Job Opportunities
Demand for data analysts, scientists, and engineers will keep growing as
data becomes more important.

Key Challenges to IoT Analytics

1. Huge Amount of Data

IoT devices like sensors, cameras, and smart gadgets generate massive
amounts of data continuously. For example, a single sensor can send
thousands of data points every minute. Handling this big volume of data
requires large storage capacity and powerful processing tools, which can be
expensive and complex to manage.

2. Different Types of Data

 IoT devices produce various kinds of data:

• Numbers (temperature, speed)

• Images or videos (security cameras)
• Text or logs (device messages)

Since this data comes in different formats and styles, combining and
analyzing it all together is challenging. You need special software and
methods to make sense of these mixed data types.

3. Data Quality Issues

Data collected from IoT devices can be incomplete, incorrect, or noisy

because of:

• Faulty or aging sensors

• Poor network connections causing data loss
• Environmental interference (e.g., weather affecting sensors)

Bad data leads to wrong insights, so cleaning and verifying data is a big
challenge.

4. Security and Privacy Risks

IoT devices are often connected to the internet, making them targets for
hackers. If devices or data are not secure:

• Personal or sensitive information could be stolen.

• Devices might be controlled by unauthorized users, causing harm.

Ensuring strong security and protecting privacy is essential but difficult

due to the variety and number of devices.

5. Real-Time Processing
 Some IoT applications need data to be analyzed instantly for quick
decisions, such as:

• Self-driving cars reacting to obstacles

• Health monitors alerting doctors immediately

Real-time analytics require fast, reliable computing power and networks,

which can be costly and technically hard to implement.

6. Scalability

As the number of IoT devices grows (from hundreds to millions), the data
analytics system must also grow without slowing down. Building systems
that can scale efficiently to handle this increasing load is a big technical
challenge.

7. Integration Problems

Often, IoT data needs to be combined with existing business systems (like
customer databases or manufacturing controls). Different systems may use
different data formats or software, making it hard to integrate everything
smoothly.

8. Lack of Standards

There are many types of IoT devices made by different companies, and they
often use different communication protocols and data formats. Without
common standards, it’s difficult for devices and systems to work together
seamlessly.

Summary

Managing IoT analytics is tough because of:

 • Massive, fast data
• Mixed and poor-quality data
• Security risks
• Need for fast, real-time answers
• Growing number of devices
• Integration and standardization issues

🔄 What is Streaming Analytics?

Streaming Analytics is the process of collecting, processing, and analyzing live data — data
that is being created right now, second by second.

Instead of waiting until all data is collected (like in traditional data analysis), streaming analytics
helps you see what’s happening immediately and respond fast.

.example
Patient Health Monitoring

What happens:
A heart rate monitor tracks a patient's heartbeat live.

How streaming analytics helps:

If the heartbeat becomes too fast or too slow, the system instantly alerts the nurse or doctor.

Why is Streaming Analytics Important?

Because in many areas, you can’t afford to wait. You need to take action
immediately — for safety, money, or performance.

Examples:

➢ Hospitals need to react quickly if a patient’s heart rate changes.

➢ Banks need to block a transaction if it looks like fraud.
➢ Factories need to stop machines if something goes wrong.

How Does Streaming Analytics Work?

1. Data Generation

➢ Data comes from sources like:

➢ IoT sensors (temperature, motion, speed)
➢ Mobile apps (user activity, clicks)
➢ Social media (comments, likes)
➢ Machines or smart devices (errors, alerts)
➢ GPS or traffic systems
➢ These sources send data non-stop (like a stream of water).

2. Data Ingestion

➢ This means collecting the live data and sending it to a system that can
process it fast.
➢ Tools used:
➢ Apache Kafka
➢ Amazon Kinesis
➢ Azure Event Hubs

3. Real-Time Processing

➢ The system checks the incoming data:

➢ Is it normal?
➢ Is something wrong?
➢ Should an alert be sent?
➢ Tools used:
➢ Apache Flink
➢ Apache Spark Streaming
➢ Google Dataflow

4. Real-Time Actions

➢ If the system finds something important, it reacts instantly:

➢ Sends a notification or alert
➢ Changes machine settings
➢ Blocks a user or transaction
➢ Logs the issue for a human to check

5. Results are Shown

➢ The system shows results using:
➢ Dashboards
➢ Graphs
➢ Reports
➢ People can see what’s happening in real time

1. 📡 Data is Created

• Devices like sensors, apps, or machines create live data.

• Example: A sensor checks temperature every second.

2. 🚀 Data is Sent (Streaming)

• The data is sent immediately (like a live stream of water).

• It keeps flowing, non-stop.

3. 🧠 Real-Time Processing

• The system looks at the data right away.

• It checks:
o Is the data normal?
o Is something wrong?
o Should we take action?

4. 🚨 Instant Action (if needed)

• If a problem is found, the system:

o Sends an alert
o Stops a machine
o Sends a message
o Fixes the issue automatically

5. 📊 Results are Shown

• The system shows results using:

o Dashboards
o Graphs
o Reports
• People can see what’s happening in real time.
• Here are the main features of Streaming Analytics in very simple words:
 Features of Streaming Analytics
1. Real-Time Data Processing

• It works with live data, not old data.

• It gives results in seconds, not hours or days.

2. Continuous Data Flow

• Data keeps coming in non-stop, like a river.

• The system watches this stream all the time.

3. Fast Decision-Making

• It finds problems or patterns quickly.

• Helps people or systems make fast choices.

4. Instant Alerts & Actions

• Sends messages or warnings right away if something is wrong.

• Can even stop machines or fix issues automatically.

5. Live Dashboards

• Shows results using live charts, graphs, or reports.

• People can see what’s happening right now.

6. Continuous Monitoring

• Keeps checking data all the time, 24/7.

• Great for factories, hospitals, banking, and more.
• 7. Smart Detection
• Uses rules, statistics, or AI to find unusual patterns or mistakes.
• Example: Finds fake transactions in online banking.

8. Works with Many Data Sources

 • Can take data from:
• Sensors
• Apps
• Cameras
• Social media
• Machines

Benefits of Streaming Analytics

1. Fast Decision-Making

• You get information immediately.

• You can act quickly — no need to wait.

Example: Stop a machine as soon as it gets too hot.

2. Instant Problem Alerts

• Warns you right away if something is wrong.

• Helps prevent damage or danger.

Example: Alerts a doctor if a patient's heartbeat is too high.

3. Saves Money

• Fixes small issues before they become big problems.

• Saves time and reduces repair or loss costs.

Example: Detects fraud early and blocks it.

4. Better Monitoring

• Keeps watching everything live (24/7).

 • Helps you know what’s happening right now.

Example: Live traffic updates to avoid jams.

5. Improved Performance

• Finds ways to work smarter and faster.

• Tracks what’s working and what’s not.

Example: In factories, it helps improve machine use.

6. Better Customer Experience

• Gives people what they want in real time.

• Makes services quicker and smarter.

Example: Shows product suggestions while you're shopping online.

7. Competitive Advantage

• Companies that use real-time data can stay ahead.

• Makes smarter decisions than competitors.

Example
Fraud Detection with Streaming Analytics
1. Continuous Monitoring of Transactions

• Every transaction (credit card purchase, online payment, bank

transfer) is checked as it happens.
 • Data like amount, location, time, device, and user behavior flows
continuously.

2. Use of Rules and Patterns

• The system uses rules and patterns to detect fraud.

For example:
o Is the transaction amount unusually large?
o Is the purchase happening from a new or suspicious
location?
o Is the user behavior different from usual?
• These patterns are learned over time from past fraud cases.

3. Machine Learning and AI

• Advanced systems use machine learning models to spot fraud

more accurately.
• These models learn from millions of transactions to recognize
subtle fraud patterns that rules might miss.
• They update continuously as new fraud tactics emerge.

4. Instant Risk Scoring

• Each transaction gets a risk score in milliseconds.

• High risk means the transaction is likely fraudulent.
• Low risk means it’s probably safe.

5. Immediate Action
 • Based on the risk score, the system can:
o Allow the transaction automatically
o Ask for extra verification (like sending a code to your phone)
o Block the transaction completely
o Alert the bank’s fraud team for further investigation

6. Real-Time Alerts

• When fraud is suspected, alerts are sent immediately to:

o Customers (to confirm if they did the transaction)
o Bank employees (to take further steps)
o Automated systems (to freeze accounts or cards)

7. Benefits of Streaming Analytics in Fraud Detection

• Speed: Fraud is detected instantly, minimizing loss.

• Accuracy: AI and rules together reduce false alarms.
• Scalability: Can handle millions of transactions every second.
• Adaptability: Learns and evolves to catch new fraud methods.

Case study
Streaming Analytics Market Growth

1. Big Market Growth

• In 2021, the streaming analytics market was worth $15.4 billion.

 • By 2026, it is expected to grow to $50.1 billion.
• That means the market is getting much bigger very fast!

2. Why Is It Growing?

Streaming analytics is used in many important areas, such as:

• Fraud detection (finding fake activities fast)

• Sales and marketing (understanding customers better)
• Predictive asset management (fixing machines before they break)
• Risk management (finding and reducing risks)
• Network management (keeping computer networks running smoothly)
• Location intelligence (tracking places and movement)
• Supply chain management (making sure products arrive on time)
• Product innovation and customer management (creating better products
and services)

3. Who Uses Streaming Analytics?

• Big companies use it the most because they collect huge amounts of data.
• They need fast ways to understand and use this data.

4. Trends Helping Growth

• More digital technology being used everywhere.

• New tech like IoT (Internet of Things) and AI (Artificial Intelligence).
• Better ways to connect and share data.
• Growing need to analyze data in real time (instantly).

5. Challenges Slowing Growth

• Data security rules: Companies must keep data safe and follow laws, which
can be tricky.
 • Managing lots of data: Data often comes from many places, making it hard
to handle.
• Old systems: Many companies still use old computer systems that are hard
to connect with new streaming analytics tools.
• Here’s a simple and clear definition of Spatial Analytics:
• Here’s a simple and clear definition of Spatial Analytics:

Spatial Analytics
Spatial analytics = Location + Data + Smart decisions
Spatial analytics is the process of collecting, analyzing, and understanding
data that is connected to a specific location or place. It helps people see
patterns, trends, and relationships on a map so they can make better
decisions.

In Simple Words:

It’s like using a smart map to understand what is happening, where, and
why — such as traffic in a city, health issues in a region, or crop growth on a
farm.

Example:

If many accidents happen at one road intersection, spatial analytics helps

find that spot on the map and understand what’s causing it — like poor
lighting or bad road design.

Why Is Spatial Analytics Important?

It helps people and businesses make better decisions:

Save time and money

Improve services (like traffic or healthcare)
Plan smarter for the future
Respond faster to emergencies

How Does Spatial Analytics Work?

 Step 1: Collect Data with Location

• This data can come from:

o GPS on phones
o Traffic cameras
o Drones and satellites
o Health apps
o Farm equipment

Step 2: Put Data on a Map

• Use GIS (Geographic Information Systems) to layer data.

o One layer might show roads.
o Another shows hospitals.
o Another shows where people live.

Step 3: Analyze the Data

• Look for patterns, distances, and connections.

o Example: Are car crashes happening near schools?
o Example: Are crops growing better on one side of the field?

Step 4: Make Smart Decisions

• Maps and results help:

o City planners
o Doctors and health workers
o Farmers and business owners

Example 1: Traffic Flow Management

What Happens:

• Traffic sensors, GPS, and cameras collect real-time data.

• Data includes: vehicle speed, road usage, traffic jams.

How Spatial Analytics Helps:

• Shows where and when traffic is slow.

• Helps design better road systems and signal timing.
• Suggests alternate routes during peak hours.

Benefits:

• Less traffic, safer roads, faster travel.

Example 2: Public Health

What Happens:

• Health data is collected from hospitals and mobile apps.

• Includes: illness cases, vaccine coverage, water quality.

How Spatial Analytics Helps:

• Tracks where disease is spreading.

• Finds areas without enough clinics.
• Targets high-risk places for quick action.

Benefits:

• Stops disease early, saves lives, sends help faster.

Example 3: Farming and Agriculture

What Happens:

• Drones, GPS tractors, and soil sensors collect data.

• Measures: soil type, crop health, rainfall.

How Spatial Analytics Helps:

• Shows which parts of a field need more water or fertilizer.

• Predicts how much food will grow this season.
• Protects crops from weather damage.

Benefits: Bigger harvests, lower costs, smarter farming.