Scribd
Upload
32 views7 pages

Bae Cia 3

Uploaded by

jaijoce2003
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PDF, TXT or read online on Scribd
32 views7 pages

Bae Cia 3

Uploaded by

jaijoce2003
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PDF, TXT or read online on Scribd
You are on page 1/ 7

BUSINESS ANALYTICS SKILLS

CIA – 3

Questions and Answers

Submitted by:

Name: JOCELYN A
USN: JU2024MBA19015
SEC: D
1.Compare and contrast Structured, Semi-Structured, and Unstructured
Data.
Data can be compared to a library. Structured data resembles a well-organized
bookshelf, semi-structured data is like a bookshelf that’s somewhat organized,
and unstructured data is akin to a chaotic pile of books on the floor.
1. Structured Data (Organized Folder)
- Easy to locate and access
- Everything is clearly labeled and categorized
- Structured data facilitates automation and analysis
- Examples: Phonebook, Bank Statements, Spreadsheets

2. Semi-Structured Data (Partly Organized Folder)


- Some labels and categories, but not entirely perfect
- Still relatively easy to search and find what you need
- Semi-structured data requires some human intervention
- Examples: Email headers, CSV files, XML documents

3. Unstructured Data (Messy Folder)


- Lacks labels or categories
- Difficult to find specific information
- Unstructured data necessitates advanced tools and techniques
- Examples: Photos, Videos, Social Media Posts, Text Messages

Examples
- Structured: contacts list
- Semi-Structured: email inbox
- Unstructured: Social media feed
2. Explain the differences between various types of databases and storage
systems like Datawarehouse, Data Mart, and Data Lake.

Data Warehouse
• A data warehouse stores data from various sources in one place.
• It's like a library, organized and structured. Helps in business
decision-making.

Data Mart
• A data mart is a smaller data warehouse focused on one area.
• It's like a bookstore, specialized and easy to navigate. Supports
specific business teams.

Data Lake
• A data lake stores raw, unorganized data from everywhere.
• It's like a vast ocean, holding treasures and surprises. Ideal for big
data analytics.

Relational Database (RDBMS)


• A relational database stores structured data in tables.
• It's like a filing cabinet, organized and secure. Suitable for
transactions.

NoSQL Database
• A NoSQL database stores unstructured or semi-structured data.
• It's like a flexible container, adapting to changing needs. Ideal for
big data.

Cloud Storage
• Cloud storage keeps data online, accessible anywhere.
• It's like a virtual file cabinet, secure and scalable. Supports
collaboration.

Object Storage
• Object storage holds unstructured data as objects.
• It's like a warehouse, storing and retrieving items easily. Optimized
for media files.
3.Differentiate between Online Analytical Processing (OLAP) and Online
Transaction Processing (OLTP).

OLTP (Transaction Store)

• OLTP is like a cashier, handling individual transactions efficiently


• Sells products one by one (transactions)
• Focuses on daily sales and customer interactions
• Fast and efficient checkout process
• OLTP ensures smooth daily operations
• fast and frequent
• User: staff and customers
• OLTP: Buying groceries online
• Example: Grocery store

OLAP (Analysis Store)

• OLAP is like a business consultant, analyzing sales data to optimize


operations.
• Analyzes sales data to understand trends
• Focuses on business decisions and strategy
• Provides insights into customer behavior
• OLAP informs strategic business decisions
• complex and periodic
• User: managers and analysts
• OLAP: Analyzing last quarter's sales data
• Example: Market research office
4. Illustrate the types of Machine Learning (ML) with examples.
1. Supervised Learning
Definition: Learning from labeled data to predict outcomes.
Example: Image recognition - teaching a model to identify dogs vs. cats
using labeled pictures.

2. Unsupervised Learning
Definition: Discovering patterns in unlabeled data.
Example: Customer segmentation - grouping similar shoppers based on
buying habits.

3. Reinforcement Learning
Definition: Learning through trial and error, rewarded for success.
Example: Self-driving cars - navigating roads, rewarded for safe decisions.

4. Semi-Supervised Learning
Definition: Combining labeled and unlabeled data for better accuracy.
Example: Speech recognition - using labeled audio clips and unlabeled
conversations.

5. Deep Learning
Definition: Using neural networks to analyze complex data.
Example: Voice assistants - understanding natural language commands.

6. Transfer Learning
Definition: Applying pre-trained models to new tasks.
Example: Using pre-trained image recognition models for medical
diagnosis.

7. Ensemble Learning
Definition: Combining multiple models for improved accuracy.
Example: Predicting stock prices using multiple machine learning models.

8. Active Learning
Definition: Selectively requesting labeled data for better accuracy.
Example: Product recommendation - asking users to rate products.
5. Discuss the different types of Artificial Intelligence (AI) and their
applications
1. Reactive AI
• Reactive AI reacts to situations without memory.
• It's like a reflex action.
• Used in self-driving cars and gaming.

2. Limited Memory AI
• Limited Memory AI learns from past experiences.
• It improves over time.
• Applied in chatbots and language translation.

3. Theory of Mind AI
• Theory of Mind AI understands human emotions.
• It empathizes with users.
• Used in social robots and mental health support.

4. Narrow or Weak AI
• Narrow AI excels in one task.
• It's specialized knowledge.
• Found in virtual assistants and image recognition.

5. General or Strong AI
• General AI thinks and learns like humans.
• It's intelligent and adaptable.
• Future applications in robotics.

6. Superintelligence AI
• Superintelligence AI surpasses human intelligence.
• It solves complex problems.
• Potential breakthroughs in science.

7. Hybrid AI
• Hybrid AI combines human and machine intelligence. It enhances
decision-making. Used in healthcare diagnosis.

You might also like