Data Analytics Fundamentals: Key Concepts and Exam Preparation

Posted on Oct 18, 2025 in Business Management and Marketing

This document covers the essential concepts from the Data Analytics Class Test (B.Tech 5th Sem, BCS-052) paper, presented in simple, exam-friendly language for easy understanding and recall.

Section A: Fundamental Definitions (5 Marks)

What is Structured Data?

  • Data stored in a fixed format (e.g., tables, rows, columns).
  • It is easy to search and analyze, typically using SQL.
  • Examples: Customer names and phone numbers stored in a relational database.

What is Machine Data?

  • Data generated automatically by machines or devices without direct human input.
  • It originates from sensors, system logs, servers, and IoT devices.
  • Examples: Web server logs, GPS data, and industrial sensor readings.

Defining Apache Hadoop

  • An open-source framework designed for storing and processing massive datasets across clusters of computers.
  • It utilizes HDFS (Hadoop Distributed File System) for storage and MapReduce for parallel processing.
  • Hadoop scales easily, making it crucial for Big Data environments.

Defining Big Data (The 3 V’s)

  • Data that is characterized by being extremely large, fast-moving, or highly diverse, making it unmanageable by traditional data processing tools.
  • It is usually described by the 3 V’s: Volume (size), Velocity (speed of creation/update), and Variety (types of data).
  • Examples: Social media feeds, online transaction records, and sensor streams.

What is Diagnostic Analytics?

  • The type of analytics used to determine the cause of a past event or problem (answering the question: “Why did it happen?”).
  • Methods include drill-down analysis, data discovery, and correlation analysis.
  • Example: An e-commerce site analyzing specific factors that caused a drop in sales last week.

Section B: Core Concepts and Explanations (15 Marks)

The Nature and Characteristics of Big Data

  • Huge Volume: Data measured in terabytes, petabytes, or even zettabytes.
  • High Velocity: Data is created and updated continuously, often requiring real-time processing.
  • Variety: Includes structured, semi-structured (e.g., JSON, XML), and unstructured data (e.g., text, video).
  • Complexity: Requires distributed storage and processing architectures (like Hadoop or Spark).
  • Value: The potential for extracting hidden insights and business intelligence.

Primary Sources of Data for Analytics

Data sources are generally categorized as follows:

  • Internal Sources: Data generated within the organization (e.g., transaction databases, CRM, ERP, HR systems).
  • External Sources: Data obtained from outside the organization (e.g., social media platforms, market research reports, public datasets).
  • Machine Data: Automated data from sensors, logs, and Internet of Things (IoT) devices.
  • Human-Generated Data: Data collected directly from people (e.g., surveys, feedback forms).

Applications of Data Analytics Across Industries

  • Business Intelligence: Customer segmentation, sales forecasting, and optimizing supply chains.
  • Healthcare: Patient monitoring, disease prediction modeling, and optimizing hospital operations.
  • Finance: Real-time fraud detection, risk assessment, and credit scoring.
  • Government & Public Sector: Smart city planning, resource allocation, and policy impact analysis.
  • E-commerce: Developing personalized recommendation systems and dynamic pricing.

The Four Types of Data Analytics (OR)

Data analytics is typically divided into four progressive types:

  1. Descriptive Analytics: Focuses on what happened (e.g., reporting, dashboards).
  2. Diagnostic Analytics: Focuses on why it happened (e.g., root cause analysis).
  3. Predictive Analytics: Focuses on what will happen next (e.g., forecasting, machine learning models).
  4. Prescriptive Analytics: Focuses on what action should be taken (e.g., optimization, decision support).

Methods of Data Collection

  • Primary Data Collection: Gathering data directly from the source (e.g., surveys, interviews, direct sensor readings).
  • Secondary Data Collection: Utilizing data that has already been collected and published (e.g., reports, existing databases, academic papers).
  • Automated Data Capture: Continuous collection via system logs, web tracking, and IoT devices.
  • Web Scraping / APIs: Programmatic methods for extracting data from websites or external services.

Section C: Advanced Topics and Tools (10 Marks)

Detailed Sources of Data for Analytics Projects

A comprehensive list of data sources includes:

  • Operational Databases: Transactional data (sales, inventory, logistics).
  • Social Media Data: User interactions, sentiment analysis, and behavioral patterns (e.g., Facebook, X/Twitter).
  • Machine/Sensor Data: Real-time telemetry from IoT devices, industrial equipment, and network logs.
  • Public/Open Data: Data made available by governments or organizations (e.g., census data, weather portals).
  • Third-Party Vendors: Purchased or licensed data sets (e.g., demographic data, specialized market research).

Key Success Factors for Data Analytics Projects

  • Clear Objectives: The project must start with a well-defined business problem or question.
  • High Data Quality: Ensuring the data used is clean, accurate, reliable, and relevant.
  • Skilled Team Composition: Requires a mix of data engineers, data analysts, data scientists, and crucial domain experts.
  • Appropriate Technology Stack: Selecting the right tools for the scale and complexity (e.g., Hadoop, Spark, cloud platforms).
  • Data Governance and Privacy: Establishing policies for data handling and ensuring compliance with legal regulations (e.g., GDPR, HIPAA).
  • Effective Communication: Translating complex analytical insights into actionable, understandable visualizations and reports for stakeholders.

Modern Data Analytics Tools and Their Features (OR)

Key tools used in contemporary data analysis:

  • Tableau / Power BI: Leading tools for interactive dashboards, data visualization, and business intelligence.
  • Apache Spark: An extremely fast, large-scale data processing engine used for complex computations and machine learning workflows.
  • Python Ecosystem: Highly flexible language utilizing libraries like pandas (data manipulation) and scikit-learn (machine learning).
  • R Language: Primarily used for statistical computing, graphical representation, and advanced statistical modeling.
  • Cloud Data Warehouses: Scalable, managed services like Google BigQuery or AWS Redshift for storing and querying massive datasets.