Data Analytics - CSBS Brief Notes, Quiz
& Assignments
UNIT I: Brief Notes
Data Analytics: Process of examining data sets to draw conclusions.
Steps: Gathering → Scrubbing → Analysis → Visualization.
Data Gathering: Collecting data from multiple sources (surveys, IoT, logs).
Data Scrubbing: Cleaning errors, removing duplicates, handling missing data.
Descriptive Analytics: Summarizes historical data using KPIs and charts.
Inferential Statistics: Uses sample data to infer population parameters.
UNIT I: Quiz Questions
1. What are the steps involved in Data Analytics?
2. What is the difference between Data Scrubbing and Data Analysis?
3. Define Descriptive Analytics with examples.
4. What are the uses of inferential statistics?
5. List common data gathering methods.
UNIT I: Assignments
1. Collect a small dataset (manually or online) and perform basic data scrubbing.
2. Create a descriptive report (mean, median, mode, standard deviation) of any sales
dataset.
UNIT II: Brief Notes
Predictive Analytics: Predicting future outcomes using historical data.
Predictive Models: Descriptive (patterns), Decision (actions).
ML Techniques:
- Linear Regression: Predicts continuous value.
- Logistic Regression: Predicts binary outcomes.
- Neural Network: Detects complex patterns.
- SVM: Classifies data with best boundary.
- Naïve Bayes: Probabilistic classifier based on Bayes' theorem.
Privacy: Ensures data confidentiality for users (respondents) and owners (holders).
UNIT II: Quiz Questions
1. What is predictive analytics?
2. Differentiate linear and logistic regression.
3. What is the role of neural networks in prediction?
4. Define respondent privacy.
5. Name three predictive approaches and explain one.
UNIT II: Assignments
1. Build a linear regression model using sample data (Excel or Python).
2. Prepare a comparison table of different ML techniques used in predictive analytics.
UNIT III: Brief Notes
Applications of Predictive Analytics:
- CRM: Analyze customer data to improve service.
- Healthcare: Disease prediction, resource optimization.
- Finance: Risk scoring, fraud detection.
- Marketing: Target segmentation, personalization.
- Business: Forecasting trends, demand planning.
- Fraud Detection: Identifying anomalies in transactions.
UNIT III: Quiz Questions
1. How is predictive analytics used in healthcare?
2. Define Analytical CRM.
3. What are key benefits of predictive analytics in business?
4. Give examples of predictive analytics in finance.
5. How does predictive analytics help in fraud detection?
UNIT III: Assignments
1. Case study: Choose any one sector and document how predictive analytics is used.
2. Design a predictive strategy for a hypothetical e-commerce company.
UNIT IV: Brief Notes
Data Visualization: Presenting data graphically for better understanding.
Chart Types:
- Stacked Bar, Histogram, Butterfly, Donut.
- Scatter, Bubble, Box Plot.
- Pareto, Bump, Maps, Gantt Chart.
Each type is suited for specific kinds of data representation.
UNIT IV: Quiz Questions
1. What is a Pareto chart used for?
2. Differentiate between scatter plot and bubble chart.
3. What does a Gantt chart show?
4. Name two charts used for showing frequency distribution.
5. What is the advantage of a stacked bar chart?
UNIT IV: Assignments
1. Visualize a dataset using any three types of charts in Excel or Power BI.
2. Identify and explain the best visualization chart for project tracking.
UNIT V: Brief Notes
Dashboard: A real-time, visual summary of key metrics.
Types: Operational (daily ops), Strategic (long-term), Analytical (in-depth).
Design Approach: Clarity, relevance, interactivity.
Examples:
- Healthcare Quality Dashboard.
- Airline Performance Dashboard.
- Manufacturing Metrics.
- Warehouse Inventory Dashboard.
UNIT V: Quiz Questions
1. What are the types of dashboards?
2. Explain the purpose of a Healthcare Quality Dashboard.
3. What elements are key to dashboard design?
4. Differentiate strategic and operational dashboards.
5. How do dashboards help in business decision-making?
UNIT V: Assignments
1. Design a sample dashboard layout (draw or use software) for any one sector.
2. Collect KPI examples and match them to appropriate dashboard types.
