Data Analytics - CSBS Notes
UNIT I: Data Analytics
Data Analytics: The science of analyzing raw data to make conclusions about
information.
Steps in Data Analytics:
• Data Gathering – Collecting data from various sources.
• Data Scrubbing – Cleaning and preparing data for analysis.
• Data Analysis – Applying statistical and logical techniques.
Descriptive Analytics: Summarizes past data to understand what has happened.
Uses: Business reporting, performance tracking.
Measures: Mean, Median, Mode, Standard Deviation.
Inferential Statistics: Drawing conclusions from sample data using probability theory.
UNIT II: Predictive Analytics
Definition: Using statistical models and forecasts to predict future outcomes.
Predictive Models:
• Descriptive Modeling – Groups data based on similarity.
• Decision Modeling – Provides recommendations for action.
Machine Learning Techniques:
• Linear Regression – Predicts numeric outcomes based on input variables.
• Logistic Regression – Predicts binary outcomes (e.g., yes/no).
• Neural Network – Mimics brain structure to detect complex patterns.
• Support Vector Machines – Finds the best boundary to classify data.
• Naïve Bayes – Applies Bayes' theorem for classification tasks.
Privacy & Disclosure:
• Respondent Privacy – Individual level privacy concerns.
• Holder Privacy – Data owner’s privacy.
Analytical Methods:
• Data-driven, Computation-driven, Result-driven approaches.
UNIT III: Application of Predictive Analytics
Applications:
• Analytical CRM – Using analytics for better customer relationships.
• Healthcare – Predicting diseases, optimizing treatments.
• Financial Sector – Credit scoring, risk analysis.
• Business – Forecasting trends and customer behavior.
• Marketing Strategies – Targeted marketing, customer segmentation.
• Fraud Detection – Identifying anomalies in financial transactions.
UNIT IV: Data Visualization
Purpose: Conveys information clearly through charts and graphs.
Charts:
• Stacked Bar Chart – Shows part-to-whole relationships.
• Histogram – Displays frequency distributions.
• Butterfly Chart – Comparison across categories.
• Donut Chart – Similar to pie chart but with center cut out.
• Scatter Plot – Shows relationships between two variables.
• Bubble Chart – Adds third variable through bubble size.
• Box Plot – Shows data distribution through quartiles.
• Pareto Chart – Combines bar and line charts.
• Bump Chart – Shows rank over time.
• Maps – Geographical data visualization.
• Gantt Chart – Visualizes project schedules.
UNIT V: Dashboards
Dashboards: Visual displays of key metrics and trends.
Types of Dashboards: Operational, Strategic, Analytical.
Design Approach: Clarity, simplicity, interactive elements.
Examples:
• Healthcare Quality Dashboard – Tracks patient outcomes and hospital metrics.
• Airline Quality Dashboard – Monitors delays, cancellations, service.
• Manufacturing Quality Dashboard – Tracks production quality and defects.
• Warehouse Quality Dashboard – Measures inventory, delivery performance.
