Titanic EDA — Step-by-step:
1. Load data using pandas
2. Check for missing values
3. Visualize survival by gender, class
4. Create age distribution plots
5. Build correlation heatmap
6. Generate summary statistics
7. Write key insights

Key insights:
- Women and children had higher survival
- First-class passengers had better survival
- Age was a significant factor

COVID-19 Data Analysis:
1. Load daily case data
2. Clean and format dates
3. Calculate 7-day rolling averages
4. Plot trends by country
5. Create animated choropleth maps
6. Analyze growth rates
7. Compare vaccination vs cases

Tools:
- Pandas, Plotly, Geopandas

Sales Data Analysis:
1. Load and clean sales data
2. Calculate total revenue by region
3. Find top-selling products
4. Analyze monthly trends
5. Customer segmentation
6. Create pivot tables
7. Generate business recommendations

Key outputs:
- Revenue dashboard
- Customer segments
- Product performance matrix

Tableau Sales Dashboard:
1. Connect to data source (CSV/Excel)
2. Create calculated fields
3. Build bar charts, line charts
4. Add filters and parameters
5. Create dashboard layout
6. Add interactive actions
7. Publish to Tableau Public

Key elements:
- Total Revenue (KPI card)
- Monthly sales trend (line chart)
- Sales by region (map)
- Top products (bar chart)
- Year-over-year comparison

Power BI HR Dashboard:
1. Import data from multiple sources
2. Clean and transform with Power Query
3. Build DAX measures
4. Create relationship model
5. Design visuals: cards, charts, matrix
6. Add slicers for filtering
7. Publish to Power BI Service

Key elements:
- Headcount by department
- Attrition rate
- Age & gender distribution
- Tenure analysis
- Drill-through pages

Python Dashboard (Streamlit):
1. Load and process data
2. Create matplotlib/seaborn plots
3. Build interactive widgets (sliders, dropdowns)
4. Use plotly for interactive visuals
5. Deploy on Streamlit Cloud

Key elements:
- Data preview
- Interactive filters
- Dynamic charts
- Summary statistics
- Download capability

Example:
- COVID-19 tracker dashboard
- Stock market visualizer

SQL Data Analysis Project:
1. Set up a database (SQLite/PostgreSQL)
2. Load sample data (e.g., Northwind)
3. Write queries to answer business questions:
   - Top 10 customers by revenue
   - Monthly sales trend
   - Product category performance
4. Use window functions for rankings
5. Create views for reporting
6. Export results to CSV

Sample queries:
- SELECT c.name, SUM(o.total) FROM customers c
  JOIN orders o ON c.id = o.customer_id
  GROUP BY c.id ORDER BY SUM(o.total) DESC;

ETL Pipeline with Python:
1. Extract: Read data from CSV/API/DB
2. Transform: Clean, aggregate, join
3. Load: Write to database or CSV

Example: E-commerce data pipeline
- Extract: Sales data from API
- Transform: Clean missing values, calculate metrics
- Load: Store in PostgreSQL

Tools:
- Python (pandas, requests, sqlalchemy)
- Airflow (optional for scheduling)
- Docker for containerization

Sentiment Analysis Project:
1. Collect product reviews (Amazon API)
2. Clean text: remove punctuation, stopwords
3. Use TextBlob or VADER for sentiment
4. Analyze sentiment distribution
5. Create word clouds (positive/negative)
6. Correlate sentiment with ratings
7. Generate insights

Output:
- Sentiment score per product
- Word cloud visualization
- Recommendation based on sentiment

Tools:
- nltk, textblob, wordcloud, pandas