End-to-end data analytics and machine learning projects completed during the Codveda Data Science internship.
| Level | Task | Dataset | Key Result |
|---|---|---|---|
| Level 1 | Data Cleaning | Iris | Removed duplicates, imputed missing values |
| Level 1 | EDA | Iris | Petal features show 0.96 correlation |
| Level 1 | Visualization | Iris | Bar, line, scatter plots |
| Level 2 | Regression | Boston Housing | R²=0.669, RMSE=$4.93k |
| Level 2 | K-Means Clustering | Iris | 83.7% accuracy, rediscovered species |
| Level 3 | Classification | Customer Churn | 95.2% accuracy, 100% precision |
| Level 3 | NLP Sentiment | Social Media | TextBlob pipeline, 46.7% agreement |
- Trained Decision Tree, Logistic Regression, Random Forest
- Tuned Random Forest with GridSearchCV (135 model fits)
- Achieved 100% precision and 95.2% accuracy
- Zero false positives on 667 test customers
- Discovered 3 species clusters WITHOUT using labels
- 83.7% overall accuracy matching botanical classifications
- Setosa identified with 100% perfect accuracy
- Adjusted Rand Score: 0.6202
- Predicted Boston house prices from 13 features
- R² = 0.669 (explains 66.9% of price variation)
- RMSE = $4.93k average prediction error
- Identified LSTAT and RM as strongest predictors
Python 3.x
pandas — data manipulation
numpy — numerical operations
matplotlib — data visualization
seaborn — statistical visualization
scikit-learn — machine learning
textblob — NLP sentiment analysis
nltk — text preprocessing
wordcloud — text visualization
jupyter — interactive notebooks
├── level1/ # Basic: Cleaning, EDA, Visualization
├── level2/ # Intermediate: Regression, Clustering
�OA�OA�OA├── level3/ # Advanced: Classification, NLP
└── requirements.txt
git clone https://github.com/YOURUSERNAME/codveda-data-analytics-internship
cd codveda-data-analytics-internship
pip install -r requirements.txt
jupyter notebook�OA�OA�OAOrganization: Codveda Technology
�OA�OA�OADomain: Data Analytics / Data Science
Duration: 1 Month
�OA�OA�OA#CodvedaJourney #CodvedaExperience #FutureWithCodveda