📚 Data Science Lab Assignments

📑 Table of Contents

1. Overview
2. Assignments
3. Experiments
4. Installation
5. Repository Structure
6. Author

🎯 Overview

This repository contains five data science assignments and seventeen experiments covering fundamental statistical analysis, machine learning, and data visualization techniques.

📊 Assignments

Assignment	Topic	Key Concepts	Tools Used
Assignment-1	Exploratory Data Analysis	Data Loading, Feature Analysis, Visualization	Pandas, Seaborn
Assignment-2	Confusion Matrix	Classification, Model Evaluation, Metrics	Scikit-learn, Matplotlib
Assignment-3	Z-Test Analysis	Hypothesis Testing, Statistical Significance	Statsmodels
Assignment-4	T-Test Analysis	Independent & Paired Tests, P-value Analysis	SciPy
Assignment-5	Linear Regression	Simple & Multiple Regression, Model Evaluation	Scikit-learn

🔬 Experiments

Experiment	Topic	Key Concepts	Tools Used
Experiment-1	Linear Regression with Scikit-learn	Model Training, Evaluation, Visualization	Scikit-learn, Matplotlib
Experiment-2	Singular Value Decomposition (SVD)	Matrix Decomposition, Reconstruction	NumPy
Experiment-3	Logistic Regression Visualization	Decision Boundary, Classification	Scikit-learn, Matplotlib
Experiment-4	T-Test Analysis with SciPy	Statistical Hypothesis Testing	SciPy
Experiment-5	Confusion Matrix Visualization	Model Evaluation, Metrics	Scikit-learn, Seaborn
Experiment-6	Decision Tree Classifier	Decision Trees, Visualization	Scikit-learn, Matplotlib
Experiment-7	[Pending Implementation]	-	-
Experiment-8	K-Means Clustering with Elbow Method	Clustering, Optimal Cluster Selection	Scikit-learn, Matplotlib
Experiment-9	DBSCAN Clustering on Customer Data	Density-based Clustering, Outlier Detection	Scikit-learn, Seaborn
Experiment-10	Gradient Boosting Classifier	Ensemble Learning, Boosting	Scikit-learn
Experiment-11	Ensemble Voting Classifier	Ensemble Learning, Voting	Scikit-learn
Experiment-12	Naive Bayes Classifier	Probabilistic Classification	Scikit-learn
Experiment-13	Linear Discriminant Analysis (LDA)	Dimensionality Reduction, Classification	Scikit-learn
Experiment-14	Hierarchical Clustering with Dendrogram	Hierarchical Clustering, Visualization	SciPy, Matplotlib
Experiment-15	Logistic Regression with Decision Boundary	Classification, Visualization	Scikit-learn, Matplotlib
Experiment-16	Hierarchical Clustering with Dendrogram	Hierarchical Clustering, Visualization	SciPy, Matplotlib
Experiment-17	[Pending Implementation]	-	-

For a detailed overview of all experiments, see Data Science Lab Experiments.

🛠️ Installation

# Clone the repository
git clone https://github.com/Mausam5055/Data-Science.git

# Navigate to the directory
cd Data-Science

# Install required packages
pip install -r Assignment-1/requirements.txt

📁 Repository Structure

.
├── Assignment-1/
│   ├── run_titanic_eda.py
│   ├── titanic.csv
│   ├── README.md
│   └── requirements.txt
├── Assignment-2/
│   ├── confusion_matrix_iris.py
│   └── README.md
├── Assignment-3/
│   ├── ztest_demo.py
│   └── README.md
├── Assignment-4/
│   ├── ttest_demo.py
│   └── README.md
├── Assignment-5/
│   ├── linear_regression_demo.py
│   └── README.md
├── Experiment-1/
│   └── main.py
├── Experiment-2/
│   └── main.py
├── Experiment-3/
│   └── main.py
├── Experiment-4/
│   └── main.py
├── Experiment-5/
│   └── main.py
├── Experiment-6/
│   └── main.py
├── Experiment-7/
│   └── main.py
├── Experiment-8/
│   └── main.py
├── Experiment-9/
│   └── main.py
├── Experiment-10/
│   └── main.py
├── Experiment-11/
│   └── main.py
├── Experiment-12/
│   └── main.py
├── Experiment-13/
│   └── main.py
├── Experiment-14/
│   └── main.py
├── Experiment-15/
│   └── main.py
├── Experiment-16/
│   └── main.py
├── Experiment-17/
│   └── main.py
└── FDS_LAB MANUAL.odt

📌 Assignment Details

1. Exploratory Data Analysis

• Dataset: Titanic Dataset
• Key Features: Passenger Demographics, Survival Analysis
• Visualizations: Count plots, Histograms, Correlation matrices

2. Confusion Matrix

• Dataset: Iris Dataset
• Model: Logistic Regression
• Metrics: Accuracy, Precision, Recall, F1-Score

3. Z-Test Analysis

• Implementation: One-sample Z-test
• Tools: Statsmodels
• Analysis: Z-score, P-value interpretation

4. T-Test Analysis

• Types: Independent and Paired T-tests
• Tools: SciPy
• Focus: Statistical significance testing

5. Linear Regression

• Types: Simple and Multiple Linear Regression
• Metrics: R² Score, MSE
• Features: Data generation, Model training, Prediction

🔬 Experiment Details

Experiment-1: Linear Regression with Scikit-learn

• Concept: Implementation of linear regression using scikit-learn
• Key Features: Data generation, model training, evaluation, and visualization

Experiment-2: Singular Value Decomposition (SVD)

• Concept: Matrix decomposition technique
• Key Features: Decomposition of a matrix and reconstruction from components

Experiment-3: Logistic Regression Visualization

• Concept: Visualization of logistic regression results
• Key Features: Decision boundary plotting (incomplete implementation)

Experiment-4: T-Test Analysis with SciPy

• Concept: Statistical hypothesis testing
• Key Features: Two-sample t-test implementation and interpretation

Experiment-5: Confusion Matrix Visualization

• Concept: Model evaluation technique
• Key Features: Confusion matrix creation and visualization using seaborn

Experiment-6: Decision Tree Classifier

• Concept: Decision tree algorithm for classification
• Key Features: Tree visualization, model evaluation with accuracy metrics

Experiment-7: [Pending Implementation]

• Status: Empty experiment requiring implementation

Experiment-8: K-Means Clustering with Elbow Method

• Concept: Unsupervised learning clustering technique
• Key Features: Optimal cluster selection using the elbow method

Experiment-9: DBSCAN Clustering on Customer Data

• Concept: Density-based clustering algorithm
• Key Features: Outlier detection and cluster visualization

Experiment-10: Gradient Boosting Classifier

• Concept: Ensemble learning technique
• Key Features: Boosting algorithm for classification

Experiment-11: Ensemble Voting Classifier

• Concept: Ensemble learning through voting
• Key Features: Combining multiple classifiers for improved performance

Experiment-12: Naive Bayes Classifier

• Concept: Probabilistic classification algorithm
• Key Features: Gaussian Naive Bayes with detailed evaluation metrics

Experiment-13: Linear Discriminant Analysis (LDA)

• Concept: Dimensionality reduction technique
• Key Features: LDA for feature reduction followed by classification

Experiment-14: Hierarchical Clustering with Dendrogram

• Concept: Hierarchical clustering algorithm
• Key Features: Dendrogram visualization of clustering results

Experiment-15: Logistic Regression with Decision Boundary

• Concept: Binary classification with visualization
• Key Features: Decision boundary plotting for logistic regression

Experiment-16: Hierarchical Clustering with Dendrogram

• Concept: Hierarchical clustering algorithm
• Key Features: Dendrogram visualization of clustering results

Experiment-17: [Pending Implementation]

• Status: Empty experiment requiring implementation

🔧 Requirements

Package	Purpose
NumPy	Numerical computations
Pandas	Data manipulation
Matplotlib	Visualization
Scikit-learn	Machine learning
SciPy	Statistical analysis
Statsmodels	Statistical models
Seaborn	Advanced visualization

👤 Author

Mausam Kar