🧠 OIBSIP — Oasis Infobyte Data Science Internship

A collection of 4 end-to-end Data Science projects completed as part of the Oasis Infobyte internship program. Each task covers a real-world problem — from classification and regression to NLP — with full EDA, model training, evaluation, and (for Task 4) a live Streamlit web app.

---

🗂️ Projects Overview

#	Project	Type	Key Algorithm	Live App
1	Iris Flower Classification	Classification	Logistic Regression, KNN, SVM	—
2	Unemployment Analysis	EDA / Time Series	Pandas, Seaborn	—
3	Car Price Prediction	Regression	Linear Regression, Random Forest	—
4	Email Spam Detection	NLP / Classification	Naive Bayes + TF-IDF	✅ Streamlit

---

📁 Repository Structure

OIBSIP/
├── .devcontainer/
│   └── devcontainer.json            # GitHub Codespaces config (auto-runs Streamlit)
│
├── DataScience-Task1-IrisFlowerClassification/
│   ├── Muhammad_Danish_Iris_Flower_Classification.ipynb
│   ├── README.md
│   ├── output/                      # saved plots & results
│   └── screenshots/
│
├── DataScience-Task2-UnemploymentAnalysis/
│   ├── Unemployment_Analysis.ipynb
│   ├── Unemployment_Rate_upto_11_2020.csv
│   ├── README.md
│   ├── outputs/
│   └── screenshots/
│
├── DataScience-Task3-CarPricePrediction/
│   ├── notebook/                    # Jupyter notebook
│   ├── dataset/                     # car data.csv (CarDekho)
│   ├── outputs/                     # saved model & metrics
│   ├── images/                      # EDA screenshots
│   ├── requirements.txt
│   └── README.md
│
└── DataScience-Task4-EmailSpamDetection/
    ├── Email_Spam_Detection.ipynb
    ├── app/
    │   ├── app.py                   # Streamlit web app
    │   └── requirements.txt
    ├── dataset/
    │   └── spam.csv
    ├── outputs/                     # model metrics & reports
    └── README.md

---

Task 1 — Iris Flower Classification

Goal: Classify Iris flowers into Setosa, Versicolor, and Virginica using petal/sepal measurements.

• Dataset loaded directly from sklearn.datasets
• EDA: pairplots, boxplots, descriptive statistics, null checks
• Models: Logistic Regression, K-Nearest Neighbors, Support Vector Machine
• Evaluation: Accuracy score, classification report, confusion matrix

📂 DataScience-Task1-IrisFlowerClassification/

---

Task 2 — Unemployment Analysis with Python

Goal: Analyze unemployment trends across Indian states and measure the COVID-19 impact.

• Dataset: Kaggle — Unemployment in India (up to Nov 2020)
• EDA: time series analysis, regional comparison, correlation heatmap
• Key finding: Sharp spike in unemployment during the COVID-19 lockdown period

📂 DataScience-Task2-UnemploymentAnalysis/

---

Task 3 — Car Price Prediction

Goal: Predict the selling price of used cars using regression models.

• Dataset: CarDekho Vehicle Dataset (Kaggle)
• Features: Brand, Car Age, Present Price, KMs Driven, Fuel Type, Transmission, Owners
• Models: Linear Regression, Random Forest Regressor
• Metrics: MAE, RMSE, R² Score
• Best model saved with Joblib

📂 DataScience-Task3-CarPricePrediction/

---

Task 4 — Email Spam Detection

Goal: Classify SMS messages as Spam or Ham using NLP + Machine Learning, with a live Streamlit app.

• Dataset: SMS Spam Collection Dataset
• Pipeline: Text cleaning → Tokenization → Stopword removal → Stemming → TF-IDF → Naive Bayes
• Metrics: Accuracy, Precision, Recall, F1-Score, Confusion Matrix
• Deployed via GitHub Codespaces — opens automatically on port 8501

📂 DataScience-Task4-EmailSpamDetection/

Run the Streamlit App Locally

cd DataScience-Task4-EmailSpamDetection
pip install -r app/requirements.txt
streamlit run app/app.py

Then visit http://localhost:8501 in your browser.

Run in GitHub Codespaces

Click "Open in Codespaces" — the environment installs all dependencies and launches the Streamlit app automatically.

---

🛠️ Tech Stack

Category	Tools
Language	Python 3.11
Notebooks	Jupyter Notebook
Data	Pandas, NumPy
Visualization	Matplotlib, Seaborn
ML	Scikit-learn, Joblib
NLP	NLTK
Web App	Streamlit
Environment	GitHub Codespaces (.devcontainer)

---

🚀 Getting Started (Local)

git clone https://github.com/alphaxt/OIBSIP.git
cd OIBSIP

# install deps for Task 3
pip install -r DataScience-Task3-CarPricePrediction/requirements.txt

# install deps for Task 4 + run app
pip install -r DataScience-Task4-EmailSpamDetection/app/requirements.txt
streamlit run DataScience-Task4-EmailSpamDetection/app/app.py

For Tasks 1 & 2, open the .ipynb files in Jupyter or VS Code — no extra requirements beyond standard data science libraries.

---

📜 License

This project is open-source under the MIT License.

---

Developed by Muhammad Danish Oasis Infobyte Data Science Internship