This data science project implements linear and logistic regression models to solve two practical business problems: restaurant profit prediction based on city population and candidate hiring prediction based on interview scores. The implementation uses Python with scikit-learn, pandas, and matplotlib to build robust predictive models with clear visualizations.
✔ Linear Regression Analysis – Predicts restaurant profitability based on population data of cities.
✔ Logistic Regression Classification – Determines hiring decisions based on interview performance scores.
✔ Data Visualization – Creates scatter plots and regression lines to illustrate model performance.
✔ Multi-class Classification Explanation – Detailed explanation of One-vs-Rest and One-vs-One strategies.
📌 Python Script (main.py) – Implements linear and logistic regression models with visualizations.
📌 Datasets (RegressionData.csv, LogisticRegressionData.csv) – Contains training data for both regression tasks.
📌 Documentation – Includes detailed explanations of multi-class classification approaches.
The first part of the project focuses on:
- Loading and preprocessing population and profit data
- Visualizing data distribution through scatter plots
- Implementing linear regression using least squares optimization
- Predicting restaurant profitability in new locations
- Interpreting model coefficients (intercept and slope)
The second part of the project demonstrates:
- Binary classification of job candidates based on two interview scores
- Visualization of hiring decisions with color-coded scatter plots
- Training a logistic regression model to predict hiring outcomes
- Evaluating model performance on training data
- Visualizing classification errors
The project concludes with detailed explanations of:
- One-vs-Rest (OvR) method – Training N binary classifiers for N classes
- One-vs-One (OvO) method – Training N*(N-1)/2 binary classifiers for pairwise comparisons
- Strengths and limitations of each approach
🔹 Python (scikit-learn, pandas, matplotlib)
🔹 Linear Regression (Least squares optimization)
🔹 Logistic Regression (Binary classification)
🔹 Data Visualization (Scatter plots, regression lines)
1️⃣ The script loads and preprocesses data from CSV files.
2️⃣ Linear regression model fits the relationship between city population and restaurant profit.
3️⃣ Logistic regression model learns to classify candidates based on interview scores.
4️⃣ Visualizations illustrate data patterns and model performance.
5️⃣ Detailed comparisons of multi-class classification approaches are provided.
📊 Setup: pip install pandas scikit-learn matplotlib numpy
🚀 Execute: python main.py
📈 Results: View console output and visualizations showing linear regression coefficients, profit predictions, and logistic regression classification results