srikanthnakka/laptop-price-prediction-ml
Machine learning project to predict laptop prices using linear Regression,Random Forest and Gradient Boosting
Laptop Price Prediction Using Machine Learning
Project Summary
Laptop prices depend on multiple hardware features such as processor type, RAM size, storage capacity, display quality, and brand reputation. Understanding how these specifications influence price can help businesses build competitive pricing strategies and assist consumers in selecting laptops that match their needs and budgets.
This project builds a machine learning regression model to predict laptop prices using hardware specifications and engineered features derived from the dataset.
The project follows a complete end-to-end machine learning workflow, including data cleaning, exploratory analysis, feature engineering, model training, and evaluation.
Project Metrics
| Metric | Value |
|---|---|
| Dataset Size | 1,273 laptops |
| Original Features | 11 |
| Engineered Features | 5+ |
| Machine Learning Task | Regression |
| Libraries Used | Pandas, NumPy, Scikit-Learn |
Dataset Overview
The dataset contains specifications for multiple laptop brands and configurations.
| Feature | Description |
|---|---|
| Company | Laptop brand (Dell, HP, Apple, Lenovo, etc.) |
| TypeName | Laptop category (Gaming, Notebook, Ultrabook, etc.) |
| CPU | Processor model |
| RAM | Memory size |
| Storage | HDD / SSD configuration |
| GPU | Graphics processor |
| Operating System | Installed OS |
| Weight | Laptop weight |
| Display | Screen size and resolution |
| Price | Target variable |
After preprocessing, the dataset contains 1,273 usable laptop records with multiple hardware features used for prediction.
Tools & Technologies
Programming Language
- Python
Libraries Used
- Pandas
- NumPy
- Matplotlib
- Seaborn
- Scikit-Learn
Techniques Applied
- Data Cleaning
- Exploratory Data Analysis
- Feature Engineering
- Correlation Analysis
- Regression Modeling
- Model Evaluation
Data Preprocessing
The raw dataset required several preprocessing steps before training the model.
Missing Value Handling
Rows containing null values were removed to maintain data consistency.
Data Cleaning
Several columns contained text-based values such as:
8GB RAM
1.37kg weight
256GB SSD
These values were converted into numeric features for machine learning.
Column Transformation
Unnecessary columns and duplicate indexing columns were removed to simplify the dataset.
Feature Engineering
Feature engineering played a critical role in improving the predictive power of the model.
1️⃣ Touchscreen Feature
A new binary feature was created to identify laptops that support touchscreen displays.
| Touchscreen | Meaning |
|---|---|
| 1 | Touchscreen laptop |
| 0 | Non-touchscreen laptop |
Touchscreen laptops tend to have higher prices.
2️⃣ IPS Panel Feature
Display specifications were analyzed to determine whether the laptop includes an IPS display panel.
IPS displays generally provide better color accuracy and viewing angles, leading to higher prices.
3️⃣ Display Pixel Density
Screen resolution and screen size were combined to calculate display pixel density (PPI).
Display = sqrt(width² + height²) / screen_size
This provides a better representation of display quality.
4️⃣ CPU Categorization
Processor models were grouped into major categories:
| CPU Category |
|---|
| Intel Core i3 |
| Intel Core i5 |
| Intel Core i7 |
| Other Intel Processors |
| AMD Processors |
This transformation reduces feature complexity while preserving meaningful information.
Exploratory Data Analysis
Exploratory analysis revealed several interesting patterns.
🏷 Brand vs Laptop Price
Brands such as Apple and Razer consistently show higher average laptop prices.
Major laptop brands in the dataset include:
- Lenovo
- Dell
- HP
- Asus
- Apple
These brands also dominate the dataset in terms of product availability.
Laptop Type vs Price
Laptop categories show significant variation in pricing.
| Laptop Type | Pricing Trend |
|---|---|
| Notebook | Most common, moderate price |
| Ultrabook | Premium pricing |
| Gaming | High performance, high price |
| Workstation | Very high price |
RAM vs Price
RAM size has one of the strongest correlations with laptop price.
Higher RAM configurations significantly increase laptop prices.
Display Features vs Price
Display features also influence laptop pricing.
Laptops with:
- Touchscreen displays
- IPS panels
- High-resolution screens
generally have higher market prices.
Machine Learning Model
The project uses regression-based machine learning models to predict laptop prices.
Modeling Workflow
- Data preprocessing and feature transformation
- Feature selection
- Train-test data split
- Model training
- Model evaluation
The trained model learns the relationship between laptop hardware specifications and market prices.
Model Evaluation
Model performance was evaluated using standard regression metrics.
| Metric | Purpose |
|---|---|
| MAE | Average prediction error |
| MSE | Penalizes larger errors |
| R² Score | Measures model accuracy |
These metrics help determine how well the model predicts laptop prices.
🔍 Key Insights
From the analysis, several key insights were discovered.
RAM is the strongest price driver
Higher RAM configurations significantly increase laptop prices.
Premium brands charge higher prices
Apple and Razer laptops appear consistently in the high-price range.
Display quality impacts pricing
High-resolution displays and IPS panels increase laptop value.
Gaming laptops are expensive
Gaming and workstation laptops contain high-performance hardware, leading to higher prices.
Business Applications
Laptop price prediction models have several real-world applications.
- E-commerce price recommendation systems
- Retail product pricing strategies
- Laptop price comparison tools
- Market trend analysis
Skills Demonstrated
This project highlights the following data science skills:
- Data Cleaning and Transformation
- Feature Engineering
- Exploratory Data Analysis
- Data Visualization
- Machine Learning Modeling
- Regression Analysis
- Business Insight Generation
Future Improvements
Possible future enhancements include:
- Deploying the model as a web application
- Creating an interactive laptop price prediction tool
- Training advanced regression models
- Integrating real-time pricing datasets
Conclusion
This project demonstrates how machine learning can be applied to analyze laptop specifications and predict their market prices.
By combining data preprocessing, feature engineering, exploratory analysis, and regression modeling, the project provides valuable insights into how different hardware components influence laptop pricing.