Devices Price Classification System

Project Overview

The Devices Price Classification System is an AI-driven application designed to predict mobile device prices based on their specifications. This system utilizes machine learning to classify prices into categories (0-3), helping sellers accurately price their products.

Project Components

This project consists of two main components:

1. Mobile Phone Price Prediction API:

   • Provides endpoints to manage and predict mobile phone prices based on specifications.
   • Built using Django REST Framework.

2. Mobile Device Price Classification System:

   • Implements machine learning algorithms to classify mobile device prices.
   • Built using Python and various machine learning libraries.

API Documentation

Overview

The API allows users to create, retrieve, and predict mobile phone records.

Base URL

• http://your-domain.com/ or http://localhost:8000/

Endpoints

1. List/Create Mobile Phones

   • URL: /api/devices/
   • Methods: GET, POST
   • Description: Retrieve a list of all mobile phones or create a new entry.

   GET Response Example:

   [
       {
           "id": 1,
           "battery_power": 1000,
           "blue": 1,
           "clock_speed": 2.2,
           ...
           "price_range": 3
       }
   ]

   POST Request Body Example:

   {
       "battery_power": 1000,
       "blue": 1,
       "clock_speed": 2.2,
       "dual_sim": 1,
       "fc": 8,
       "four_g": 1,
       "int_memory": 64,
       "m_dep": 0.7,
       "mobile_wt": 180,
       "n_cores": 8,
       "pc": 12,
       "px_height": 1920,
       "px_width": 1080,
       "ram": 4096,
       "sc_h": 15.5,
       "sc_w": 7.5,
       "talk_time": 20,
       "three_g": 1,
       "touch_screen": 1,
       "wifi": 1,
       "price_range": null
   }

2. Retrieve Mobile Phone

   • URL: /api/devices/<device_id>/
   • Methods: GET
   • Description: Retrieve an individual mobile phone record.

3. Predict Price

   • URL: /api/predict/<device_id>/
   • Method: POST
   • Description: Predict the price and save the result in the device entity.

Data Models

The API uses the following data model for mobile phones:

Field	Type	Description
battery_power	Integer	Total energy a battery can store in mAh
blue	Integer	Has Bluetooth or not (0/1)
clock_speed	Float	Speed at which microprocessor executes
dual_sim	Integer	Has dual sim support or not (0/1)
...	...	...
price_range	Integer	Price range category (0-3)

Price Range Categories

• 0: Low Cost
• 1: Medium Cost
• 2: High Cost
• 3: Very High Cost

Machine Learning Implementation

Overview

The machine learning component of the project classifies mobile device prices based on their specifications.

Data Processing

• Datasets are loaded from CSV files (train.csv and test.csv).
• Null values are removed, and features are standardized.
• Data is split into training (70%), validation (15%), and testing (15%).

Model Architecture

• Algorithm: Logistic Regression
• Performance: Achieved 97% accuracy on test data with an average confidence score of 0.92.

Model Persistence

• The model and scaler are saved using joblib for future predictions.
• Best Model
• Standard Scaler

Conclusion

This project provides a comprehensive solution for mobile device price classification, integrating a robust API with a powerful machine learning model. The system is designed to assist sellers in accurately pricing their products based on detailed specifications.

For further details, please refer to the individual documentation files:

• Mobile Phone Price Prediction API Documentation
• Mobile Device Price Classification System Documentation