CarQuest

Project Overview:

This program is a Car Recommender System designed to help users find suitable cars based on specific criteria, such as brand, price range, and location. By processing user input and querying a database, the system retrieves and displays matching vehicles from the dataset. Additionally, it answers user queries, like identifying the most fuel-efficient car, offering valuable insights for informed decision-making.

View the Project here: https://github.com/azrabano23/carquest

Technical Overview:

Features:

Dynamic Car Recommendations

  • Allows users to search for cars by brand, price range, and city.

  • Queries an SQLite database to retrieve and display matching vehicles.

Data Insights

  • Provides additional insights by answering questions, such as identifying the most fuel-efficient car in the dataset.

Model Integration

  • Trains a Random Forest Regressor to predict car prices based on features like brand, fuel type, mileage, and transmission.

  • Evaluates the model using metrics like Mean Squared Error (MSE) and R-squared to ensure accuracy.

User Interaction

  • Collects user input for brand, price range, and city to tailor recommendations.

  • Handles invalid inputs with error messages and guidance for users.

Database Management

  • Saves preprocessed data to an SQLite database for efficient querying and storage.

Programming Insights:

  • Coded in Python

    • Pandas: For data cleaning, preprocessing, and analysis.

    • Scikit-Learn: For training and evaluating the Random Forest Regressor model.

    • SQLite3: To create and manage a database for car storage and queries.

    • LabelEncoder: Encodes categorical variables, such as brand and city, for machine learning and database querying.

  • Database Querying

    • Converts user inputs into encoded values for efficient querying.

    • Utilizes SQL queries to filter cars based on user-specified criteria.

  • Machine Learning Pipeline

    • Trains a regression model to predict car prices based on multiple features.

    • Splits the data into training and testing sets to validate performance.

  • Interactive Features

    • Enables dynamic user interactions to tailor car recommendations.

    • Answers specific queries, such as finding the most fuel-efficient car.

Key Applications:

  1. Personalized Car Recommendations: Helps users find cars that match their budget and preferences.

  2. Market Insights: Provides valuable data insights, such as identifying fuel-efficient cars or price trends.

  3. Enhanced Customer Experience: Supports car buyers in making informed decisions by combining analytics and personalized recommendations.

This project demonstrates a strong integration of data science, machine learning, and database management, providing a user-centric solution for car recommendations and automotive insights.