1. Abstract

Car acceptability prediction is an important task in the automobile industry to evaluate whether a car meets customer expectations after manufacturing or servicing. Predicting the acceptability of a car helps manufacturers and service centers ensure that vehicles meet quality and usability standards before being delivered to customers.

This project focuses on building a Car Acceptability Prediction System using machine learning techniques. The system analyzes various car attributes such as buying price, maintenance cost, number of doors, number of persons the car can accommodate, luggage capacity, and safety level to determine whether the car is acceptable.

The dataset used in this project contains multiple classes such as acceptable, unacceptable, good, and very good, making it a multi-class classification problem. The project involves several steps including data preprocessing, exploratory data analysis (EDA), feature selection, model training, and algorithm evaluation.

Multiple machine learning algorithms such as Random Forest Classifier are used to train the model. The trained model is integrated into a web application using Flask and Bootstrap. The system allows users to input car features through a form and receive predictions about the car's acceptability. Finally, the application is deployed on Google Cloud Platform, making it accessible online.

 2. Objectives

The main objectives of this project are:

1. To analyze car evaluation data using data science techniques.
2. To build a machine learning model that predicts car acceptability.
3. To perform exploratory data analysis to understand the dataset.
4. To select the most suitable machine learning algorithm for classification.
5. To develop a web application interface using Flask and Bootstrap.
6. To integrate the machine learning model with the web application.
7. To deploy the application on Google Cloud Platform.

3.Existing System

In traditional automobile evaluation systems, car acceptability is usually determined manually by engineers, testers, or customers after inspecting various car features such as safety, maintenance cost, and passenger capacity.

Limitations of the Existing System

1. Manual evaluation can be time-consuming.
2. Difficult to analyze large datasets of vehicle features.
3. Human judgments may vary and can introduce bias.
4. Lack of automated systems for predicting car acceptability.
5. Hard to predict customer satisfaction before releasing the vehicle.

Because of these limitations, an automated prediction system is required.

4. Proposed System

The proposed system uses machine learning algorithms to automatically predict the acceptability of a car based on its features.

In this system:

1. A dataset containing different car attributes is used.
2. Data preprocessing is performed to clean and prepare the data.
3. Exploratory Data Analysis (EDA) is conducted to understand the relationships between features.
4. Feature selection techniques are applied to identify the most relevant attributes.
5. Machine learning algorithms are used to train a classification model.
6. The trained model predicts the acceptability category of the car.
7. A web interface is created using Flask and Bootstrap to allow user interaction.
8. The final system is deployed on Google Cloud Platform.

5. Implementation Procedure

The implementation of this project is carried out in several steps.

Step 1: Data Collection

The dataset used for this project contains car attributes such as:

1. Buying price
2. Maintenance cost
3. Number of doors
4. Passenger capacity
5. Luggage capacity
6. Safety level

The dataset includes four target classes:

1. Acceptable
2. Unacceptable
3. Good
4. Very Good

Step 2: Data Loading

The dataset is loaded into data frames using Python libraries such as Pandas.

Step 3: Data Preprocessing

Data preprocessing includes:

1. Cleaning the dataset
2. Handling missing values
3. Converting categorical variables into numerical form
4. Preparing data for machine learning algorithms

Step 4: Exploratory Data Analysis (EDA)

Exploratory Data Analysis is performed to understand the distribution of features.

EDA helps in:

1. Understanding relationships between variables
2. Identifying patterns in the dataset
3. Visualizing feature distributions

For example, features such as maintenance level, safety level, and number of persons significantly influence car acceptability.

Step 5: Feature Selection

Important features that influence the acceptability of the car are selected.

Key features include:

1. Buying price
2. Maintenance cost
3. Number of doors
4. Passenger capacity
5. Luggage size
6. Safety level

Step 6: Model Training

Machine learning algorithms are applied to train the classification model.

Algorithms used may include:

1. Random Forest Classifier
2. Decision Tree Classifier
3. Other classification algorithms

The model learns patterns from the training data to classify car acceptability.

 Step 7: Model Evaluation

The performance of the model is evaluated using accuracy metrics.

For example, the Random Forest model achieved an accuracy of approximately 98.55% in predicting car acceptability.

Step 8: Web Application Development

A Flask web application is created with Bootstrap templates.

The application includes:

1. A user form to input car attributes
2. Data description and dataset information
3. Visualization for exploratory data analysis

Step 9: Prediction Output

Users enter car features such as:

1. Buying price
2. Maintenance level
3. Number of doors
4. Passenger capacity
5. Luggage size
6. Safety level

The system processes the input and predicts whether the car is:

1. Acceptable
2. Unacceptable
3. Good
4. Very Good

Step 10: Deployment

The complete application is deployed on Google Cloud Platform (GCP) so users can access the system through a web browser.

6.Software Requirements

The software required for this project includes:

1. Python Programming Language
2. Jupyter Notebook / VS Code
3. Flask Framework
4. Bootstrap (for web templates)
5. Machine Learning Libraries (Scikit-learn)
6. Pandas and NumPy
7. Google Cloud Platform
8. Web Browser

7.Hardware Requirements

Minimum hardware requirements include:

1. Processor: Intel i3 or higher
2. RAM: 4 GB minimum (8 GB recommended)
3. Storage: 256 GB hard disk or higher
4. System: Laptop or Desktop computer

8.Advantages of the Project

1. Automates the process of evaluating car acceptability.
2. Provides quick and accurate predictions.
3. Helps manufacturers understand vehicle quality before release.
4. Supports multi-class classification of car quality levels.
5. Integrates machine learning with web technologies.
6. Accessible online through cloud deployment.
7. Useful for beginners learning data science and machine learning projects.