1. Abstract

The video game industry is one of the fastest growing entertainment industries in the world. Game development companies constantly analyze sales data to understand market trends and predict future sales. Predicting video game sales helps companies make better business decisions regarding marketing strategies, production planning, and distribution.

This project focuses on predicting video game sales in different regions using machine learning techniques. The dataset used in this project contains information about video games, including attributes such as game title, platform, genre, publisher, release year, and sales in different regions.

Although the dataset was originally created for data analysis and market trend identification, it can also be used to build predictive models. In this project, machine learning algorithms such as Linear Regression are used to predict the sales of video games in a specific region.

After building the machine learning model, the project continues with the development of a web application using the Django framework. The trained model is integrated into the Django application to allow users to input game details and obtain predicted sales results.

Finally, the web application is deployed on the Heroku cloud platform using GitHub integration, making the system accessible online.

This project demonstrates how machine learning models can be integrated with web frameworks to build real-world predictive applications.

2. Objectives

The main objectives of this project are:

1. To understand video game sales trends using historical data.
2. To perform data preprocessing and feature engineering on the video game dataset.
3. To build a machine learning model for predicting video game sales.
4. To implement Linear Regression algorithm for sales prediction.
5. To evaluate the performance of the prediction model.
6. To develop a web application using Django framework.
7. To integrate the trained machine learning model into the Django application.
8. To deploy the web application on Heroku cloud platform using GitHub.

3. Existing System

Traditionally, video game companies analyze sales data manually or through basic statistical methods to understand market trends.

However, these methods have several limitations:

1. Manual data analysis is time-consuming.
2. Large datasets are difficult to process without automated systems.
3. Traditional analysis methods cannot accurately predict future sales.
4. Market trends and hidden patterns are difficult to identify manually.
5. Decision-making becomes slower without predictive systems.

Therefore, machine learning based prediction systems are required to improve sales forecasting.

4. Proposed System

The proposed system uses Machine Learning and Web Application technologies to predict video game sales.

In this system:

1. A video game dataset containing information about game sales is used for analysis.
2. Data preprocessing and feature engineering techniques are applied to prepare the dataset.
3. A Linear Regression machine learning model is trained to predict video game sales in a specific region.
4. A Django web application is developed to provide a user interface for the prediction system.
5. The trained machine learning model is integrated with the Django application.
6. The final web application is deployed on Heroku cloud platform using GitHub.

This system allows users to easily predict video game sales through an online interface.

5. Implementation Procedure

The implementation of this project is carried out in the following steps:

Step 1: Data Collection

1. Collect the Video Game Sales dataset.
2. The dataset contains attributes related to video games such as platform, genre, publisher, and regional sales.

Step 2: Data Analysis

1. Analyze the dataset to understand sales patterns.
2. Study relationships between game features and sales in different regions.

Step 3: Data Preprocessing

1. Handle missing values.
2. Convert categorical features into numerical values.
3. Prepare the dataset for machine learning model training.

Step 4: Model Building

1. Train a machine learning model using Linear Regression.
2. Use the dataset to predict video game sales in a specific region.

Step 5: Model Evaluation

1. Evaluate the model performance using evaluation metrics such as:
2. Mean Squared Error (MSE)
3. Root Mean Squared Error (RMSE)
4. R² Score

Step 6: Web Application Development

1. Develop a web application using the Django framework.
2. Create a user interface where users can input video game attributes.

Step 7: Model Integration

1. Integrate the trained machine learning model with the Django application.
2. Enable the web application to generate predictions based on user inputs.

Step 8: Deployment

1. Upload the project to GitHub repository.
2. Deploy the Django web application on Heroku cloud platform using GitHub integration.
3. Make the application accessible online.

6. Software Requirements

The software used in this project includes:

Operating System

1. Windows / Linux / macOS

Programming Language

1. Python 3.x

Development Environment

1. Jupyter Notebook
2. VS Code
3. Django Framework

Libraries and Frameworks

1. Scikit-learn
2. NumPy
3. Pandas
4. Matplotlib
5. Django

Deployment Platform

1. Heroku

Version Control

1. GitHub

Web Browser

1. Chrome / Firefox

7. Hardware Requirements

The hardware required for this project includes:

1. Processor: Intel i3 / i5 or higher
2. RAM: Minimum 4 GB (8 GB recommended)
3. Storage: Minimum 128 GB free space
4. System: Laptop / Desktop Computer
5. Internet Connection

8. Advantages of the Project

1. Helps predict video game sales in different regions.
2. Uses machine learning techniques for sales forecasting.
3. Provides an interactive web application using Django.
4. Integrates machine learning models with web technologies.
5. Enables easy access to predictions through a web interface.
6. Deploys the system on cloud platform (Heroku).
7. Helps game companies analyze potential market demand.
8. Can be extended for advanced game market analytics systems.