1. Abstract

Mushrooms are widely consumed as food across the world, but some species are highly poisonous and can cause serious health risks if consumed. Identifying whether a mushroom is edible or poisonous is an important classification problem in biology and food safety. Manual identification requires expert knowledge and can be difficult for common people.

This project focuses on predicting whether a mushroom is edible or poisonous using Machine Learning techniques. The dataset used in this project contains descriptions of hypothetical samples corresponding to 23 species of gilled mushrooms in the Agaricus and Lepiota family. Each mushroom sample is classified as either edible or poisonous based on its physical characteristics such as cap shape, colour, odour, gill size, and other features.

Since the dataset consists entirely of categorical attributes, preprocessing techniques such as Label Encoding are used to convert categorical data into numerical values that can be processed by machine learning algorithms. A Random Forest Classifier model is used to train the classification system.

Finally, the trained model is integrated into a Django-based web application, allowing users to input mushroom characteristics and determine whether the mushroom is edible or poisonous. The application is then deployed on Heroku using GitHub, enabling online access to the prediction system.

This project demonstrates the practical application of machine learning in biological classification and food safety.

2. Objectives

The main objectives of this project are:

1. To understand the mushroom classification problem using machine learning.
2. To study different mushroom characteristics and their impact on classification.
3. To analyze a mushroom dataset containing categorical features.
4. To preprocess categorical data using Label Encoding techniques.
5. To understand classification algorithms in machine learning.
6. To implement a Random Forest Classifier for mushroom classification.
7. To evaluate the performance of the classification model.
8. To deploy the trained model using a Django-based web application.

3. Existing System

In the existing system, mushroom identification is typically performed through:

1. Manual inspection by experts
2. Botanical classification methods
3. Field guides and reference books
4. Visual pattern recognition

Limitations of Existing Systems

1. Requires expert knowledge in mycology.
2. Difficult for non-experts to correctly identify mushrooms.
3. Manual identification is time-consuming.
4. Risk of incorrect classification leading to consumption of poisonous mushrooms.
5. No automated tools for quick classification.

These limitations highlight the need for an automated machine learning-based classification system.

4. Proposed System

The proposed system uses Machine Learning techniques to automatically classify mushrooms as edible or poisonous.

In this system:

1. Mushroom dataset is collected containing categorical attributes.
2. Data preprocessing is performed using Label Encoding to convert categorical features into numerical format.
3. A Random Forest Classifier model is trained using the dataset.
4. The model learns patterns between mushroom characteristics and their class labels.
5. The trained model predicts whether a mushroom is edible or poisonous.
6. The prediction system is integrated into a Django web application.
7. The web application is deployed on Heroku using GitHub.

This system provides an automated, accurate, and user-friendly solution for mushroom classification.

5. Implementation Procedure

The implementation of this project consists of the following steps:

Step 1: Data Collection

The mushroom dataset is collected. It includes descriptions of 23 species of mushrooms belonging to the Agaricus and Lepiota families.

The dataset contains attributes such as:

1. Cap shape
2. Cap colour
3. Odour
4. Gill size
5. Gill colour
6. Stalk shape
7. Habitat
8. Population

Step 2: Data Preprocessing

The dataset is prepared by:

1. Checking for missing values
2. Encoding categorical variables using Label Encoder
3. Converting categorical attributes into numerical values

Step 3: Exploratory Data Analysis (EDA)

1. Understanding the distribution of edible and poisonous mushrooms
2. Identifying relationships between features
3. Visualizing categorical feature patterns

Step 4: Feature Engineering

1. Selecting relevant attributes
2. Preparing encoded features for model training

Step 5: Model Development

A Random Forest Classifier model is developed including:

1. Training dataset preparation
2. Model training using encoded mushroom features
3. Learning patterns for classification

Step 6: Model Evaluation

The performance of the model is evaluated using metrics such as:

1. Accuracy Score
2. Confusion Matrix
3. Precision and Recall

Step 7: Model Deployment

1. The trained model is integrated into a Django framework
2. A web interface is created
3. Users input mushroom characteristics
4. The system predicts whether the mushroom is edible or poisonous

6. Software Requirements

The software tools used in this project include:

1. Python – Programming language
2. Jupyter Notebook / Google Colab – Development environment
3. Django – Web framework for application development
4. NumPy – Numerical computation
5. Pandas – Data manipulation and analysis
6. Matplotlib / Seaborn – Data visualization
7. Scikit-learn – Machine learning algorithms
8. Pickle / Joblib – Model saving and loading
9. GitHub – Version control system
10. Heroku – Cloud deployment platform

 7. Hardware Requirements

       Minimum Hardware Requirements:

1. Processor: Intel i5 or higher
2. RAM: 8 GB or higher
3. Storage: 256 GB or higher
4. Laptop or Desktop Computer
5. Internet connection for deployment

8. Advantages of the Project

1. Provides accurate classification of mushrooms.
2. Helps prevent accidental consumption of poisonous mushrooms.
3. Automates the mushroom identification process.
4. Easy-to-use web interface for users.
5. Reduces dependency on expert knowledge.
6. Demonstrates practical application of machine learning in biological classification.
7. Deployable as a real-time web application.