angelpatriciads / mushroom-characteristic-classification

Create a deep neural network to perform the process classification of mushrooms characteristic and using hyperparameter tuning to get the best model. Then, do analysis on accuracy, precision, recall and, f1-score of the model and also determine the features that are strongly suspected of being related with toxics of a mushroom.

Mushroom Characteristic Classification using Deep Learning

-- Project Status: Completed

Project Objective

Create a deep neural network to perform the process classification of mushrooms characteristic and using hyperparameter tuning to get the best model. Then, do analysis on accuracy, precision, recall and, f1-score of the model and also determine the features that are strongly suspected of being related with toxics of a mushroom.

Methods Used

• Deep Learning
• Neural Network

Language

• Python

Module

• Pandas
• Matplotlib
• Numpy
• Seaborn
• Keras
• Tensorflow
• Sklearn

Step-by-step

1. Dataset analysis
2. Cleaning data
   • Checking missing values
   • Checking duplicates data
3. Data Visualization
4. Encoding
   • Data nominal : onehot encoding
   • Data binary : ordinal encoding
5. Sampling data
6. Feature Selection
   • Using Pearson Correlation
7. Data preparation
   • Train-test-val split with the proportion 7.5:1.25:1.25
8. Tuning hyperparameter
   • Best number of layer : 3
   • Best number of neuron : 50
9. Build the model
10. Model evaluation
    • Precision : 0.99
    • Recall : 0.99
    • F1-score : 1.00
    • Accuracy : 99%

Getting Started

1. You can access the raw data here within this repo.
2. All of the scripts are being kept here.