Advanced Machine Learning Techniques for Waste Classification and Employee Ratings Prediction

This repository contains the code for training and evaluating a Convolutional Neural Network (CNN) to classify waste images from the RealWaste dataset. The RealWaste dataset can be downloaded from UCI Machine Learning Repository.

Project Overview

The project is divided into two parts, each documented in the provided Jupyter notebooks and PDFs:

Part 1: Detailed in ML2_Part1_Machine-Learning-for-Business-Intelligence-2--Spring,-2024-.pdf and its corresponding R Markdown file.
Part 2: Detailed in ML2_Part2_Machine Learning for Business Intelligence.pdf and its corresponding Jupyter notebook ML2_Part2_Machine Learning for Business Intelligence.ipynb.

Dataset

To use this repository, you need to download the RealWaste dataset and place it in the root folder of the project. The dataset should be organized as follows:

root/
│
├── RealWaste/
│ ├── Cardboard/
│ ├── Food Organics/
│ ├── Glass/
│ ├── Metal/
│ ├── Miscellaneous Trash/
│ ├── Paper/
│ ├── Plastic/
│ ├── Textile Trash/
│ └── Vegetation/
│
├── RealWaste_Sampled/
│ ├── Cardboard/
│ ├── Food Organics/
│ ├── Glass/
│ ├── Metal/
│ ├── Miscellaneous Trash/
│ ├── Paper/
│ ├── Plastic/
│ ├── Textile Trash/
│ └── Vegetation/
│
├── Data.xls
├── ML2_Part1_Machine-Learning-for-Business-Intelligence-2--Spring,-2024-.pdf
├── ML2_Part1_Machine-Learning-for-Business-Intelligence-2--Spring,-2024-.Rmd
├── ML2_Part2_Machine Learning for Business Intelligence.pdf
├── ML2_Part2_Machine Learning for Business Intelligence.ipynb
├── requirements.txt
└── README.md

Instructions

Download and Prepare the Dataset
- Download the RealWaste dataset from UCI Machine Learning Repository.
- Extract the dataset and place the RealWaste folder in the root directory of this repository.
Install Required Libraries
- Install the necessary Python libraries by running:
```
pip install -r requirements.txt
```
Run the Jupyter Notebooks
- Open and run all cells in ML2_Part2_Machine Learning for Business Intelligence.ipynb. This notebook includes all steps from loading the dataset to evaluating the trained model.
Run the R Markdown File
- Open and run the R Markdown file ML2_Part1_Machine-Learning-for-Business-Intelligence-2--Spring,-2024-.Rmd to execute Part 1 of the project.

Results

The trained model achieves a validation accuracy of approximately 52.54%. Detailed results and model evaluation metrics are provided in the respective notebooks and PDF documents.

Contributing