Attentiveness Detection in a Metaverse Classroom

For my final year project, I worked on developing a Machine Learning model to detect the attentiveness of students in a Metaverse classroom using sensory data from Meta Quest 2. We created our own dataset by conducting a Continuous Performance Test (CPT) inside the VR classroom. Our final model achieved a 21.92% mean absolute percentage error.

Table of Contents

• Introduction
• Dataset
• Technologies and Libraries Used
• Methodologies and Techniques Applied
• Installation
• Usage
• Contributing
• License

Introduction

The objective of this project was to leverage machine learning techniques to assess the attentiveness of students in a virtual reality classroom environment. Using sensory data from Meta Quest 2, we conducted experiments and developed a model to predict attentiveness levels accurately.

Dataset

We collected data by performing a Continuous Performance Test (CPT) within the VR classroom environment. The dataset consists of various sensory inputs recorded during the tests.

Technologies and Libraries Used

The project utilizes the following technologies and libraries:

• Programming Languages: Python
• Data Processing and Analysis:
  • pandas
  • numpy
  • scipy
• Machine Learning and Deep Learning:
  • scikit-learn
  • TensorFlow
  • Keras
  • statsmodels
• Visualization:
  • matplotlib
  • seaborn

Methodologies and Techniques Applied

The project followed these key methodologies and techniques:

1. Data Collection:

   • Conducted Continuous Performance Tests (CPT) within the VR classroom.
   • Recorded sensory data from Meta Quest 2.

2. Data Pre-processing:

   • Cleaned and pre-processed the sensory data.
   • Extracted relevant features from the raw sensory data.

3. Model Development:

   • Implemented various machine learning models, including Support Vector Machines (SVM) and Neural Networks.
   • Built a Sequential model using Keras with layers including Dense and Dropout.

4. Model Evaluation:

   • Evaluated models using metrics such as accuracy, F1 score, and mean absolute percentage error (MAPE).
   • Achieved a final MAPE of 21.92%.

Installation

To run this project, you will need to install the following dependencies:

pip install pandas numpy scipy seaborn matplotlib scikit-learn tensorflow keras statsmodels

Usage

1. Clone the repository:

git clone https://github.com/yourusername/metaverse-classroom-attentiveness.git
cd metaverse-classroom-attentiveness

2. Run the Jupyter notebooks to process data and train models:

jupyter notebook

3. Open and execute the notebooks in the following order:
   • parse.ipynb
   • analysis.ipynb

Contributing

Contributions are welcome! Please fork the repository and submit a pull request with your changes.