Adhithpasu/PredictingStudentStressLevels

Student Stress Prediction Model

Model Summary

This model predicts stress levels in students based on behavioral, academic, and psychological indicators. It was developed as part of research into the implications of AI on student stress — supporting a forthcoming peer-reviewed publication on the topic.

The model is built with TensorFlow/Keras and is designed for binary or multi-class classification of student stress levels from structured survey/feature data.

---

Model Details

• Developed by: Chandrasekar Adhithya Pasumarthi (@Adhithpasu)
• Affiliation: Frisco ISD, TX | AI Club Leader | Class of 2027
• Model type: Deep Neural Network (TensorFlow/Keras)
• Language: English
• License: Apache 2.0
• Related research: "AI's Implications on Student Stress Levels" — forthcoming publication (expected April 2026)
• Related prior work: "Comparing Vision Transformers and Convolutional Neural Networks: A Systematic Analysis", JCSTS Vol. 8(2), January 2026

---

Intended Uses

Direct use:

• Predicting stress risk levels in students based on academic and behavioral features
• Supporting school counselors or ed-tech platforms in early identification of at-risk students
• Research into AI-driven mental health monitoring in educational settings

Out-of-scope use:

• Clinical diagnosis of mental health conditions
• Use without informed consent from students
• Deployment in high-stakes decision-making without human oversight

---

Training Data

The model was trained on a student stress dataset containing features such as:

• Academic performance indicators (GPA, study hours, deadlines)
• Behavioral signals (sleep patterns, extracurricular load)
• Self-reported psychological indicators (anxiety levels, social pressure)

Note: Dataset details and preprocessing steps are documented in the associated research paper.

---

Evaluation

Metric	Value
Accuracy	TBD
F1 Score	TBD
Precision	TBD
Recall	TBD

(Fill in with your actual results before publishing)

---

How to Use

import tensorflow as tf
import numpy as np

# Load the model
model = tf.keras.models.load_model("student_stress_model")

# Example input — replace with your actual feature vector
# Features: [study_hours, sleep_hours, gpa, extracurriculars, anxiety_score, ...]
sample_input = np.array([[6.0, 5.5, 3.8, 3, 7]])

# Predict
prediction = model.predict(sample_input)
stress_label = np.argmax(prediction, axis=1)

labels = {0: "Low Stress", 1: "Moderate Stress", 2: "High Stress"}
print(f"Predicted stress level: {labels[stress_label[0]]}")

---

Model Architecture

Input Layer  →  Dense(128, relu)  →  Dropout(0.3)
             →  Dense(64, relu)   →  Dropout(0.2)
             →  Dense(num_classes, softmax)

(Update this section to match your actual architecture)

---

Limitations & Bias

• The model is trained on a specific student population and may not generalize across different demographics, school systems, or cultures.
• Self-reported features are subject to response bias.
• Stress is a complex, multifactorial construct — model predictions should be treated as probabilistic indicators, not ground truth.

---

Ethical Considerations

Student mental health data is sensitive. This model is intended for research and educational tool development only. Any real-world deployment should:

• Obtain explicit informed consent from students (and guardians for minors)
• Be reviewed by a qualified mental health professional
• Comply with FERPA and applicable data privacy regulations

---

Citation

If you use this model in your research, please cite:

@misc{pasumarthi2026studentstress,
  author    = {Chandrasekar Adhithya Pasumarthi},
  title     = {Student Stress Prediction Model},
  year      = {2026},
  publisher = {Hugging Face},
  url       = {https://huggingface.co/Adhithpasu/student-stress-prediction}
}

---

Contact

• GitHub: @Adhithpasu
• Research: See linked publication for full methodology and dataset details.