Source code, cron job scripts, and database schema for the predictive chatbot for student success
Description
This dataset contains the full source code, cron job scripts, and database schema used in the study “Early detection and personalized academic support using a predictive chatbot for student success” (submitted to PeerJ Computer Science). The system integrates a high-performance LightGBM predictive model with an AI-powered WhatsApp® chatbot based on the GPT-4o-mini model to identify at-risk students and deliver early, personalized academic support.
Files
Steps to reproduce
1. Clone or download the repository containing the chatbot backend code and cron job scripts. 2. Install the required dependencies using npm install as indicated in the README file. 3. Set up a PostgreSQL database and import the db_schema.sql file to create the required tables. 4. Configure the environment variables (database connection parameters, OpenAI API key, and WhatsApp® session credentials) following the instructions in the README. 5. Run the predictive model and cron job scripts to populate the table of at-risk students using institutional grade data. 6. Start the chatbot backend to enable WhatsApp® conversations with at-risk students. 7. Use the provided example configuration files as a template to adapt the system to a new institutional context while preserving student privacy.
Institutions
- Universidad Estatal de Milagro
- Universitat Politecnica de Catalunya