Palomar project: predicting school renouncing dropouts, using the artificial neural networks as a support for...

Area of Science:

• Educational Psychology
• Artificial Intelligence in Education
• School Guidance Systems

Background:

• The Italian school system faces challenges with student disengagement, including dropout, grade repetition, and study delays.
• Effective educational guidance is crucial for addressing these complex issues and supporting student success.

Purpose of the Study:

• To develop and implement the "Palomar" project, a system for predicting and optimizing interventions for students at risk of school failure or dropout.
• To provide educators and administrators with tools to enhance school guidance services and improve student outcomes.

Main Methods:

• Utilizing Artificial Neural Networks (ANN) for predictive modeling to identify students at risk of school destabilization.
• Developing a system to analyze student characteristics, predict potential success in different study paths, and identify key factors for intervention.
• Implementing a "follow-up" verification process to monitor intervention efficacy and allow for adjustments.

Main Results:

• The Palomar project's prediction system can identify students at risk of school failure with high approximation.
• It optimizes interventions by pinpointing personal factors needing reinforcement and simulating support measures.
• The system can be applied to individual students or groups, optimizing class and school-wide formative routes.

Conclusions:

• The Palomar project offers a robust framework for proactive educational guidance, leveraging AI to mitigate school dropout and failure.
• The system empowers educators to make data-driven decisions, personalize student support, and enhance overall educational efficacy.
• Continuous verification and adaptation of interventions ensure sustained improvement in student academic trajectories.