In Intelliboard, situations of students at risk of not completing a subject or course can be evaluated using 3 methodologies:

1. Rule-based models: each institution establishes with its best judgment what those rules should be, quantifying the limit values of different variables. It does not require historical data and is based on criteria of the type IF ->THEN. 

2. Predictive analytics based on "machine learning": uses an algorithm that, based on historical data from the institution's students, calculates through artificial intelligence the weight of each parameter used to predict whether a student is at risk or not. Therefore, it provides a higher degree of objectivity and also statistically quantifies the reliability of its predictions. This methodology is available as an add-on module to the standard Intelliboard Pro solution.

3. Risk Points: statistically calculates the student's risk using three main categories: Participation (visits, time, participations), Attendance (days since last visit, % days connected...), and Progress (current grade, % of activities completed...). It uses Z scores, revealing whether a student's performance is typical or atypical compared to the course average, allowing identification of students who may need additional support.