RIACT: A Responsible AI System for Personalized Study Habit Tracking and Early Burnout Signal Detection in University Students

When university life piles on lectures, assignments, and part‑time jobs, students often lose sight of how their study habits affect their well‑being. Burnout can creep in unnoticed, showing up only after grades slip or motivation vanishes. Researchers have built a tool that aims to make those hidden patterns visible before they become a problem.

What You Need to Know

RIACT stands for Record, Insight, Analyze, Coach, Track. It is a web‑based application that asks students to log each study session by noting the start time, end time, and physical location (e.g., library, dorm, café). From these logs the system calculates a metric called net focus time, which subtracts periods of distraction or multitasking detected through optional self‑reports or passive cues like keyboard idle periods.

The logged data feeds a hybrid AI architecture. A rule‑based component first cleans and aggregates the raw logs, flagging obvious inconsistencies such as overlapping sessions or unusually short breaks. A machine‑learning layer then looks for personal trends—variations in focus duration, timing shifts, and location changes—over weeks. When the model detects a deviation that matches known burnout precursors (e.g., a steady drop in net focus time combined with irregular sleep‑adjacent study blocks), it surfaces a personalized insight.

Beyond detection, RIACT offers lightweight coaching suggestions tailored to the student’s current pattern. If the system notices that late‑night library sessions correlate with lower focus, it might recommend shifting a portion of work to earlier hours or inserting a short break. All recommendations are presented with a brief explanation of the data that triggered them, supporting transparency and student agency.

Why It Matters

Burnout in higher education is not just a personal health issue; it correlates with lower academic performance, higher dropout rates, and long‑term mental‑health challenges. Existing productivity apps often record screen time or task lists but stop short of interpreting what those numbers mean for a student’s energy levels or stress. By turning raw logs into actionable feedback, RIACT bridges the gap between self‑monitoring and proactive self‑care.

Early detection matters because interventions are most effective before exhaustion reaches a crisis point. When students receive a timely signal—such as a notification that their average focus time has fallen 15 % over the past two weeks—they can adjust their schedule, seek support, or adopt healthier study habits before burnout compromises their coursework or well‑being.

Key Details

• Students log sessions via a simple web form: date, start time, end time, and location tag.
• Net focus time = logged duration minus estimated distraction periods (self‑reported or inferred from idle input).
• Hybrid AI: rule‑based preprocessing + a lightweight temporal‑pattern model (e.g., sliding‑window anomaly detection).
• Insight dashboard shows weekly focus trends, location‑based focus scores, and burnout‑risk indicators.
• Coaching module generates context‑specific tips (e.g., “Try a 10‑minute walk after 9 pm sessions”) with evidence citations from the student’s own data.
• Privacy design stores logs locally by default; optional cloud sync uses end‑to‑end encryption and allows data export or deletion at any time.

What’s Next

The research team plans to run a longitudinal trial with undergraduate cohorts to validate the burnout‑signal accuracy against established surveys like the Maslach Burnout Inventory‑Student Survey. They also aim to expand the location sensing to include Wi‑Fi SSID changes for passive logging, reduce entry friction, and explore how the coaching messages affect long‑term study habits and self‑reported stress levels. Outcomes from these steps will inform whether a lightweight, responsible AI tool like RIACT can become a routine part of student wellness services.

📌 Source: Arxiv Ai