0% found this document useful (0 votes)
3 views10 pages

Project Defense

Uploaded by

info.pacatech
Copyright
© All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PDF, TXT or read online on Scribd
0% found this document useful (0 votes)
3 views10 pages

Project Defense

Uploaded by

info.pacatech
Copyright
© All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PDF, TXT or read online on Scribd

FEDERAL UNIVERSITY OF LAFIA

DEPARTMENT OF COMPUTER SCIENCE

A PREDICTIVE INTELLIGENCE
SYSTEM FOR ANALYZING AND
VISUALIZING STUDENT
ACADEMIC PERFORMANCE
PRESENTED BY:
2021/CP/CSC/0083
SHIDOIKU DANIEL SIMON
1ST APRIL, 2026
ABOUT THE TOPIC
a smart computer system that uses Artificial
Intelligence (machine learning) to predict how well
students will perform academically — and then
displays those predictions on an easy-to-read
dashboard so teachers and students can take action
early
STATEMENT OF THE
PROBLEM
1 conventional evaluation only identifies problems
after the students fall behind. too late for timely
help.

2 Educational data from LMs, onine courses and


attendace log is collected but rarely converted to
actionalbe insights.

3 Existing predictive model produce results that are


too complex for educators to understand or act
upon.

4 current dashboards serves administration, not


individual students. No tailored interventions exit.
AIM
To develop a predictive intelligence system that
analyze and visualizes student academic performce
to provide actionable insights for educators and
learners.
Objectives
Objective 1 Objective 2 Objective 3 Objective 4

Collect and integrate Design and implement Develop an interactive Evaluate the effectiveness
diverse student academic macine learning models visualization dashboard of the system in supporting
data (assessments, capable of accurately presenting predictions and timely interventions and
attendance, engagement forecasting students trends intuitively. improving academic
metrics). performance. outcomes.
EARLY DETECTION
1 Attendance logs, submission rates & interacton data are

Literature Review
realiable early indicators of at risk students. (Rozario &
GunaSundair, 2024)

Key Findings EXPLAINABLE AI


2 Educators need transparent, interpretable predictions, not
just accuracy. (Chen, 2025; Davalos et al,. 2025)

LEARNING DASHBOARD
3 Dashboards that combine visualization + prediction
improve educator decision making in real time (Pei et al.
2024)

PERSONALIZED LEARNING

4 Analytics driven personalization improves students


motivation and engagement. (Okoye et al. 2022)

ETHICAL AI

5 system must protect privacy, reduce bias, and build trust to


be widely adopted. (Alotaibi et al. 2026).
What Existing System Miss
Knowledge LOW INTERPRETABILITY
Model generates predictions but can’t

Gap
67%
explain them in a way educators can act on

LIMITED DATA INTEGRATION


50% Most systems only use exam scores,
ignoring attendace, assesmentand.

POOR SCALABILITY
50% System are institution specific and can’t
easily transfer to other education settins.

NO STUDENT-CENTERED DESIGN
70% Dashboards serve adiminstrators, not
students who need self-reflection tools.

ETHICAL AI
IMPROVED SYSTEM
COMPONENT EXISTIG MODEL IMPROVED MODEL

Exams, Attendance, assignments,


Data Source Exam, Scores, Attendance only
engagement

Algorithms Single ML Algorithm Hybrid ML

Visualization Static Dashboards Interactive, real-time dashboards

personlization Limited / none Tailoared interventions per student

Scalability Institution-specific Adoptable across insitutions


METHODOLOGY
This project uses the Agile Methodology, particularly
Scrum, due to its suitability for dynamic, user-feedback-
driven development. Agile allows for incremental
improvements based on evolving needs.
ank Yo
h
T

You might also like