Agile Methodology

There are several phases involved in this methodology such as information gathering, analysis, design, development, testing, and feedback. Agile methodology then be modified into information gathering, analysis, design, development, testing, and feedback to meet the activities carried out throughout developing the system.

Collect and build student’s profile and gather information

Based on the student’s profile, design the algorithm system that describes the procedures of the assessment. 

Encourage students to reflect on their learning process and provide feedback that guides them to higher levels of understanding and application.

Predictive Analysis System

We apply Machine Learning algorithms such as decision trees and linear regression to predict student performance, using data from academic records and assessments. This approach enables the identification of patterns and trends, informing personalized educational strategies through continuous model refinement with new insights.

Profile Analysis

Evaluate a student's skills and academic performance by considering their coursework, projects, extracurricular activities, major, and school enrollment details, along with their career goals and aspirations.


Customize assessment tests for students based on their profiles, focusing on five key areas: cultural adaptability, language communication, cognitive ability, technical skills relevant to their major or career, and overall career readiness.

Data Analysis

Utilize institutional and industry data to conduct predictive analysis, enabling the evaluation and monitoring of each student's performance and trends.

Custom Assestment

Develop a personalized assessment system using an algorithm that tailors tests to each student's profile and institutional data, incorporating adaptive testing and providing detailed feedback and reports for self-improvement.

Predictive Analysis

Integrate diverse data sources and apply machine learning algorithms for predictive analysis of student performance and career success, with a focus on continuous system improvement through regular updates and feedback.


Conduct pilot tests of the system with a select group of students for initial feedback and adjustments, followed by a full-scale implementation across the wider student body, accompanied by continuous support and updates.


Continuously monitor and evaluate the system's effectiveness in predicting student performance and career outcomes, while establishing a feedback loop with students and educators for ongoing refinement.

Students' Grades Prediction using Machine Learning

We use the decision tree approach and the linear regression model to predict students' academic achievement using machine learning. The algorithm has been assessed based on the accuracy score. This will help us understand the key factors of success and failure that influence student performance, assisting them in achieving high grades for their post-secondary studies.

Copyright © 2024 Bluekey AI Technology. All rights reserved. Content on this site, including text and images, may not be used for any commercial purpose without our explicit permission. Some content may be sourced from partners. If you have concerns regarding any content or images that may originate from another source, please contact us at