Moodle Analytic Admin Tool Plugin for Student Performance Predict
Download as PPTX, PDF0 likes20 views
The document outlines a research project focused on developing a Moodle analytics plugin to identify students at risk of poor performance in e-learning environments. It details the project's aims, objectives, and methodology, including the use of machine learning to predict student performance and provide actionable insights. The initiative seeks to enhance Moodle's learning management system by offering a user-friendly analytics solution for administrators.
Moodle Analytic Admin Tool Plugin for Student Performance Predict
- 1. MOODLE ANALYTIC PLUGIN TO IDENTIFY STUDENTS PERFORMANCE 2nd Progress Presentation EEY7881 Engineering Research Project (Computer Engineering) Supervisor, Dr. D.D.M. Ranasinghe. YML Kumara | 50864636 Dept. Electrical & Computer Engineering
- 2. Content • Problem Background • Overview of Project • Aim • Objectives • Methodology – Block Diagram – Indicator for Model – Analyzer – Machine Learning Model • Features of the system • Time Plan
- 3. Problem Background 3 • Performance of student is risk at E-learning. • Moodle is widely used for E-learning. • No mechanism to identify low performance student in Moodle environment and make proper intervention to fetch students up to course complete level.
- 4. Overview of Project • Early detection of student at low performance • Clustering students with feature matrix. • Specific algorithm for every clusters. • Machine learning backend prediction model. – Both Python and PHP • Extra user actions based on prediction result. • Proper intervention where necessary. 4
- 5. Aim • Identification of existing analytic solution in the Moodle. • Developed a mechanism to Identify student at low performance. • Developed a mechanism to engage students and teachers with LMS effectively. • Developed a dashboard widget for quick access. • Evaluate the developed system for accuracy. • To enhance the Moodle Learning Management System (LMS) by providing administrators with a powerful and user-friendly analytics solution. Objectives
- 6. Learning analytics • Descriptive -what happened? • Predictive -what will happen next? • Diagnostic -why did it happen? • Prescriptive -What should be done? Appelbaum et al., 2017; Davenport & Harris, 2017; Delen & Demirkan, 2013; Banerjee et al., 2013
- 8. Potential Cognitive Depth # Criteria Depth i Learner has not even viewed the activity 0 ii The learner has viewed the activity details. 1 iii The learner has submitted content to the activity. 2 iv The learner has viewed feedback from an instructor 3 v The learner has provided feedback to the instructor 4 vi The learner has revised and/or resubmitted content to the activity. 5
- 9. Potential Social Breadth # Criteria Depth i The learner has not interacted with anyone 0 ii The learner read the page but has not interacted with any other participant in this activity 1 iii The learner has interacted with at least one other participant such as they have submitted an assignment or attempted a self-grading quiz providing feedback. 2 iv The learner has interacted with multiple participants in this activity such as posting to a discussion forum or extra assistance done. 3 v The learner has interacted with participants in at least one of communications back and forth. 4 vi The learner has interacted with people outside the class like in an authentic community of practice. 5
- 10. General Indicator for the Model Attribute Type Description First visit date. Numeric No of days passed after start date Before start access. Nominal Course accessed before start. Write action in site. Numeric If the user has completed a saved content anywhere on the site. Write action in enrolled course. Numeric If the user has completed a saved content in the enrolled course. Read actions. Numeric Estimates the amount of content the user has accessed. Answer submitted. Nominal If the user has activities due and not yet submitted. View announcements. Numeric No of time check for new announcements Discussion group. Numeric No of time participate Absence days. Numeric Number of absence days
- 11. Analyzer • Pre-process – – cleaning the data – Cognitive depth and social breadth simulation – transforming into form that can be feed to the algorithm. • Distinguish training set and a testing set. • Parameter tuning using grid search. • Action Generation for prediction.
- 12. Machine Learning Model • K-Means - Clustering students • Support Vector Machines • Decision trees • Naïve Baye • Grid Search for parameter tuning
- 13. Open University Learning Analytics Dataset • Original owners :- The Open University, Walton Hall, Milton Keynes, United Kingdom. – courses.csv – assessments.csv – vle.csv (Virtual Learning Environment data) – studentInfo.csv – studentRegistration.csv – studentAssessment.csv – studentVle.csv • Missing Attribute Values: Yes
- 14. Dataset Summary Description Amount Students enrolled 32953 Number of courses 22 VLE pages 6364 VLE log entries 10655280 Registration entries 32953 Assessments 206 Assessment entries 173912 Number of attributes 43 Distinction 3024 Fail 7052 Pass 12361 withdrawn 10156
- 15. Features of the system Develop the dashboard widgets, enabling quickly access the analyzed data. Notification Alert, actions and grading of alert Teacher task simulation Email for user define alert Early detection students at risk
- 16. Time Plan ID Action TIMELINE 2023 - 2024 JUNE JULY AUG SEP OCT NOV DEC JAN FEB 1 Proposal. Completed 2 Literature survey. 3 Project Planning and Analysis. 4 Analytics Framework Development. 5 Testing and Quality Assurance 6 Deployment and Documentation.
- 17. Thank You