Download presentation
Presentation is loading. Please wait.
Published byErnest Watkins Modified over 6 years ago
1
University of Colombo School of Computing, Colombo, Sri Lanka
A Personalized Web Content Recommendation System for e-Learners in e-Learning Environment Daminda Herath University of Colombo School of Computing, Colombo, Sri Lanka
2
Introduction Nowadays e-Learning becomes one of the social aspects of the web and it plays a major role. Education institutions, universities and business organizations use e-Learning system to disseminate resources among e-Learners. Those parties struggle to give quality and useful required information to the e-Learners. On the other hand, e-Learners will be dissatisfied about given information and lose focus on the e-Learning.
3
Introduction According to the current e-Learning system, some problems are identified. Navigation patterns of e-Learners. E-learners’ level of knowledge. Association between subjects contents and questions. Relation between subject contents and assignment results by e-Learners. Tracking assignments completions and grades. Attention about posts in forums and frequently asked questions.
4
Introduction e-Learners weakness based on contents.
Unattended assignments by e-Learners. Analyzing the e-Learners feedback. Monitoring the progress of e learners. Predictions about enrollment, completions of course, accessing of contents and feedbacks. Recommending course contents.
5
Introduction In this research, a web content recommendation system is proposed to solve problems. A new system with a personalized and adaptive features will be introduce to e-Learners to promote e-Learning education. It is more dynamic and intelligent It uses to identify e-Leaner's level of knowledge and give good recommendations.
6
Methods To achieve this, web mining is used as a popular technique in Data mining. Web Mining Web Content Mining Web Usage Mining Techniques Clustering,Classification, Association, Sequential pattern analysis. Recommendation Collaborative and Content Filtering
7
A proposed Framework for A Personalized Web Content Recommendation System
8
Course →Concept → Learning Object
Course Content Hierarchy
9
Levels of Question Levels of Knowledge / Skill Level Proficiency Energy Point (EP) 1 Very Easier 10 2 Easy 15 3 Medium 20 4 Hard 25 5 Very Hard 30 Energy Point Range Proficiency Not Applicable 1-19 Beginner 20-39 Limited 40-59 Moderated 60-79 Competence 80-100 Good TYPES OF QUESTION Initial Skill Level Practical Level Final Skill Level Assignment Level All Questions are based on Learning Objects
10
Result
11
Concepts
12
Learning Objects
13
Review the answers
14
Skill Level and Recommendations
15
After Skill Level Test
16
Practical Level Test
17
Practical Level Attempts and eLearners’ Emotions
18
Learners’ Feedback
19
Initial Skill Level Test Summery
Introduction to Python Learning Object
20
Test Case Step Test case Expected result Actual Outcome Status 1
Select the course Display course Display course web content Pass 2 Try Initial skill level quiz Login into the system Enroll with the course Display login interface Display enroll button 3 Try practical quiz Check whether skill test finished Display a message (Try practical quiz/Finish skill test) 4 Try Final skill level quiz Check whether practical quiz finished (Try Final Skill level quiz/Finish practical quiz) Try assignment quiz Check whether Final skill level quiz finished (Try assignment quiz/Finish final skill level quiz)
21
Conclusion In this research, a personalized web content recommendation system proposed to encourage the e-Learners to pro-actively engage in e-learning environment to improve their education. This system was used web mining techniques such as web content and usage mining. Web content mining was used to identify relevant web contents and specifically web usage mining was used to identify e-Learners navigational patterns which could help to identify interests and weaknesses of e-Learners, frequently visited web contents, predicting performances of e-Learners. Then recommendation system could give an effective and efficient personalized web contents with aid of content and collaboration filtering.
22
Future Works In future work, it is going to implement for number of other courses in the e-learning environment and introduce biometric identification to identify the authorized e learners. Apart from that web mining and semantic web technologies can be incorporated to create an enhanced system with more intelligence. Also it is planning to implement this recommendation system with Massive Open Online Courses (MOOC). Machine learning (ML) and big data analytics techniques are used to make recommendation more accurately and enhance the performance in large heterogeneous environment.
23
Questions ?????????????? Thank You
Similar presentations
© 2025 SlidePlayer.com. Inc.
All rights reserved.