Novel online Recommendation algorithm for Massive Open Online Courses (NoR-MOOCs)

Area of Science:

• Educational Technology
• Computer Science
• Artificial Intelligence

Background:

• Massive Open Online Courses (MOOCs) are increasingly popular, leading to information overload.
• Existing recommender systems struggle with scalability, data sparsity, and dynamic updates in MOOC environments.
• Traditional recommendation techniques are ill-suited for the rapidly growing and changing nature of MOOC data.

Purpose of the Study:

• To propose a novel online recommender system, NoR-MOOCs, for Massive Open Online Courses.
• To develop a system that is accurate, scalable, and handles incremental data updates.
• To overcome the limitations of traditional recommendation algorithms in the MOOC context.

Main Methods:

• Development of a novel online recommender system named NoR-MOOCs.
• Extensive experimentation using the COCO dataset.
• Empirical comparison against traditional KMeans and Collaborative Filtering algorithms.

Main Results:

• NoR-MOOCs demonstrates superior performance compared to KMeans and Collaborative Filtering.
• The system achieves significant improvements in predictive accuracy metrics.
• The system shows significant improvements in classification accuracy metrics.

Conclusions:

• NoR-MOOCs effectively addresses the challenges of recommending learning resources in MOOCs.
• The proposed system offers a scalable and accurate solution for dynamic online learning environments.
• NoR-MOOCs provides a viable alternative to traditional recommender systems for MOOC platforms.