Utilizing Learners' Negative Ratings in Semantic Content-based Recommender System for e-Learning Forum

Nowadays, most of e-learning systems embody online discussion forums as a medium for collaborative learning that supports knowledge sharing and information exchanging between learners. The exponential growth of the available shared information in e-learning online discussion forums has caused a difficulty for learners in discovering interesting information. This paper introduces a novel recommendation architecture that is able to recommend interesting post messages to the learners in an e-learning online discussion forum based on a semantic content-based filtering and learners' negative ratings. We evaluated the proposed e-learning recommender system against exiting e-learning recommender systems that use similar filtering techniques in terms of recommendation accuracy and learners' performance. The obtained experimental results show that the proposed e-learning recommender system outperforms other similar e-learning recommender systems that use non-semantic content-based filtering technique (CB), non-semantic content-based filtering technique with learners' negative ratings (CB-NR), semantic content-based filtering technique (SCB), with respect to system accuracy of about 57%, 28%, and 25%, respectively. Furthermore, the obtained results also show that the learning performance has been increased by at least 9.84% for the learners whom are supported by recommendations based on the proposed technique as compared to other similar recommendation techniques.