Publication Information

Authors:

• Jaurès S. H. Kameni,
• Bernabé Batchakui,
• Roger Nkambou,

Pages:

• 65-95

Keywords:

• Artificial Intelligence, Digital Education and Educational Technology, Search Engine, Learning, Bloom’s Taxonomy, Machine Learning, Natural Language Processing

Abstract:

• The majority of Sub-Saharan African countries are facing a very negative teacher-learner ratio: one teacher for over 120 learners. In order to support the learner training, we propose optimizing search engines for learning contexts, to enable learners to take optimal advantage of the vast reservoir of Open Educational Resources (OER) available on the Internet. It bears noting that search engines serve as the primary means of accessing OER on the Internet. Indeed, search engines often present a large volume of content that is sometimes unsuited for learning contexts, potentially resulting in cognitive overload and a considerable loss of time for learners and teachers seeking appropriate resources among so many options. To this end, this work proposes an additional layer for Information Retrieval Systems (IRS) whose architecture features include: a match to the Bloom’s Taxonomy of the query, a pedagogical filter, and an informative presentation of results. The implementation of a system underpinning this model utilizes Google APIs. It should be noted that, search results are re-ranked after classifying the user query and initial results according to Bloom's taxonomy, performed through machine learning-based Natural Language Processing (NLP). In addition, an integrated intelligent filter removes irrelevant search results for learning purposes. On the Human-Machine Interaction (HMI) side, this layer relies on an educational content ontology to generate results that are more informative and semantically meaningful. An empirical evaluation of the system involving 84 students using a comparative questionnaire indicates the additional layer is favorable for educational use.