A Neural Network for Generating Adaptive Lessons

Abstract

Traditional sequencing technology developed in the field of intelligent tutoring systems
 have not find an immediate place in large-scale Web-based education. This study investigates the use
 of computational intelligence for adaptive lesson generation in a distance learning environment over
 the Web. An approach for adaptive pedagogical hypermedia document generation is proposed and
 implemented in a prototype called KnowledgeClass. This approach is based on a specialized artificial
 neural network model. The system allows automatic generation of individualised courses according to
 the learner’s goal and previous knowledge and can dynamically adapt the course according to the
 learner’s success in acquiring knowledge. Several experiments showed the effectiveness of the
 proposed method

Artificial Intelligence and Machine Learning Approaches in Digital Education: A Systematic Revision

“…This theme consists of studies that use different algorithms for adaptive and predictive learning as well as for addressing different learning styles needed for digital education. For this theme (see Table 5), ten papers have been classified: six experiment-based studies [64][65][66][67][68][69], three case studies [70][71][72], and one survey-based study [73]. The ANN model is studied to generate adaptive lessons for an individual [64].…”

“…For this theme (see Table 5), ten papers have been classified: six experiment-based studies [64][65][66][67][68][69], three case studies [70][71][72], and one survey-based study [73]. The ANN model is studied to generate adaptive lessons for an individual [64]. The model generates a set of documents that is adapted to learners' needs by searching for the best route to connect the known concepts.…”

“…The model generates a set of documents that is adapted to learners' needs by searching for the best route to connect the known concepts. The learner self-defines the learning goals, where selection algorithms aim to present the most suitable didactic plan based on the goal and considering the learner's actual knowledge [64]. Learner modeling and resource modeling are important aspects to deploy adaptive mechanisms [73].…”

Concept-Based Classification for Adaptive Course Sequencing Using Artificial Neural Network

“…The content selection rules are designed according to the cognitive style or learning preferences of the learner [2]. Though most of these rules are domain independent, there are still an absence of well defined and commonly accepted rules on the selection and sequencing of the content in a way to produce a sense of instructional value in most educational system ( [3], [5], [6]). Furthermore, the complexity of dependencies between the content or learning object characteristics and learners usually requires a complicated and huge set of rules to be defined.…”