Using an instructional expert to mediate the locus of control in adaptive e-learning systems

Instructional designers are under increasing pressure to enhance the pedagogical quality and technical richness of their learning content offerings, while the task of authoring for such complex educational frameworks is expensive and time consuming. Personalisation and reusability of learning contents are two main factors which can be used to enhance the pedagogical impact of e-learning experiences while also optimising resources, such as the overall cost and time of designing materials for different e-learning systems. However, personalisation services require continuous fine tuning for the different features that should be used, and e-learning systems need sufficient flexibility to offer these continuously required changes. The semantic modelling of adaptable learning components can highly influence the personalisation of the learning experience and enables the reusability, adaptability and maintainability of these components. Through the discrete modelling of these components, the flexibility and extensibility of e-learning systems will be improved as learning contents can be separated from the adaptation logic which results in the learning content being no longer specific to any given adaptation rule, or instructional plan. This thesis proposes an innovative semantic rule-based approach to dynamically generate personalised learning content utilising reusable pieces of learning content. It describes an ontology-based engine that composes, at runtime, adapted learning experiences according to learner’s interaction with the system and learner’s characteristics. Additionally, enriching ontologies with semantic rules increases the reasoning power and helps to represent adaptation decisions. This novel approach aims to improve flexibility, extensibility and reusability of systems, while offering a pedagogically effective and satisfactory learning experience for learners. This thesis offers the theoretical models, design and implementation of an adaptive e-learning system in accordance with this approach. It also describes the evaluation of developed personalised adaptive e-learning system (Rule-PAdel) from pedagogical and technical perspectives.

Adaptive e-learning is considered a stimulus to assist learning and increase student engagement. Hence, building suitable adaptive e-learning environments helps to individualize education to reinforce learning goals. This work aims to create an adaptable e-learning environment based on students' learning styles and examine the effect of the adaptive e-learning environment on student engagement. This study also contrasts the suggested adaptive e-learning environment with the standard e-learning strategy. The suggested adaptive e-learning strategy and findings may assist e-learning institutions in building and implementing more individualized and adaptable e-learning environments to increase student engagement. The work is based on mixed research methodologies used to examine the effect in the following manner: The adaptive e-learning environment is designed using the development technique, a quasi-experimental research methodology for performing the research experiment. The student engagement scale measures the following affective and behavioral factors: skills, participation/interaction, performance, and emotions. According to the data, the experimental group is statistically substantially higher than the control group. These experimental findings suggest that an adaptable e-learning environment can motivate pupils to learn. This study makes many practical recommendations: how to develop a foundation for adaptive e-learning based on learning styles and how to execute it; how to improve the impact of adaptive e-learning in education; and how to boost the cost-effectiveness of education.

Deployment of information processing theory to support adaptive e-learning systems: Feasibility study

rpose: This study aims to develop and evaluate an adaptive e-learning model tailored to the diverse learning styles of students, particularly within the context of distance learning at STIE Krakatau. The objective is to overcome the limitations of conventional e-learning systems that deliver uniform content to all users, regardless of their individual learning needs. Methodology: The study employed a development-based approach using an engineering model consisting of analysis, design, implementation, and evaluation phases. The adaptive e-learning system was created using multimedia authoring tools and tested in a seminar and training session lasting three hours. Results/Findings: The results demonstrated that the adaptive e-learning system effectively tailored instructional materials to the students' preferred learning styles (visual, auditory, and kinesthetic). The system successfully identified students’ learning preferences through a questionnaire and adjusted content presentation accordingly. Ongoing evaluations confirmed that the system functioned as intended and aligned with the designed functionalities. Conclusions: The implementation of an adaptive e-learning model significantly enhances the personalization and relevance of instructional materials, leading to better engagement and potentially improved learning outcomes in distance education. The seminar and training program proved effective in introducing and applying this adaptive model in a higher education setting. Limitations: The study was limited to formative evaluations and alpha testing. Further testing, including beta testing and broader implementation, is needed to determine user acceptance and the system's effectiveness in real classroom settings. Additionally, the current version relies on self-reported learning styles which may affect the accuracy of adaptation. Contributions: This research contributes to the advancement of personalized e-learning technologies by proposing a dual-level adaptation model incorporating both adaptivity (system-driven) and adaptability (user-driven).

Application of a fuzzy controller in adaptive e-learning content used to evaluate student activity

This chapter revolves around the synthesis of three research areas- data mining, personalization, recommendation systems and adaptive e-Learning systems. It also introduces a comprehensive list of parameters, extricated by reviewing the existing research intensity during the period of 2000 to October 2014, for understanding what should be essential parameters for adapting an e-learning. In general, we can consider and answer few questions to answer this body of literature ‘what' can be adapted? What can we adapt to? How do we adapt? This review tries to answer on ‘what' can be adapted. Thus, it advances earlier personalization studies. The gaps in the previous studies in building adaptive e-learning systems were also reviewed. It can help in designing new models for adaptation and formulating novel recommender system techniques. This will provide a foundation to industry experts and scientists for future research in adaptive e-learning.

Over the last years, sustainable learning has adopted various dimensions aimed at limiting negative impacts at the individual, social, and technological levels. For this reason, formal classrooms have been partially replaced by e-learning systems in order to ensure the sustainability of these systems. In addition, the MOOC (Massive Open Online Course) revolution is gaining an increasing popularity since everyone can improve their knowledge and skills by using these courses. Moreover, MOOCs are, especially interesting, as they handle not only a large amount of teaching data, but also a large number of users with diverse educational and cultural backgrounds. However, the retention rate of learners, which is generally around 10%, raises the question of the sustainability of this mode of teaching. In this paper, we suggest a sustainable e-learning system to tackle school dropout. The idea is to provide courses that correspond to the way in which learners could complete adequately their learning process. To this end, the model will be realized through an adaptive e-learning system by exploiting the traces left by users’ interactions with their learning environment. Depending on the learners’ profile, the system will automatically determine the path and recommend the appropriate courses for them using the ant colony algorithm.

Learner style data is increasingly being incorporated into adaptive eLearning (electronic learning) systems for the development of personalized user models. This practice currently relies heavily on the prior completion of questionnaires by system users. Whilst potentially improving learning outcomes, the completion of questionnaires can be time consuming for users. Recent research indicates that it is possible to detect a user's preference on the Global / Sequential dimension of the FSLSM (Felder-Silverman Learner Style Model) through a user's mouse movement pattern, and other biometric technology including eye tracking and accelerometer technology. In this paper we discuss the potential of eye tracking technology for inference of Visual / Verbal learners. The paper will discuss the results of a study conducted to detect individual user style data based on the Visual / Verbal dimension of the FSLSM.

Adaptive e-learning systems enhance the efficiency of online education by providing personalised, adaptive contents and user interfaces which change with respect to learner's requirements. In order to understand the learner's requirements, learners with similar learning behaviour have to be grouped into clusters based on the usage data of each learner. In this paper, a clustering technique to group learner's profiles is proposed where learners will be grouped based on similar sequences of accesses to learning material and time spent. A learner's model is designed based on Felder and Silverman learning style model. The clustering algorithm has two different phases, where the first phase considers the all sequences of access of learners which are in the chronological order of accessing the learning components and learning materials on the portal. The second phase considers the time spent on each learning components as a fuzzy membership function and groups the similar sequences of learners into three clusters. Learners in the clusters have similar learning behaviour for providing adaptive interfaces and contents.

These days modifications and supplements to e-learning are not enough to make them successful since flexibility and adaptability is missing. Different e- Learning architecture provide different kinds of learning experience to the learners. The lack of adaptive system is due to ―One size fits for all‖ concept. Currently the research is towards to learner oriented platforms putting student motivation, habits, expectations and learning styles (additional Koper, R.; Burgos, D (2005)). The concept of adaptation has become a prominent issue in e- learning. Systems allow the users to change some system parameters and adapt their behavior accordingly is known as adaption process. Systems that adapt based on the systems assumption are called as adaptive systems (additional santally mohammed issack). Since the system adapts to a person and personalizes the learning process it can be also known as personalized system (additional Paramythis, A.; Loidl-Reisinger). So to serve the learning community better there was a necessity to profile them and identify their requirements and provide an adaptive e- learning architecture which suits the individuals needs better. So the main objective of this research was to provide an adaptive e-learning system architecture which satisfies Felder Silverman learning style. Felder- Silverman's learning model (FSLM) was chosen since FSLM synthesis of these theories can be designed to easily translate them into strategies into a university setup. Index Terms—Learning, e-Learning, Virtual Learning, e- Learning Standards, Adaptive e-learning systems.

Online learners are increasingly diverse (NCES, 2022), which underlines the need for instructors to be inclusive and equitable in online teaching. Inclusion refers to providing opportunities for all learners in the online course, so they can actively participate and feel welcomed and belong in the course, and equity ensures that all learners have fair treatment and access to the opportunities and resources needed to succeed. In this survey-based research, we developed an Inclusive and Equitable Online Teaching Strategies (IEOTS) instrument with 45 strategies and examined instructor perceptions of the helpfulness of these strategies. These strategies focused on instructor self-awareness and commitment, getting to know the learners, course design, course facilitation, and evaluation. Based on the 478 online instructor survey responses, descriptive statistics showed that the instructors rated the strategies between somewhat helpful and helpful. In the open-ended question, student choice was described as an important aspect of the online course being inclusive and equitable. Analysis conducted based on the learner (student level), instructor (gender, ethnicity, teaching experience and teaching expertise), course (delivery modality), and organizational differences (required training, collaboration with instructional designer) found that instructor perceptions of helpfulness was higher for the course design subscale for instructors who taught online asynchronously rather than synchronously; higher for the know your learner subscale for instructors who taught graduate students rather than those who taught undergraduate students, and between those who attended training for online teaching compared to those who had not. In addition to supporting diverse online students, this study has implications for online instructors, instructional designers, and administrators who provide support to integrate these strategies effectively.

Adaptive E-learning systems, which provide personalized learning experiences based on learner's specific characteristics (e.g. knowledge, skills, and competencies), are essential for developing effective learning and teaching. This paper presents a conceptual proposition consisting in developing adaptive E-learning system by incorporating egocentric network-based user profiling, an existing contribution of our research team. Actually, this contribution has been proved in the online social network context with empirical results on different social networks data (Facebook, Delicious, Twitter and DBLP). We aim to apply the existing algorithms underlying this contribution in E-learning context and present in this work a conceptual model and the construction process of learner's user profile, called profile. The learner profile can be exploited in adaptive E-learning systems to provide different personalized/adapted services (books recommendation, personalized courses search, ...) to the learner.

Artificial intelligence tools like ChatGPT can impact an individual's cognitive processes, such as thinking, learning, and problem-solving. Recent studies have explored ChatGPT in education, but they mainly rely on educator and learner surveys for feedback and report qualitative data findings. There is also a lack of controlled experiments measuring the effect of integrating ChatGPT into Adaptive E-learning Systems (AESs) on learning. This paper aims to fill this research gap by evaluating AESs incorporating ChatGPT on learning outcomes and learner satisfaction. A framework for ChatGPT-enhanced AESs was proposed. An AES was developed based on this framework and served as the primary learning platform in a controlled experiment to teach Java programming. The experiment involved 54 male first-year undergraduate students majoring in Business Administration and primarily focused on evaluating learning outcomes and learner satisfaction. The results indicated that the AES with ChatGPT led to better outcomes and higher learning satisfaction than the AES without ChatGPT. This paper discusses these findings and suggests future research directions.

Personality and its effects on learning performance: Design guidelines for an adaptive e-learning system based on a user model

The domain of education has taken great leaps by capitalizing on technology and the utilization of modern devices. Nowadays, the established term "one size fits all" has begun to fade. The research focuses on personalized solutions to provide a specially designed environment on the needs and requirements of the learner. The adaptive platforms usually use Learning Styles to offer a more effective learning experience. This review analyzes the learner model, adaptation module, and domain module, originating from the study of 42 papers published from 2015 to 2020. As more modern techniques for adaptation get incorporated into e-learning systems, such techniques must be compliant with educational theories. This review aims to present the theoretical and technological background of Adaptive E-learning Systems while emphasizing the importance and efficiency of the utilization of Learning Styles in the adaptive learning process. This literature review is designated for the researchers in this field and the future creators and developers of adaptive platforms.