Personalized and adaptive learning

A learning style classification mechanism for e-learning

Learning style detection is essential for developing adaptive learning tailored to learner preferences. Extensive research has been conducted to address the limitations of direct learner engagement, behavioral-based approaches, and the lack of learner-teacher interaction caused by the shift to online learning. Many learning style models have been widely used and evaluated for their performance in supporting adaptive learning. However, implementing Multiple Intelligence Learning Styles (MILS) remains limited in online learning due to low detection performance. This study employs machine learning approaches to enhance the detection performance of Multiple Intelligence Learning Styles. This study proposes an integrated machine learning framework to enhance MILS classification by combining data preprocessing, log-modulus transformation, and class imbalance handling via Synthetic Minority Over-Sampling Technique (SMOTE), Adaptive Synthetic Sampling Approach (ADASYN), and SMOTE-Tomek. Recursive Feature Elimination (RFE) with Support Vector Machine (SVM) is used for feature selection while eight machine learning models are used for classification. Interactions between learners and the Learning Management System (LMS) comprise the dataset, which are labeled using the Multiple Intelligence Inventory. The SVM model using ADASYN outperforms previous studies and other models, achieving the highest F1-Score of 89%, according to experimental data using 5-fold cross-validation. Statistical tests (Shapiro-Wilk, Analysis of variance (ANOVA), and paired t-tests) confirm significant differences in performance between models. The proposed approach demonstrates enhanced detection performance and can be adapted to broader e-learning applications, supporting adaptive and personalized learning systems.

Learning style is an important factor that accounts for the individual student learning in any learning environment. Every student has a different learning style and different ways to percept, process, retain and understand new information. In this paper, a new approach is proposed to classify students learning style automatically and dynamically depending on their learning behavior in a learning management system (LMS). There have been several approaches proposed for automatic learning style detection. One of the widely accepted and frequently used classification techniques is the decision tree classifier. The decision tree classifier mainly depends on the construction of strong decision rules which are required to identify learning styles accurately. The lack of strong decision rules would lead to the misclassification of individual students learning style. Hence, this paper mainly focuses on the construction of strong decision rules to strengthen the existing decision tree classifier to accurately and precisely classify students learning style thereby improving the classification accuracy. The proposed approach has experimented with an average of 300 students enrolled for the online courses in Moodle LMS. Initially, the students' behavior are extracted from the web log files of LMS and then preprocessed to build decision tree classifier using strong decision rules based on the three learning dimensions of standard Felder Silverman learning style model (FSLSM). The evaluation result is obtained using the inference engine with forward reasoning searches of the rules until the correct learning style is determined. Based on the result obtained, the prediction of learning style is done for the new students automatically and accurately using the significant rules built in the decision tree classifier. The experimental result proved that the processing dimension shows variance in classification whereas perception and input dimension shows less variance with an average accuracy of 87%.

Although the results of numerous studies have indicated that providing for the learning styles of students results in marked improvement in behavior and achievement, the results of this study indicate that the relationship between middle level students, their learning styles, their academic success and self–reporting learning style instruments is confusing and tenuous at best. Neither statistical differences nor linear relationships between achievement and learning styles were confirmed and interview data findings indicate students fail to recognize their learning style preferences. It is possible that the traditional classroom does not provide sufficient environmental options to enable students to recognize their learning style preferences. It is also possible that the Dunn and Dunn Learning Styles Inventory may not accurately profile the learning style preferences of middle level students; in fact, the LSI may profile only students' perceptions of appropriate learning behavior.

This study aims to examine the effect of using Canva-based learning media on learning outcomes of Islamic Religious Education (PAI) in class III SDIT Al-Insan Pinrang, as well as how it relates to individual learning styles of students. The problem raised is the extent to which the use of Canva media can improve students' learning outcomes compared to PowerPoint media, taking into account visual, auditory, and kinesthetic learning styles. This study used a quantitative approach with a quasi-experimental non-equivalent control group design, where learners were divided into two groups, namely the experimental group using Canva and the control group using PowerPoint. The research design also involved a 2x3 factorial analysis to see the interaction between learning media and learning styles. Data was collected through pre-test and post-test to measure learning outcomes, as well as a questionnaire to identify learners' learning styles. Hypothesis testing was conducted using two independent samples t-test, one-way ANOVA, and Kruskal-Wallis to see the differences between experimental and control groups, as well as differences based on learning styles. The results showed that the use of Canva had a significant effect on PAI learning outcomes, especially for students with visual and kinesthetic learning styles. However, there was no significant difference between the auditory learning style groups using Canva and PowerPoint. This finding reinforces the importance of considering learning styles in choosing learning media, and shows that visual design-based learning media, such as Canva, can increase learners' motivation and active participation, especially in learning that requires greater visual engagement. This research is beneficial for the development of more adaptive and effective learning methods in primary schools.

The ability to detect learners’ learning styles based on their learning behaviors is of utmost importance for online educational systems, as it greatly enhances student engagement, motivation, and overall learning outcomes. Knowing the learning preferences of learners may significantly aid in creating personalized learning recommendations and empower learners to identify their own learning styles. However, learners exhibit diverse behaviors in an online setting, which poses significant difficulties in detecting their learning style. This paper proposes a novel approach for detecting learning styles using graph representation learning techniques and machine learning algorithms. While our approach is not reliant on a particular learning style model, our approach may be divided into two distinct parts. Initially, we represented the behavior of learners as a bipartite graph and then transformed this graph into a lower-dimensional representation using the graph embedding approach. This lower-dimensional representation was then utilized for machine learning tasks. Furthermore, we categorize the encoded learner’s sequence using a clustering technique based on the selected learning style model. Our methodology uses the Felder-Silverman model as the learning style model and the Fuzzy C-means algorithm as a clustering technique. Our approach was evaluated using the 2015 KDD Cup dataset through a series of comprehensive experiments to showcase its effectiveness. The findings demonstrate that our approach surpasses the previous approach, achieving an average precision of 0.8737 and an accuracy of 0.9182.

Learning styles are increasingly being integrated into computational-enhanced earning environments and a great deal of recent research work is taking place in this area. The purpose of this study was to examine the impact of the computational experiment approach, learning styles, epistemic beliefs, and engagement with the inquiry process on the learning performance of pre-service engineering students. The study used the Felder-Silverman learning style model (FSLSM), in order to provide information for the relation of FSLSM with the learning environment in order to examine whether the strength of learning styles has an effect on the students' learning performance in mismatched courses. Our objective was: a) to investigate whether students with a strong preference for a specific learning style have more difficulties in learning, if their learning style is not supported in the learning environment; b) if the methodology of the computational experiment has an impact on students independently of their learning style; and c) if the epistemological beliefs are related to different learning styles. The learning environment was based on the methodology of the computational experiment and applications were developed using the Easy Java Simulator software, while the inquiry based teaching and learning process was adopted. The questionnaire responses were gathered from 79 pre-service engineering students in a higher education institute in Greece. Results indicate that students with no preferred learning style have a better learning performance in mismatched courses.

Learning styles are determined by how individuals approach learning, understanding, and remembering information. The researcher employed descriptive type research. The respondents of the study were composed of one hundred twenty-nine (129) Grade V and VI learners of Manolo Fortich District IV namely, Impakibel Elementary School and Santiago Integrated School. The researcher selected the respondents through universal sampling method. The researcher adopted the learning style questionnaire of Ramos and Lopez (2020). Data were evaluated employing mean and standard deviation. Pearson Product Moment Correlation was used to distinguish the significant relationship between the learning styles and academic performance of IP learners. The study revealed that Visual Learning Styles was highly preferred learning style by IP learners. Kinesthetic Learning Style was undecided by IP learners. Furthermore, there is no significant relationship between auditory and visual learning style and academic performance. However, kinesthetic learning style and academic performance of IP learners have a significant relationship. Thus, teachers may design classes and activities that suit to the learning style preference of their learners. Therefore, action research is recommended

Various types of learners can be observed in today’s e-learning environments. Learning Analytics can offer insights on a student’s actions and behaviour. We can correlate this data with the identification of learning styles like in the Felder-Silverman Learning Styles Model (FSLSM), and use it for user modelling. Supported by visualizations on learning styles and learning behaviour students and instructors can reflect the learning process. Based on our previously published conceptual model for linking learning analytics and learning styles in e-learning environments, we present an Integrated Learning Styles Analytics system (ILSA), which supports identification of learning styles as well as analysis and visualization of activity data in the Moodle LMS. ILSA consists of a questionnaire for learning style identification derived from FSLSM and offers immediate results to its participants. Next, by utilizing various data sources in Moodle, e.g. log data or the grade book, we are able to correlate a user’s activity with their learning styles. This work offers details on the finalized concept as well as its implementation. By providing insights on data sources in Moodle and presenting various visualizations, this work allows teachers to reuse the system in their e-learning courses.

Due to the widespread of online learning, learning management systems (LMSs) contain many of online courses but very little attention is paid to how well these courses actually support learners. Teachers build courses according to their preferred teaching methods; on the other hand, learners have different learning styles. The harmony between the learning styles that a course supports and the actual learning styles of students can help to magnify the efficiency of the learning process. In this chapter, an interactive tool is presented for analyzing existing course contents in learning management systems based on learning styles. This tool allows teachers to be aware of the course support level for different learning styles. It visualizes the suitability of a course for students’ learning styles and helps teachers to improve the course support level of their courses. It aims at supporting teachers in adaptive and personalized learning environments to decide-making efficient modifications in the course structure in order to meet the need of different students’ learning styles.KeywordsInteractive course analyzerLearning stylesLearning management systems

The Society 5.0 era demands inclusive and effective education, yet diverse student needs remain unmet due to limited personalised approaches. The visual, auditory, and kinesthetic (VAK) model offers a promising solution to improve engagement and achievement, especially in biology learning at MAN 1 Kerinci, Jambi, Indonesia. This study aimed to examine the effect of VAK (Visual, Auditory, and Kinesthetic) learning styles on students’ learning achievement in biology subjects in Madrasah Aliyah Negeri 1 Kerinci. Using a quantitative approach with a correlational design, 60 students in class VIII were selected as samples through a purposive sampling technique. Research instruments in the form of learning style questionnaires and student learning achievement data were analysed using descriptive statistics, one-way ANOVA, and Tukey’s HSD post hoc tests. The results showed that learning styles had a significant influence on student learning achievement (p < 0.05), with the auditory learning style producing the highest average achievement compared to the Visual and Kinesthetic learning styles. These findings support the importance of customising learning strategies based on students' learning styles to improve learning outcomes. Recommendations from this study emphasise the need for varied, adaptive, and inclusive learning approaches to optimise students' academic potential.

Currently, e-learning is becoming an option as it can save the cost of education, time, and more flexible in its implementation. The main problem that arises is how to create e-learning content that is interesting and really fit the needs of the users. One way that can be done to optimize the content of e-learning is to analyze the user behavior. This study aims to analyze user (student) behavior in KALAM UMP, based on logs report (activity history), which is often called as behavioral tracking. First, the learning style of the students is determined based on Honey and Mumford Learning Styles Model by using Learning Styles Questionnaire. The analysis is done using SPSS 16.0 for Windows. The results shows that student with Reflector and Theorist learning styles access e-learning materials the most. From Spearman Correlation analysis, the relationship between learning styles and students’ behavior in e-learning is found to be very weak (r<sub>s</sub>=.276, p=.000), but statistically significant (p<0.05). In other words, students’ learning styles and behavior in e-learning have significant impacts on the improvement or degradation of students’ performance. Therefore, from the results of this study, an adaptive KALAM e-learning system which can suits the learning styles of UMP students is proposed. In adaptive e-learning system, students can access learning materials that match the students' learning needs and preferences.

The present investigation sought to explore the relationship between learning styles and writing behaviors of EFL learners in a blended environment. It also aimed to identify the learning style types best predicting writing behaviors. Initially, the participants' preferred learning styles were identified through the Kolb’s learning style inventory (Kolb, 1984). Secondly, data were obtained through analyzing the Stat counter and Input log data to reveal the pausing, revising and switching behaviors of the participants who attended a writing course in which they developed their writing texts using an online module. The results indicated a negative and significant correlation between the accommodator learning style and the revision behavior. A statistically significant and positive relationship was also found between the converger learning style and the pausing behavior, and between the converger learning style and the revision behavior Furthermore, a positive and significant relationship between the accommodator learning style and the switching behavior was revealed. The accommodator learning style was found as the best predictor for the switching behavior and the converger learning style turned to predict the revision and pausing behavior at an optimal level. The findings suggest that internal factors, cognitive and learning styles, play a significant role in the learning behaviors of English writing learners. The results encourage writing educators to take into account students’ learning style and provide more flexible and rigorous learning environment in which all learners can take benefit.

This study investigates the long-term impact of adaptive digital learning ecosystems on students' self-regulated learning (SRL) behaviors and academic persistence. Adaptive learning systems personalize the learning experience by adjusting content and feedback to meet individual students' needs, preferences, and performance. These systems enhance engagement, motivation, and learning outcomes through real-time adjustments and continuous feedback. The research aims to explore how adaptive learning systems influence SRL and academic persistence in university courses over time. Using a longitudinal quantitative design, the study tracks SRL behaviors and academic persistence at multiple points during the semester. Results show significant improvements in SRL behaviors such as goal setting, planning, self-monitoring, and reflection among students engaged with adaptive learning environments. These students exhibited greater autonomy, improved metacognitive awareness, and higher motivation. Additionally, students in adaptive systems demonstrated greater academic persistence, as indicated by more time spent on tasks, higher assignment completion rates, and sustained engagement. The findings suggest that adaptive learning platforms promote SRL and academic persistence by offering personalized, responsive learning experiences. Unlike static, non-adaptive environments, adaptive systems provide dynamic support, enhancing students' ability to regulate their learning and remain engaged despite challenges. The study concludes that adaptive learning systems are vital for long-term academic success, though further research is needed to assess the sustainability of these effects in various educational settings and among diverse student populations.

This study aims to examine and formulate an adaptive learning framework in Islamic-based character education integrated with child-friendly school leadership. The primary focus lies on how Islamic values—including aqidah, sharia, and akhlakul karimah—are internalized through a learning approach that is responsive to the individual needs of students, and managed in a leadership climate that supports child protection and participation. This study employs a qualitative approach, utilizing a case study method at the madrasah. Data were collected through in-depth interviews, participant observation, and documentation studies. The results of the study show four main findings: (1) the foundation of Islamic values ​​as the moral framework of the institution; (2) an adaptive learning approach that adapts to students’ learning styles and backgrounds; (3) child-friendly leadership practices that foster a sense of security and active involvement of students; and (4) a holistic evaluation that assesses academic and character development in an integrative manner. This study recommends the development of an Islamic character-based adaptive learning model, supported by a participatory and inclusive leadership culture, within the school environment.