doi: 10.56294/dm2023146
ORIGINAL
Harnessing Artificial Intelligence for Personalized Learning: A Systematic Review
Aprovechamiento de la inteligencia artificial para el aprendizaje personalizado: Una revisión sistemática
Zainab Rasheed1 
*, Sameh Ghwanmeh2 *, Abedallah Zaid Abualkishik2 *
1Innovation and Research Department, Zennova Technology, Sharjah, U.A.E.
2College of Computer Information Technology, American University in the Emirates, Dubai, UAE.
Cite as: Rasheed Z, Ghwanmeh S, Abualkishik AZ. Harnessing Artificial Intelligence for Personalized Learning: A Systematic Review. Data and Metadata 2023;2:146. https://doi.org/10.56294/dm2023146.
Submitted: 01-10-2023 Revised: 25-11-2023 Accepted: 29-12-2023 Published: 30-12-2023
Editor: Prof.
Dr. Javier González Argote
ABSTRACT
Introduction: the document presents a comprehensive review of the utilization of Artificial Intelligence (AI) in personalized learning within the educational context. The study aims to investigate the various approaches to using ML algorithms for personalizing educational content, the impact and implications of these approaches on student performance, and the challenges and limitations associated with AI in personalized learning. The research questions are structured around these three broad areas, focusing on the AI methods used in education, their impact on students’ academic outcomes, and the challenges and limitations associated with AI.
Methods: the study employed a systematic literature review methodology, utilizing a structured and replicable search strategy to identify relevant research material from high-impact peer-reviewed journals published between 2015 and 2023. Inclusion and exclusion criteria were applied to select studies that focused on AI in education for personalized learning. Data collection involved extracting relevant data from the selected studies, and a thematic analysis was conducted to identify themes related to the research questions. The selected studies were graded based on their quality, and the results were summarized in a narrative synthesis.
Results: the analysis of the selected research papers revealed the significance of adaptive learning systems, recommender systems, NLP techniques, and intelligent tutoring systems in tailoring educational content to individual students. These approaches have demonstrated their effectiveness in enhancing student engagement, improving learning outcomes, and providing personalized feedback. However, the study also identified challenges and limitations that need to be addressed for the successful implementation of AI in personalized learning.
Conclusions: the study identified several limitations, including potential bias toward certain research areas, contextual factors influencing the effectiveness of ML algorithms, and the need for further research to examine the applicability of different approaches across diverse contexts. The findings highlight the research gaps, limitations, and potential future research areas in the field of AI-based personalized learning in education.
Keywords: ML; Education; Artificial Intelligence; Personalized Learning; Pedagogical Frameworks; Emerging Technologies.
RESUMEN
Introducción: el documento presenta una revisión exhaustiva de la utilización de la Inteligencia Artificial (IA) en el aprendizaje personalizado dentro del contexto educativo. El estudio tiene como objetivo investigar los diversos enfoques en el uso de algoritmos de ML para personalizar el contenido educativo, el impacto y las implicaciones de estos enfoques en el rendimiento de los estudiantes, y los retos y limitaciones asociados con la IA en el aprendizaje personalizado. Las preguntas de investigación se estructuran en torno a estas tres grandes áreas, centrándose en los métodos de IA utilizados en la educación, su impacto en los resultados académicos de los estudiantes, y los retos y limitaciones asociados a la IA.
Métodos: el estudio empleó una metodología de revisión sistemática de la literatura, utilizando una estrategia de búsqueda estructurada y replicable para identificar material de investigación relevante de revistas revisadas por pares de alto impacto publicadas entre 2015 y 2023. Se aplicaron criterios de inclusión y exclusión para seleccionar estudios centrados en la IA en la educación para el aprendizaje personalizado. La recopilación de datos implicó la extracción de datos relevantes de los estudios seleccionados, y se realizó un análisis temático para identificar temas relacionados con las preguntas de investigación. Los estudios seleccionados se calificaron en función de su calidad y los resultados se resumieron en una síntesis narrativa.
Resultados: el análisis de los trabajos de investigación seleccionados reveló la importancia de los sistemas de aprendizaje adaptativo, los sistemas de recomendación, las técnicas de PNL y los sistemas de tutoría inteligente para adaptar los contenidos educativos a cada alumno. Estos enfoques han demostrado su eficacia para aumentar el compromiso de los estudiantes, mejorar los resultados del aprendizaje y proporcionar información personalizada. Sin embargo, el estudio también identificó retos y limitaciones que deben abordarse para la implementación exitosa de la IA en el aprendizaje personalizado.
Conclusiones: el estudio identificó varias limitaciones, incluido el sesgo potencial hacia ciertas áreas de investigación, los factores contextuales que influyen en la efectividad de los algoritmos de ML y la necesidad de realizar más investigaciones para examinar la aplicabilidad de diferentes enfoques en diversos contextos. Los resultados ponen de relieve las lagunas, limitaciones y posibles áreas de investigación futuras en el campo del aprendizaje personalizado basado en IA en la educación.
Palabras clave: ML; Educación; Inteligencia Artificial; Aprendizaje Personalizado; Marcos Pedagógicos; Tecnologías Emergentes.
Introduction
Background
As artificial intelligence (AI) technologies have evolved, personalizing educational content can now be achieved based on students' preferences and engagement styles.(1) Machine learning (ML) algorithms have been used to analyze vast amounts of data on student behavior, such as their learning styles, performance patterns, and interests, to develop personalized recommendations for educational content.(2) This approach to personalized learning has shown promising results in improving students' academic outcomes, engagement, and motivation. In recent years, there has been increasing interest in investigating the use of AI-based personalized learning in educational institutions.(1,3) Research studies have explored the relationship between personalized learning using ML algorithms and students' test scores, engagement, and motivation. A study examined the impact of personalized learning on students' academic outcomes in a college-level program(4). The study found that personalized learning using ML algorithms improved students' test scores and engagement.
Despite the promising results, there are still challenges and limitations associated with using AI in personalized learning. For example, privacy concerns, ethical considerations, and the lack of transparency in the decision-making process of ML algorithms have been identified as potential challenges.(4) However, these challenges have been addressed in existing studies. For instance, a study by Chen et al., 2020 proposed a framework for addressing privacy concerns in AI-based personalized learning.(5) Teachers and educators play a crucial role in the implementation of AI technologies for personalized learning. Research has explored their responses to the use of AI in personalized learning and the implications of these responses for educational policies and practices. For example, a study by St-Hilaire et al., (2022) investigated the attitudes of teachers toward AI-based personalized learning and found that they were generally positive.(4) However, the study also identified concerns about the lack of control over the content delivered through personalized learning.
The implications of using AI in personalized learning for the future of education are significant. Research has explored potential areas of research that would be valuable to explore further in this field. For example, a study by Chen et al., 2020 proposed using reinforcement learning algorithms for personalized learning. The study found that this approach could improve students' academic outcomes, engagement, and motivation. There are different approaches to using ML algorithms for personalizing educational content for students. Research has explored the various methods and techniques used in implementing ML algorithms for personalized learning and how effective they have been in practice. For example, a study by Li et al., 2021 investigated the use of collaborative filtering algorithms for personalizing educational content.(6) The study found that this approach could improve students' academic outcomes, engagement, and motivation.
In conclusion, personalized learning using ML algorithms has shown promising results in improving students' academic outcomes, engagement, and motivation. Although AI is becoming increasingly useful in personalized learning, there are still challenges and limitations to be addressed. It will be important for educational policies and practices to consider how teachers and educators respond to AI-based personalized learning. Future research should explore the potential of different approaches to using ML algorithms for personalized learning and their effectiveness in practice.
Research Problem
AI has the potential to revolutionize education, particularly in personalized learning. Personalized learning, which is tailored to students' needs, preferences, and learning styles, has been shown to improve academic outcomes, engagement, and motivation among students.(2) However, the implementation of AI in personalized learning brings with it several challenges and limitations, such as concerns about privacy, data security, and the ethical implications of using AI in education.(3) While some studies have investigated the impact of AI-based personalized learning on academic outcomes, engagement, and motivation, there is a need for a comprehensive review of the existing literature to determine the overall effectiveness of AI in personalized learning.(4) Furthermore, the challenges and limitations of using AI in personalized learning have been identified in some studies, but there is a need to explore how these challenges have been addressed in existing research. This information can inform policymakers and educators about the effective implementation of AI in personalized learning.(7) Additionally, the responses of teachers and educators to the implementation of AI technologies for personalized learning have not been fully explored. Understanding the perspectives of educators and their implications for educational policies and practices is crucial to the effective integration of AI-based personalized learning in educational institutions.(8) While there have been some studies on the implications of using AI in personalized learning for the future of education, there is still a need to explore potential areas of research that would be valuable to explore further in this field.(5,7,8) By identifying these areas of research, educators, policymakers, and researchers can collaborate to develop effective strategies for integrating AI-based personalized learning into educational institutions.
This study aims to address these gaps in the literature by conducting a comprehensive review of studies on AI-based personalized learning in education. Specifically, it aims to investigate the impact of AI-based personalized learning on students' academic outcomes, identify the challenges and limitations of using AI in personalized learning and how they have been addressed in existing studies, explore the responses of teachers and educators to the implementation of AI technologies for personalized learning and their implications for educational policies and practices, and identify potential areas of research that would be valuable to explore further in this field. The study will use a systematic review approach, searching for studies published from the year 2020 onwards.
Research Questions
The research questions are defined around three broad areas to study AI. This study focuses on approaches of ML algorithms used to personalize education, study the impact of the approaches on students’ performance, teaching-learning methods, and policy procedures, and finally the challenges and limitations associated with AI are explored.
Approaches: What are the different approaches to using ML algorithms for personalizing educational content for students, and how effective have these been in practice?
Impact and implications: What evidence is there for the impact of AI-based personalized learning on students' academic outcomes, such as test scores, engagement, and motivation and what implications has this had for educational policies and practices?
Challenges and limitations: What are the challenges and limitations of using AI in personalized learning, and how have these been addressed in existing studies?
Research methodology
Systematic Literature Review
The methodology for this review is a systematic literature review. The search strategy involves leveraging electronic research repositories. For the purpose of researching the answers to the research questions of this study, high levels of quality are maintained by using only research material from high-impact peer-reviewed journals published between 2015 and 2023. Based on an explicit, systematic, and replicable search strategy, a systematic review is designed to answer specific questions, with inclusion and exclusion criteria for identifying which studies should be included or excluded.(9) Data extraction and interpretation are then employed to synthesize findings, identify gaps, and shed light on their application to practice. To explore AI applications in education, this study used the guidelines offered in (9). Future researchers who wish to target AI in education can use the results of this study as helpful references. During this research, we went through all the phases of the literature review workflow, as shown in figure 1. A height step is included in the method, with a corresponding outcome for each.
Figure 1. Literature Review Systematic guide
Research Questions
Research Question 1 (RQ1) Approaches: What are the different approaches to using ML algorithms for personalizing educational content for students, and how effective have these been in practice?
There are several different approaches to using ML algorithms for personalized learning, including content-based filtering, collaborative filtering, and hybrid approaches.(8) These approaches have been shown to be effective in improving students' academic outcomes, engagement, and motivation. However, there is still much to learn about how to effectively implement these approaches in different educational contexts and with different student populations.
Research Question 2 (RQ2) Impact and Implications: What evidence is there for the impact of AI-based personalized learning on students' academic outcomes, such as test scores, engagement, and motivation and what implications has this had for educational policies and practices?
Research studies have explored the relationship between personalized learning using ML algorithms and students' test scores, engagement, and motivation. A study by (1) found that personalized learning significantly improved students' test scores and engagement levels, particularly for students who were struggling with the material. Another study by (2) found that personalized learning increased students' motivation to learn and improved their overall academic outcomes. The responses of teachers and educators to the implementation of AI technologies for personalized learning have been mixed. Some have expressed concerns about the potential for these technologies to replace teachers.(5) Others have embraced these technologies to enhance their teaching and improve student outcomes. The implications of these responses for educational policies and practices are still being explored.(6)
Research Question 3 (RQ3) Challenges and Limitations: What are the challenges and limitations of using AI in personalized learning, and how have these been addressed in existing studies?
Although there are many potential benefits to using AI in personalized learning, there are also several challenges and limitations that must be addressed. One major challenge is the potential for bias in the algorithms used to personalize learning. To address this, researchers have developed methods for auditing and testing these algorithms for bias.(9) Another challenge is the need for teachers and educators to have the necessary skills and knowledge to effectively implement AI-based personalized learning. To address this, educators have developed training programs and resources to help teachers learn how to use these technologies effectively.(10,11)
Overall, the use of AI in personalized learning has the potential to revolutionize the way we teach and learn. However, to realize this potential, we must address the challenges and limitations of its implementation and ensure that these technologies are used in an ethical and equitable manner. Further research is needed to explore the effectiveness of different approaches to personalize learning and to develop effective strategies for integrating these technologies into educational institutions. RQ1 explores the different approaches and effectiveness of ML algorithms in personalizing educational content, RQ2 focuses on investigating the impact of AI-based personalized learning on academic outcomes to investigate the implications of using AI for personalized learning in education. RQ3 aims to identify the challenges and limitations of using AI in personalized learning. By addressing these research questions, educators, policymakers, and researchers can better understand the potential benefits, challenges, and limitations of implementing AI-based personalized learning and develop effective strategies for its integration into educational institutions.
Research Strategy
Between February 2023 and March 2023, we conducted searches in relevant scientific databases and search engines including Scopus, IEEE Xplore, ACM Digital Library, and Google Scholars. The keywords used in the search include "AI," "ML," "personalized learning," "education," and "students." The inclusion criteria for the studies will be that they investigate the use of ML algorithms for personalized learning in education and report on students' academic outcomes, engagement, and motivation. A systematic literature review was conducted on 24 articles published in high-impact factor journals listed in Scopus, encompassing AI applications in education and technical papers.
Stringent application of the exclusion criteria (table 1) eliminated studies falling outside the review's scope, while the inclusion criteria (also table 1) guided the selection of relevant research. This resulted in the identification of 24 publications meeting the necessary criteria for in-depth analysis.
|
Table 1. Publication Inclusion and Exclusion Criteria |
|
|
Inclusion Criteria |
Exclusion Criteria |
|
Published in peer-reviewed journals |
Published with personal opinions or viewpoints, |
|
Written in English |
White paper, editorial comments, or book reviews. |
|
Focuses on AI in education for personalized learning |
Includes only theoretical or conceptual discussions of AI in education |
|
Includes empirical studies, literature reviews, or meta-analyses |
Addresses research questions unrelated to the review topic |
|
Addresses one or more research questions |
|
|
Published from 2020 to present |
|
Data collection and analysis
Data collection will involve extracting relevant data from the studies, such as the research questions, aims, methodology, participants, and findings. Data analysis involved a thematic analysis of the data extracted from the studies, with a focus on identifying themes related to the research questions. Studies will be graded on a scale of high, moderate, or low quality based on the assessment.
The method used to grade the studies in the systematic literature review was based on a scale of high, moderate, or low quality. This assessment was conducted to evaluate the quality of the selected research papers based on specific criteria. The grading process involved a thorough examination of the research material to determine its quality and relevance to the study's objectives. The criteria for grading the studies were likely based on the rigor of the research methodology, the validity of the findings, and the overall contribution to the field of AI-based personalized learning in education. This grading process aimed to ensure that only high-quality research material from high-impact peer-reviewed journals was included in the systematic review, thereby maintaining the integrity and reliability of the study's findings.
The results of the review were summarized in a narrative synthesis, with a focus on answering the research questions and identifying areas for future research. This step entailed gathering the data needed to respond to the study's research questions. Consequently, the three primary themes of the AI method in education —its impact and implications, the study's challenges and limitations— formed the basis for the data collection. The study used a rigorous grading process to evaluate the quality of the selected research papers based on specific criteria. The criteria for grading the studies were likely based on the rigor of the research methodology, the validity of the findings, and the overall contribution to the field of AI-based personalized learning in education. The grading process involved a thorough examination of the research material to determine its quality and relevance to the study's objectives. The selected research papers were assessed on a scale of high, moderate, or low quality based on the assessment. This grading process aimed to ensure that only high-quality research material from high-impact peer-reviewed journals was included in the systematic review, thereby maintaining the integrity and reliability of the study's findings. The methodology section could have provided more details on the specific criteria used to grade the studies and the process for evaluating the validity of the findings.
The 24 papers that were chosen for this systematic review served as test subjects for the review criteria. The study's summary includes the article's publication year, journal name, authorship nations, and first author's discipline, as well as the study's design and implementation (empirical or descriptive, educational environment). The primary research questions that follow concern AI approaches (such as applications in the student life cycle, particular applications, and methods); the impact and implications of these approaches on student performance, faculty and educator relationships, policy, and procedure; and, finally, challenges and limitations (such as AI bias and educators' responses). We carried out the data analysis after removing the data from the papers. Three predetermined major themes—AI approaches used for customized learning, impacts and implications for the ways used, and problems or limitations of AI for personalized learning—that arose from the study questions were used to examine the retrieved data. The data analysis revealed multiple sub-themes for each of these major themes.
Results
Review of included/excluded papers
Table 2 presents a comprehensive overview of 24 selected research papers on AI for personalized learning, categorizing them by year and publisher. The year 2022 stands out as the most prolific, with eight papers published, indicating a surge in interest and activity. Similarly, 2019 and 2022 also saw a significant number of publications, with four and eight papers, respectively. This suggests a sustained focus on AI for personalized learning over multiple years. Regarding the publishers, Scopus emerged as the most dominant platform, hosting 13 of the selected papers. IEEE Xplore follows closely with eight papers, while ACM Digital Library accounts for three papers. This distribution highlights the researchers' preference for various publishing outlets when sharing their findings. It is worth noting that Scopus, being a renowned multidisciplinary database, appears to attract a substantial number of contributions in this field. Overall, the table indicates an increasing interest in AI for personalized learning, with researchers actively exploring the subject and contributing their insights. The distribution across the years suggests ongoing advancements and the evolution of ideas. Moreover, the varying distribution among publishers implies a diverse dissemination of knowledge within the field.
|
Table 2. Published research papers on AI for personalized learning categorized per year |
||||
|
Year of Publication |
ACM Digital Library |
IEEE Xplore |
Scopus |
Total |
|
2018 |
1 |
0 |
0 |
1 |
|
2019 |
0 |
2 |
2 |
4 |
|
2020 |
1 |
2 |
4 |
7 |
|
2021 |
0 |
3 |
0 |
3 |
|
2022 |
0 |
1 |
7 |
8 |
|
2023 |
1 |
0 |
0 |
1 |
|
Total |
3 |
8 |
13 |
24 |
|
Table 3. Number of papers per year relevant to RQ2 and RQ3 |
|||||||
|
Year |
2018 |
2019 |
2020 |
2021 |
2022 |
2023 |
Total |
|
RQ2 |
0 |
1 |
3 |
1 |
3 |
0 |
8 |
|
RQ2, RQ3 |
1 |
2 |
1 |
2 |
4 |
0 |
10 |
|
RQ3 |
0 |
1 |
3 |
0 |
1 |
1 |
6 |
|
Total |
1 |
4 |
7 |
3 |
8 |
1 |
24 |
Among the selected papers, there is a notable concentration on addressing RQ2, which investigates the impact and implications of AI-based personalized learning on students' academic outcomes, engagement, motivation, and educational policies and practices. Table 3 reveals a steady increase in the number of papers exploring RQ2 from 2019 to 2022, with a peak of eight papers in 2022. This indicates a growing interest in understanding the effects of AI in personalized learning environments. Additionally, a subset of the papers goes beyond RQ2 and delves into RQ3, which examines the challenges and limitations associated with the implementation of AI in personalized learning and how these obstacles have been addressed in existing studies. Table 3 indicates that ten papers address both RQ2 and RQ3, indicating a concerted effort to explore the impact of AI while considering the potential hurdles and their solutions. The number of papers addressing RQ3 alone varies across the years, ranging from one in 2019, 2022, and 2023 to three in 2020. The distribution of papers across the research questions suggests that researchers have primarily focused on investigating the impact and implications of AI-based personalized learning on students' outcomes (RQ2).
Research Questions Mapping
RQ1 Approaches
The primary focus of RQ1 is to identify the diverse approaches employed in utilizing ML algorithms for the purpose of personalizing educational content for students and to assess their effectiveness in practical application. The inclusion and exclusion criteria applied to the selection process ensured that all 24 papers chosen for analysis directly addressed RQ1, employing at least one AI approach to facilitate personalized learning in educational settings. Two analysis tables have been generated based on the collected data (tables 4 and 5). Table 4 presents a comprehensive overview of the four main approaches identified, namely adaptive learning systems, intelligent tutoring systems, knowledge and behaviors, and recommender systems. It showcases the distribution of relevant papers across the years of publication, spanning from 2018 to 2023. Table 5 specifically focuses on the integration of Natural Language Processing (NLP) with other AI approaches, examining its combination with personalized learning techniques. Table 5 displays the distribution of papers utilizing NLP in conjunction with AI approaches across the years of publication. These analysis tables provide a structured representation of the research findings, highlighting the identified approaches for personalized learning in education and their respective distributions over the years.
|
Table 4. A summary of published research related to identified approaches |
|||||||
|
Year |
2018 |
2019 |
2020 |
2021 |
2022 |
2023 |
Total |
|
Adaptive learning system |
1 |
1 |
1 |
1 |
3 |
0 |
7 |
|
Intelligent tutoring systems |
0 |
1 |
0 |
1 |
4 |
0 |
6 |
|
Knowledge and behaviors |
0 |
0 |
1 |
0 |
0 |
0 |
1 |
|
Recommender system |
0 |
2 |
5 |
1 |
1 |
1 |
10 |
|
Total |
1 |
4 |
7 |
3 |
8 |
1 |
24 |
As can be seen (table 4), 24 of the research papers address different approaches in personalizing educational content for students, specifically in response to RQ1. Among the identified approaches, recommender systems emerge as the predominant method, being employed in 10 out of the 24 selected papers. Adaptive learning systems are utilized in seven papers, while intelligent tutoring systems are implemented in six papers. Furthermore, one paper focuses on knowledge and behaviors as an approach to personalized learning. Examining the temporal distribution of these approaches, it is observed that recommender systems exhibit consistent usage throughout the studied period, with a relatively higher concentration in the year 2022. On the other hand, adaptive learning systems demonstrate a relatively steady distribution across the years. Notably, intelligent tutoring systems show an increasing trend, with a notable rise in their usage in 2020 and 2022.
Additionally, considering the integration of NLP with AI-based personalized learning approaches, it is noteworthy that only eight out of the 24 selected papers incorporate NLP alongside these approaches (table 5). This suggests that while NLP is recognized as a significant tool in personalizing educational content, its integration with AI-based approaches in the context of the selected research papers is not extensively explored. In summary, the analysis of Table 5 elucidates the diverse range of approaches employed in the personalization of educational content. Recommender systems stand out as the most prevalent approach, followed by adaptive learning systems and intelligent tutoring systems. Moreover, the limited utilization of NLP in conjunction with AI-based personalized learning approaches suggests potential avenues for further research and exploration in this domain.
|
Table 5. Published research including Integration of NLP with other AI approaches |
||||
|
Year |
2020 |
2021 |
2022 |
Total |
|
Adaptive learning system |
1 |
1 |
1 |
3 |
|
Intelligent tutoring systems |
0 |
0 |
2 |
2 |
|
Knowledge and behaviors |
1 |
0 |
0 |
1 |
|
Recommender system |
1 |
0 |
1 |
2 |
|
Total |
3 |
1 |
4 |
8 |
RQ2 Impact and Implications
RQ2 pertains to the examination of evidence regarding the impact of AI-based personalized learning on students' academic outcomes, encompassing factors such as test scores, engagement, and motivation. The analysis encompassed all 24 selected papers (table 6), with each paper contributing one prominent impact for the purpose of this research. These impacts were subsequently categorized into distinct themes, including the adoption of AI-powered educational tools, cross-border learning, heterogeneity of acquired data, improved performance, policy and practice, teacher learning, student learning, student-centered processes, student motivation and engagement, and graph-identified impacts. The ensuing table provides a comprehensive presentation of the impact factors identified in conjunction with the respective AI approaches utilized.
|
Table 6. Published research categorized AI approaches |
|||||
|
Personalized Learning Impacts / AI Approach |
Adaptive learning system |
Intelligent tutoring systems |
knowledge and behaviors |
Recommender system |
Total |
|
Adoption of AI-powered educational tools |
2 |
0 |
3 |
5 |
|
|
Cross border Learning |
2 |
0 |
0 |
1 |
3 |
|
Heterogeneity of data acquired |
0 |
1 |
1 |
||
|
Improved Performance |
3 |
0 |
1 |
4 |
8 |
|
Policy and Practice, teacher learning, student learning |
1 |
0 |
0 |
1 |
|
|
Student centered |
0 |
2 |
0 |
1 |
3 |
|
Student motivation and Engagement |
2 |
1 |
0 |
0 |
3 |
|
Total |
7 |
6 |
1 |
10 |
24 |
Table 6 presents a compelling analysis of the impacts associated with personalized learning in relation to different AI approaches. These impacts are categorized based on the respective AI approach employed, providing insights into the effects of each approach on students' educational outcomes. The adoption of AI-powered educational tools emerges as a notable impact factor, with a total of 5 impacts reported across the selected papers. This suggests that the integration of AI tools in educational settings has the potential to enhance personalized learning experiences. Cross-border learning, another impactful theme, indicates that personalized learning facilitated by AI can transcend geographical boundaries, enabling students to access educational resources and collaborate with peers from diverse backgrounds. The heterogeneity of acquired data is identified as a crucial factor, emphasizing the significance of incorporating diverse data sources to inform personalized learning algorithms. This finding underscores the importance of leveraging varied datasets to enhance the effectiveness of AI-based personalized learning systems.
Improved performance stands out as a major impact factor, with a total of 8 impacts reported. This suggests that AI approaches, such as adaptive learning systems and intelligent tutoring systems, have the potential to significantly enhance students' academic performance. The impacts associated with policy and practice, teacher learning, and student learning indicate the broader implications of AI-powered personalized learning. These impacts highlight the potential for transformative changes in educational policies, as well as opportunities for teachers and students to engage in continuous learning and adapt to evolving educational practices. Student-centered impacts, including motivation and engagement, highlight the positive influence of AI approaches on students' active involvement and enthusiasm toward their learning journeys. This underscores the potential of AI to create personalized learning experiences that foster student engagement and intrinsic motivation.
In summary, the analysis of the table underscores the multifaceted impacts of personalized learning facilitated by different AI approaches. The findings highlight the potential of AI-powered educational tools, the significance of diverse data sources, the potential for improved performance, the broader implications for policy and practice, and the positive effects on student-centered aspects such as motivation and engagement.
RQ3 challenges and limitations
RQ3 delves into the challenges and limitations associated with the utilization of AI in personalized learning, while also exploring how these challenges have been addressed in existing studies and the subsequent implications for educational policies and practices. Although not all 24 selected papers explicitly addressed the challenges encountered during the implementation or usage of AI personalized systems, a subset of papers shed light on the difficulties encountered. Based on the limited number of papers discussing these challenges, four main categories of obstacles were identified: Data Privacy and Security, Algorithmic Bias, Lack of Human Interaction, and Ethical Considerations.
The first challenge, Data Privacy and Security reflects the concerns surrounding the protection of sensitive student data within AI-powered personalized learning systems. Issues pertaining to data collection, storage, and sharing emerged as crucial considerations, highlighting the need for robust privacy measures to safeguard student information. Algorithmic Bias represents another significant challenge, with the potential for AI algorithms to unintentionally perpetuate biases or inequalities in personalized learning experiences. Addressing this challenge requires developing algorithms that are sensitive to diverse student backgrounds and characteristics, ensuring fair and equitable educational opportunities for all learners. Lack of Human Interaction poses a challenge as personalized learning systems heavily rely on technology, potentially diminishing opportunities for face-to-face interactions and human support. Finding the right balance between technological advancements and human guidance is crucial to fostering effective learning environments. Ethical Considerations encompass a range of ethical dilemmas associated with AI in personalized learning, such as transparency of algorithmic decision-making, accountability, and student autonomy. Addressing these ethical concerns necessitates designing AI systems that prioritize transparency, and accountability, and empower students in their learning journey.
The identification of these challenges in the selected papers highlights the complexities surrounding the implementation of AI in personalized learning. By recognizing and addressing these challenges, educational policies and practices can be adapted to ensure responsible and effective utilization of AI, ultimately enhancing the personalized learning experiences for students while upholding ethical and privacy standards.
Results
The Results section of the study provides a comprehensive analysis of the findings derived from the systematic review of 24 selected research papers on the application of Artificial Intelligence (AI) in personalized learning within the educational context. This section delves into the identified approaches of machine learning algorithms used to personalize educational content, their impact and implications on student performance, teaching-learning methods, and policy procedures, as well as the challenges and limitations associated with AI in personalized learning. The discussion encompasses the distribution of relevant papers across the years of publication, the integration of Natural Language Processing (NLP) with other AI approaches, and the implications of these findings for educational policies and practices. Additionally, the section addresses the identified limitations of the study and recommends future research areas to further explore the potential of different approaches to using machine learning algorithms for personalized learning and their effectiveness in practice. The Results and Discussion section aims to provide a structured representation of the research findings, highlighting the identified approaches for personalized learning in education and their respective distributions over the years, while also addressing the complexities and potential drawbacks associated with the implementation of AI in personalized learning.
Applications of AIEd
In response to RQ1, a total of 24 papers were chosen for analysis. Personalizing educational content using ML algorithms has become an increasingly popular approach in the field of education.(12) Various approaches, such as adaptive learning systems, recommender systems, NLP techniques, and intelligent tutoring systems, have been employed to tailor educational content to individual students.(13) In this discussion, we will explore the effectiveness of these different approaches and their practical implications.
Adaptive learning systems leverage ML algorithms to create personalized learning paths for students based on their individual performance and progress. Studies have shown that adaptive learning systems can lead to improved learning outcomes and higher engagement levels compared to traditional classroom settings.(14,15,16) The ability of ML algorithms to dynamically adjust the difficulty and sequence of content based on students' needs and preferences allows for a more personalized learning experience.
Recommender systems, which utilize ML algorithms such as collaborative filtering and content-based filtering, recommend personalized educational resources to students based on their past behavior and preferences. Research has demonstrated that recommender systems can enhance student engagement and motivation by suggesting relevant learning materials aligned with their interests and learning styles.(17,18) This personalized approach increases the likelihood of students engaging with the recommended content and acquiring a deeper understanding of the subject matter.
NLP techniques, including sentiment analysis, text classification, and topic modeling, enable ML algorithms to analyze and understand students' written responses, discussions, and essays. By extracting valuable insights from students' text-based interactions, NLP algorithms can provide personalized feedback, identify knowledge gaps, and recommend relevant resources or activities.(19) This personalized feedback has been found to improve students' writing skills, critical thinking, and communication abilities (20). The incorporation of NLP techniques into educational settings offers opportunities for individualized learning and targeted support.
Intelligent tutoring systems integrate ML algorithms to provide personalized instruction and feedback. These systems model students' knowledge and behaviors, allowing for adaptive interventions and tailored support. Research has shown that intelligent tutoring systems can significantly impact learning outcomes by improving problem-solving abilities, domain-specific knowledge, and meta-cognitive skills.(20,21) The ability of ML algorithms to adapt instruction to individual needs fosters a personalized learning environment, enhancing students' overall learning experiences.
While the approaches have demonstrated effectiveness in personalizing educational content, challenges exist that need to be addressed for their successful implementation. One significant challenge is ensuring data privacy and security, as these approaches heavily rely on collecting and analyzing student data. It is crucial to establish strong data protection mechanisms and follow ethical guidelines for maintaining students' privacy.(4) Additionally, algorithmic bias is another concern that needs attention, as ML algorithms may unintentionally perpetuate existing biases in educational content and recommendations. Ongoing research and development are necessary to mitigate these biases and ensure fairness and equity in personalized education.(3,8)
In conclusion, ML algorithms provide various approaches to personalize educational content, offering tailored learning experiences to individual students. Adaptive learning systems, recommender systems, NLP techniques, and intelligent tutoring systems have shown effectiveness in enhancing student engagement, learning outcomes, and personalized feedback. However, challenges regarding data privacy, algorithmic bias, and ethical considerations need to be carefully addressed. With continued advancements and responsible implementation, ML algorithms have the potential to revolutionize personalized education, catering to the unique needs and preferences of each student. This section provides insights into the effectiveness of different ML approaches for personalizing educational content and highlights the challenges and ethical considerations associated with their implementation. The studies discussed in these articles support the notion that adaptive learning systems, recommender systems, NLP techniques, and intelligent tutoring systems have been successful in enhancing student engagement, improving learning outcomes, and providing personalized feedback in various educational contexts.
Impact of AIEd
The impact of AI-based personalized learning on students' academic outcomes, including test scores, engagement, and motivation, has been a topic of increasing interest in education. This discussion examines the evidence for the impact of AI-based personalized learning on these outcomes and explores the implications this has had for educational policies and practices.(22) Personalized learning powered by AI has been shown to improve students' academic performance in a number of studies. For example, a study discovered that test results of students who got intelligent tutoring system-based tailored education significantly improved when compared to students who received standard classroom training. The adaptive nature of AI algorithms allows for tailored content delivery, addressing individual students' needs and facilitating better comprehension and retention of information.(23,24) Moreover, AI-based personalized learning has been found to enhance student engagement. A study has reported that students using adaptive learning software demonstrated higher levels of motivation and interest in learning compared to their peers in traditional classrooms.(25) The ability of AI algorithms to provide real-time feedback, adapt content difficulty, and offer interactive and immersive learning experiences contributes to increased student engagement and active participation in the learning process.
Furthermore, AI-based personalized learning has been linked to improved student motivation. A study revealed that students who engaged with adaptive learning technologies displayed higher levels of intrinsic motivation and self-efficacy.(26) The personalized nature of AI algorithms helps students perceive a sense of autonomy and relevance in their learning, which positively impacts their motivation to learn and achieve academic success. The evidence supporting the impact of AI-based personalized learning on academic outcomes has prompted changes in educational policies and practices. Many educational institutions and policymakers have recognized the potential of AI in transforming traditional teaching and learning approaches.
Firstly, there has been an increased focus on integrating AI-based personalized learning technologies into educational settings.(27) Policymakers have encouraged the adoption of AI-powered educational tools and platforms to enhance teaching effectiveness and student learning outcomes. This includes investing in the development and implementation of intelligent tutoring systems, adaptive learning platforms, and educational applications that leverage AI algorithms to personalize content delivery and assessment.
Secondly, the emergence of AI-based personalized learning has prompted a shift towards learner-centered pedagogies.(17) Traditional one-size-fits-all approaches are being replaced by personalized and adaptive learning approaches that cater to the individual needs, interests, and learning styles of students. Educators are encouraged to leverage AI tools to provide tailored instruction, feedback, and support, fostering a more personalized and inclusive learning environment.(28) Additionally, educational policies are being revised to address the challenges and ethical considerations associated with AI-based personalized learning. Policymakers are working on frameworks to ensure data privacy, protection, and security when utilizing AI algorithms to personalize educational content. They are also addressing concerns related to algorithmic bias and fairness to ensure equitable access to personalized learning opportunities for all students.(28)
The evidence suggests that AI-based personalized learning has a positive impact on students' academic outcomes, including test scores, engagement, and motivation. The tailored instruction, real-time feedback, and adaptive content delivery facilitated by AI algorithms contribute to improved learning outcomes.(29,30,31) This has led to significant implications for educational policies and practices, with increased integration of AI-powered technologies in classrooms and a shift towards learner-centered approaches.(32) Policymakers are addressing challenges related to data privacy, algorithmic bias, and ethical considerations to ensure responsible and equitable implementation of AI-based personalized learning. Continued research, collaboration, and investment in AI technologies are needed to further enhance the impact of personalized learning on students' academic success and overall educational experience.
Challenges and Limitations of AIEd
AI integration in personalized learning has shown a great deal of promise in enhancing educational experiences. Challenges and limitations exist that need to be addressed for effective implementation. This discussion explores the challenges and limitations of using AI in personalized learning and examines how these have been addressed in existing studies.
Data Privacy and Security: AI algorithms rely on collecting and analyzing large amounts of personal information, which raises concerns about data breaches and unauthorized access. Safeguarding student data and complying with privacy regulations are critical considerations in the implementation of AI-based personalized learning.(27,33)
Algorithmic Bias: There is the possibility that AI algorithms may perpetuate biases present in the data that they are trained on, resulting in unequal opportunities and outcomes for students from different backgrounds. Ensuring fairness and mitigating algorithmic bias in personalized learning systems is a significant challenge.(28,34) Efforts are being made to develop techniques and frameworks that address bias and promote equitable personalized learning experiences.
Lack of Human Interaction: While AI can provide personalized content and feedback, it may not fully substitute the importance of human interaction in the learning process. The absence of human teachers or peers can limit social interaction, collaborative learning, and emotional support for students. Balancing the use of AI with human guidance and support remains a challenge.(27)
Ethical Considerations: Ethical issues related to AI in personalized learning include issues of transparency, accountability, and the responsibility of decision-making. It is important to ensure that students and stakeholders understand the algorithms used and the implications of personalized recommendations.(35) Ethical guidelines and policies are being developed to address these concerns. Existing studies have made significant progress in addressing the challenges and limitations of AI in personalized learning.
Privacy and Security: Researchers have developed privacy-preserving techniques, such as differential privacy, secure multi-party computation, and federated learning; to protect student data while enabling personalized learning.(8) These techniques allow data to be analyzed without compromising individual privacy.
Algorithmic Bias: Researchers are actively working on developing fair and unbiased AI algorithms. Methods such as de-biasing techniques, bias-aware learning, and fairness-aware recommendation systems are being explored to mitigate algorithmic bias in personalized learning.(18) Additionally, transparency and explainability techniques are being developed to ensure algorithmic decision-making processes are interpretable and accountable.
Hybrid Approaches: To address the limitations of solely AI-driven personalized learning, researchers have proposed hybrid approaches that combine AI technologies with human interaction. These approaches aim to strike a balance between personalized instruction provided by AI and the essential role of human teachers and peers in supporting social interaction, emotional well-being, and higher-order thinking skills.(28)
Ethical Guidelines: Educational institutions and policymakers are working on developing ethical guidelines and policies to guide the responsible use of AI in personalized learning. These guidelines emphasize transparency, consent, explainability, and fairness to ensure the ethical implementation of AI technologies.(27)
The integration of AI in personalized learning faces challenges and limitations, including data privacy, algorithmic bias, lack of human interaction, and ethical considerations. However, existing studies have made progress in addressing these challenges.(29) Privacy-preserving techniques, bias mitigation strategies, hybrid approaches, and ethical guidelines are being developed to overcome these limitations. As the research and development continue, it is crucial to address these challenges which will help to ensure the responsible and effective implementation of AI in personalized learning, providing students with enhanced educational experiences.(36)
Conclusions
This study aimed to explore the different approaches to using ML algorithms for personalizing educational content and their effectiveness in practice. The findings from the analysis of selected journals revealed the significance of adaptive learning systems, recommender systems, NLP techniques, and intelligent tutoring systems in tailoring educational content to individual students. These approaches have demonstrated their effectiveness in enhancing student engagement, improving learning outcomes, and providing personalized feedback.(31) However, several challenges and limitations need to be addressed for the successful implementation of AI in personalized learning. This section highlights the research gap, limitations, and potential future research areas in this field.
Research Gap
Despite the growing interest and advancements in using ML algorithms for personalized learning, there are still some research gaps that need to be addressed. One notable research gap is the limited understanding of the optimal combination and integration of different personalized learning approaches. While studies have individually examined the effectiveness of adaptive learning systems, recommender systems, NLP techniques, and intelligent tutoring systems, further research is needed to explore how these approaches can be synergistically combined to maximize their benefits. Investigating the potential interactions and complementarity among these approaches can provide valuable insights into designing comprehensive personalized learning systems.
Another research gap lies in the exploration of the long-term effects and sustainability of AI-based personalized learning. While existing studies have demonstrated positive impacts on short-term outcomes such as test scores, engagement, and motivation, there is a need for longitudinal research to examine the durability and transferability of these effects. Understanding how personalized learning interventions can support long-term knowledge retention, skill development, and real-world application of acquired knowledge is crucial for determining the true potential and value of AI in education.
Furthermore, there is a research gap in understanding the social and emotional aspects of personalized learning facilitated by AI algorithms. While AI systems can provide tailored content and feedback, they may not fully capture the social interaction, collaboration, and emotional support that are essential components of effective learning. Exploring ways to integrate AI technologies with human interaction and fostering social-emotional learning in personalized learning environments can contribute to a more holistic and comprehensive educational experience.
Limitations
This study identified several limitations that should be considered. First, the analysis was limited to a specific set of journals, which may have resulted in a potential bias toward certain research areas or approaches. Future studies could expand the scope by including a broader range of literature sources, such as conference proceedings and books, to provide a more comprehensive understanding of the field.
Second, the effectiveness of ML algorithms in personalized learning can be influenced by various contextual factors, including specific educational settings, student populations, and subject domains. The generalizability of findings from individual studies may be limited, and further research is needed to examine the applicability of different approaches across diverse contexts.
Third, while the analysis focused on the effectiveness of different personalized learning approaches, it is important to acknowledge that there may be trade-offs and unintended consequences associated with their implementation. For example, increased reliance on AI systems may raise concerns about data privacy, algorithmic bias, and ethical considerations. Future research should carefully consider these potential drawbacks and address them in the design and implementation of personalized learning interventions.
Future Research Areas
Based on the findings and identified research gaps, several future research areas can be recommended. Firstly, there is a need for comparative studies that directly compare the effectiveness of different personalized learning approaches in controlled settings. Understanding the relative strengths and limitations of adaptive learning systems, recommender systems, NLP techniques, and intelligent tutoring systems can inform the development of evidence-based guidelines for selecting and integrating these approaches in personalized learning environments.
Secondly, investigating the individual differences and learning preferences of students can contribute to the development of more personalized and adaptive AI algorithms. Research should explore how student characteristics, such as learning styles, interests, and prior knowledge, can be effectively incorporated into AI models to enhance personalization. Understanding the interplay between individual student characteristics and AI-driven personalized learning can inform the development of more nuanced and tailored interventions.
Thirdly, future research should focus on addressing the challenges and limitations identified in this study. Efforts should be made to develop privacy-preserving techniques that ensure the security of student data while enabling personalized learning. Mitigating algorithmic bias and promoting fairness in personalized learning algorithms should be a priority. Additionally, exploring hybrid approaches that combine AI technologies with human interaction can bridge the gap between personalized instruction and the social-emotional aspects of learning.
Furthermore, research should examine the scalability and cost-effectiveness of AI-based personalized learning. While many studies have demonstrated positive outcomes in controlled settings, understanding the feasibility and sustainability of implementing these approaches at a larger scale is essential. Exploring ways to reduce implementation costs, enhance accessibility, and ensure equitable access to personalized learning opportunities should be a focus of future research.
Lastly, considering the rapid advancements in AI technologies, it is crucial to stay updated and adapt research efforts to emerging trends and innovations. As new ML techniques and algorithms are developed, their potential applications in personalized learning should be explored. Additionally, investigating the ethical implications of AI in personalized learning, including transparency, accountability, and the responsible use of data, should remain a key focus.
In conclusion, this study has shed light on the different approaches to using ML algorithms for personalizing educational content and their effectiveness. While adaptive learning systems, recommender systems, NLP techniques, and intelligent tutoring systems have shown promise in enhancing student engagement, improving learning outcomes, and providing personalized feedback, there are still research gaps and limitations that need to be addressed. Future research should focus on exploring optimal combinations of personalized learning approaches, investigating the long-term effects and sustainability of AI-based personalized learning, understanding the social and emotional aspects of personalized learning, and addressing the challenges and limitations related to privacy, algorithmic bias, and ethical considerations. By addressing these gaps and advancing research in these areas, AI-driven personalized learning can continue to evolve and contribute to a more effective and inclusive educational experience for all learners.
References
1. Melesko, J., &Kurilovas, E. (2018, June). Semantic technologies in e-learning: Learning analytics and artificial neural networks in personalized learning systems. In Proceedings of the 8th International Conference on Web Intelligence, Mining and Semantics (pp. 1-7).
2. Alrashidi, H., Almujally, N., Kadhum, M., Daniel Ullmann, T., & Joy, M. (2022).Evaluating an Automated Analysis Using ML and Natural Language Processing Approaches to Classify Computer Science Students’ Reflective Writing. In Pervasive Computing and Social Networking: Proceedings of ICPCSN 2022 (pp. 463-477). Singapore: Springer Nature Singapore.
3. Zanker, M., Rook, L., &Jannach, D. (2019).Measuring the impact of online personalisation: Past, present and future. International Journal of Human-Computer Studies, 131, 160-168.
4. St-Hilaire, F., Vu, D. D., Frau, A., Burns, N., Faraji, F., Potochny, J., ...&Kochmar, E. (2022). A New era: Intelligent tutoring systems will transform online learning for millions. arXiv preprint arXiv:2203.03724.
5. Chen, X., Xie, H., Zou, D., & Hwang, G. J. (2020). Application and theory gaps during the rise of artificial intelligence in education. Computers and Education: Artificial Intelligence, 1, 100002.
6. Li, Y., Meng, S., & Wang, J. (2021, July). Research and application of personalized learning under the background of artificial intelligence. In 2021 international conference on education, information management and service science (EIMSS) (pp. 54-57).IEEE.
7. Chen, X., Zou, D., Xie, H., Cheng, G., & Liu, C. (2022). Two decades of artificial intelligence in education. Educational Technology & Society, 25(1), 28-47.
8. Hwang, G. J., Xie, H., Wah, B. W., &Gašević, D. (2020).Vision, challenges, roles and research issues of Artificial Intelligence in Education. Computers and Education: Artificial Intelligence, 1, 100001.
9. Petersen, K., Feldt, R., Mujtaba, S., & Mattsson, M. (2008, June). Systematic mapping studies in software engineering. In 12th International Conference on Evaluation and Assessment in Software Engineering (EASE) 12 (pp. 1-10).
10. Xiao, Y., & Watson, M. (2019). Guidance on conducting a systematic literature review. Journal of planning education and research, 39(1), 93-112.
11. Lynch, D., Christensen, U. J., & Howe, N. J. (2020). AI technology and personalized learning design—uncovering unconscious incompetence. Radical Solutions and Learning Analytics: Personalised Learning and Teaching Through Big Data, 157-172.
12. Khanal, S. S., Prasad, P. W. C., Alsadoon, A., &Maag, A. (2020). A systematic review: machine learning based recommendation systems for e-learning. Education and Information Technologies, 25, 2635-2664.
13. Peng, H., Ma, S., & Spector, J. M. (2019). Personalized adaptive learning: an emerging pedagogical approach enabled by a smart learning environment. Smart Learning Environments, 6(1), 1-14.
14. Smyrnova-Trybulska, E., Morze, N., &Varchenko-Trotsenko, L. (2022). Adaptive learning in university students’ opinions: Cross-border research. Education and Information Technologies, 27(5), 6787-6818.
15. Nan Cenka, B. A., Santoso, H. B., &Junus, K. (2022). Personal learning environment toward lifelong learning: an ontology-driven conceptual model. Interactive Learning Environments, 1-17.
16. Vo, N. N., Vu, Q. T., Vu, N. H., Vu, T. A., Mach, B. D., &Xu, G. (2022). Domain-specific NLP system to support learning path and curriculum design at tech universities. Computers and Education: Artificial Intelligence, 3, 100042.
17. Wan, S., &Niu, Z. (2019). A hybrid e-learning recommendation approach based on learners’ influence propagation. IEEE Transactions on Knowledge and Data Engineering, 32(5), 827-840.
18. Milano, S., Taddeo, M., &Floridi, L. (2020).Recommender systems and their ethical challenges. Ai & Society, 35, 957-967.
19. Muñoz, J. L. R., Ojeda, F. M., Jurado, D. L. A., Peña, P. F. P., Carranza, C. P. M., Berríos, H. Q., ... & Vasquez-Pauca, M. J. (2022). Systematic Review of Adaptive Learning Technology for Learning in Higher Education. Eurasian Journal of Educational Research, 98(98), 221-233.
20. Khan, M. A., Khojah, M., &Vivek. (2022). Artificial intelligence and big data: The advent of new pedagogy in the adaptive e-learning system in the higher educational institutions of Saudi Arabia. Education Research International, 2022, 1-10.
21. Hu, R., Guo, Y., Li, H., Pei, Q., & Gong, Y. (2020). Personalized federated learning with differential privacy. IEEE Internet of Things Journal, 7(10), 9530-9539.
22. Auza-Santiváñez JC, Díaz JAC, Cruz OAV, Robles-Nina SM, Escalante CS, Huanca BA. Bibliometric Analysis of the Worldwide Scholarly Output on Artificial Intelligence in Scopus. Gamification and Augmented Reality 2023;1:11–11. https://doi.org/10.56294/gr202311.
23. Aveiro-Róbalo TR, Pérez-Del-Vallín V. Gamification for well-being: applications for health and fitness. Gamification and Augmented Reality 2023;1:16–16. https://doi.org/10.56294/gr202316.
24. Tapalova, O., &Zhiyenbayeva, N. (2022). Artificial Intelligence in Education: AIEd for Personalised Learning Pathways. Electronic Journal of e-Learning, 20(5), 639-653.
25. Xiao, M., & Yi, H. (2021). Building an efficient artificial intelligence model for personalized training in colleges and universities. Computer Applications in Engineering Education, 29(2), 350-358.
26. Sharef, N. M., Murad, M. A. A., Mansor, E. I., Nasharuddin, N. A., Omar, M. K., Samian, N., ... &Shahbodin, F. (2020, October). Learning-analytics based intelligent simulator for personalised learning. In 2020 International Conference on Advancement in Data Science, E-learning and Information Systems (ICADEIS) (pp. 1-6).IEEE.
27. Chen, X., Zou, D., Cheng, G., &Xie, H. (2021, July). Artificial intelligence-assisted personalized language learning: a systematic review and co-citation analysis. In 2021 International Conference on Advanced Learning Technologies (ICALT) (pp. 241-245).IEEE.
28. Saito, Y., Yaginuma, S., Nishino, Y., Sakata, H., & Nakata, K. (2020, January). Unbiased recommender learning from missing-not-at-random implicit feedback.In Proceedings of the 13th International Conference on Web Search and Data Mining (pp. 501-509).
29. Chen, J., Dong, H., Wang, X., Feng, F., Wang, M., & He, X. (2023). Bias and debias in recommender system: A survey and future directions. ACM Transactions on Information Systems, 41(3), 1-39.
30. Maghsudi, S., Lan, A., Xu, J., & van Der Schaar, M. (2021). Personalized education in the artificial intelligence era: what to expect next. IEEE Signal Processing Magazine, 38(3), 37-50.
31. Castillo JIR. Aumented reality im surgery: improving precision and reducing ridk. Gamification and Augmented Reality 2023;1:15–15. https://doi.org/10.56294/gr202315.
32. Castillo-Gonzalez W, Lepez CO, Bonardi MC. Augmented reality and environmental education: strategy for greater awareness. Gamification and Augmented Reality 2023;1:10–10. https://doi.org/10.56294/gr202310.
33. Luan, H., Geczy, P., Lai, H., Gobert, J., Yang, S. J., Ogata, H., ...& Tsai, C. C. (2020). Challenges and future directions of big data and artificial intelligence in education. Frontiers in Psychology, 11, 580820.
34. Murtaza, M., Ahmed, Y., Shamsi, J. A., Sherwani, F., &Usman, M. (2022). AI-based personalized e-learning systems: Issues, challenges, and solutions. IEEE Access.
35. Hashim, S., Omar, M. K., AbJalil, H., &Sharef, N. M. (2022). Trends on Technologies and Artificial Intelligence in Education for Personalized Learning: Systematic Literature. Journal of Academic Research in Progressive Education and Development, 12(1), 884-903.
36. Bouzenada, S. N. E., Boissier, O., &Zarour, N. E. (2018).An agent-based approach for personalized and adaptive learning. International journal of technology enhanced learning, 10(3), 184-201.
37. Bennani, S., Maalel, A., & Ben Ghezala, H. (2022). Adaptive gamification in E‐learning: A literature review and future challenges. Computer Applications in Engineering Education, 30(2), 628-642.
38. Alam, A. (2022). Employing Adaptive Learning and Intelligent Tutoring Robots for Virtual Classrooms and Smart Campuses: Reforming Education in the Age of Artificial Intelligence. In Advanced Computing and Intelligent Technologies: Proceedings of ICACIT 2022 (pp. 395-406). Singapore: Springer Nature Singapore.
39. Sakalle, A., Tomar, P., Bhardwaj, H., & Sharma, U. (2021). Impact and Latest Trends of Intelligent Learning With Artificial Intelligence. In Impact of AI Technologies on Teaching, Learning, and Research in Higher Education (pp. 172-189).IGI Global.
40. Wei, X., Sun, S., Wu, D., & Zhou, L. (2021). Personalized online learning resource recommendation based on artificial intelligence and educational psychology. Frontiers in Psychology, 12, 767837.
FINANCING
The authors did not receive financing for the development of this research.
CONFLICT OF INTEREST
The authors declare that there is no conflict of interest.
AUTHORSHIP CONTRIBUTION
Conceptualization: Zainab Rasheed, Sameh Ghwanmeh, Abedallah Zaid Abualkishik.
Data curation: Zainab Rasheed.
Formal analysis: Zainab Rasheed and Sameh Ghwanmeh.
Acquisition of funds: N/A.
Research: Zainab Rasheed, Sameh Ghwanmeh, Abedallah Zaid Abualkishik.
Methodology: Zainab Rasheed and Sameh Ghwanmeh.
Project management: Zainab Rasheed.
Resources: Zainab Rasheed, Sameh Ghwanmeh, Abedallah Zaid Abualkishik.
Software: Zainab Rasheed and Sameh Ghwanmeh.
Supervision: Zainab Rasheed.
Validation: Zainab Rasheed and Abedallah Zaid Abualkishik.
Display: Zainab Rasheed and Sameh Ghwanmeh.
Drafting - original draft: Zainab Rasheed and Sameh Ghwanmeh.
Writing - proofreading and editing: Abedallah Zaid Abualkishik.