Smart Learning Ecosystems: A SWOT Exploration of AI, IoT, and Big Data
A
AlMahdiKhaddar1
YoussefSaid1
AbdellahBakhouyi2
AmineDehdi1,3
TarikChafiq1
1Laboratory of Information Processing (LTI), Faculty of Sciences Ben M’SickUniversity Hassan II of CasablancaCasablancaMorocco
2ENSAD Casablanca, Hassan II University of CasablancaCasablancaMorocco
3EST Dakhla, Ibn Zohr University of AgadirAgadirMorocco
Al Mahdi Khaddar1, Youssef Said1, Abdellah Bakhouyi2, Amine Dehdi1,3, Tarik Chafiq1
1 Laboratory of Information Processing (LTI), Faculty of Sciences Ben M’Sick, University Hassan II of Casablanca, Casablanca, Morocco.
2 ENSAD Casablanca, Hassan II University of Casablanca, Casablanca, Morocco.
3 EST Dakhla, Ibn Zohr University of Agadir, Agadir, Morocco.
Abstract -
A
In recent years, Artificial Intelligence (AI), the Internet of Things (IoT), and Big Data have become pivotal drivers of educational innovation, redefining how learning is personalized, monitored, and scaled. This systematic literature review investigates how these emerging technologies are shaping personalized learning by conducting a comprehensive SWOT analysis based on twenty-four peer-reviewed studies published between 2020 and 2025, selected according to the PRISMA framework. The analysis reveals significant strengths, including AI-enabled adaptive learning pathways, intelligent feedback systems, and real-time analytics that enhance student engagement and learning outcomes. Conversely, major weaknesses persist, such as high implementation costs, limited data interoperability, and insufficient digital readiness among educators. On the opportunity side, the findings point to the potential for interoperable data ecosystems, modular AI-supported pedagogical designs, and the development of scalable smart learning environments. However, several threats remain most notably data privacy vulnerabilities, algorithmic bias, and a growing digital divide across institutions. The review concludes that achieving the transformative promise of these technologies requires a holistic and ethically grounded strategy that prioritizes data governance, equity, and teacher empowerment. Such an approach is essential to ensure that technology-enhanced learning evolves into an inclusive and sustainable educational paradigm.Keywords -
Personalized Learning
Smart Education
Artificial Intelligence (AI)
Internet of Things (IoT)
Big Data
SWOT Analysis
Systematic Review
A
A
1. Introduction
Over the last decade, education has undergone a profound digital transformation. Emerging technologies such as Artificial Intelligence (AI), the Internet of Things (IoT), and Big Data have become powerful catalysts for redefining how knowledge is delivered, accessed, and evaluated [1]. AI enables intelligent systems to process vast amounts of data, allowing learning platforms to adapt content dynamically to the individual needs, pace, and performance of each student [2]. Meanwhile, IoT connects smart devices including sensors, wearable tools, and interactive objects that continuously monitor learners’ activities and provide valuable real-time insights into engagement and motivation [3]. Big Data complements these systems by integrating diverse sources of educational information, helping institutions anticipate learning trends and intervene when early warning signs appear [4].
A
Although the idea of personalized learning based on individual capabilities is not new, its large-scale implementation has become feasible thanks to these converging technologies. They enable continuous assessment that extends beyond traditional tests to include everyday performance metrics. For instance, connected classrooms and biofeedback devices can monitor variables such as attention span, emotional state, or stress level [5]. By processing these data through machine learning and deep learning algorithms, educators can obtain immediate feedback and generate targeted recommendations ranging from personalized study materials to real-time alerts that signal when intervention is needed [6].However, the convergence of AI, IoT, and Big Data in education raises critical ethical and operational concerns. As these systems collect increasingly detailed data, the boundaries between learning support and surveillance become blurred. Issues of data privacy, algorithmic transparency, and technological ownership must be addressed to ensure that innovation remains responsible and inclusive [7]. Moreover, many educators still lack the technical training or confidence required to integrate these tools effectively into their pedagogical practices. This gap between technological potential and real-world application highlights the need for a stronger institutional framework that promotes teacher capacity building and ethical data use.
A
To respond to these challenges, the present study conducts a Systematic Literature Review (SLR) following the PRISMA protocol and employs a SWOT framework to assess the strengths, weaknesses, opportunities, and threats associated with the integration of AI, IoT, and Big Data in education. The main research question guiding this analysis is: Under what conditions can the integration of AI, IoT, and Big Data achieve its transformative potential for personalized learning in an equitable, ethical, and effective manner? By synthesizing the findings from recent peer-reviewed studies, this paper aims to provide a structured overview of the current state of research, identify recurring implementation barriers, and outline strategic directions for the future of smart learning systems.2. Methodology
This study adopts a Systematic Literature Review (SLR) approach to examine how Artificial Intelligence (AI), the Internet of Things (IoT), and Big Data are jointly applied to create personalized learning environments. The goal is to provide a structured and comprehensive overview of current practices, challenges, and research gaps in this rapidly evolving domain.
Several key elements were examined in the selected studies, including the types of IoT devices used, the educational contexts in which they were deployed, the nature and format of the data collected, the features extracted, preprocessing techniques, and the algorithms employed from both Big Data and machine learning domains. The aim of the literature search was to identify, select, and critically evaluate relevant publications that address specific research questions [8]. This review was conducted in accordance with the PRISMA protocol and followed the Kitchenham guidelines, ensuring methodological rigor, transparency, and traceability throughout each stage of the review process [9].
2.1 Research Questions (RQs)
The study is guided by three central research questions designed to explore the technological, pedagogical, and ethical dimensions of AI-IoT-Big Data integration in education:
RQ1: Which AI, IoT, and Big Data technologies or architectures are currently employed to develop personalized learning systems?
RQ2: What measurable impacts do these integrated systems have on learner engagement, performance, and motivation?
RQ3: What are the main technical, ethical, organizational, and financial challenges associated with combining these technologies in educational contexts?
Together, these questions establish the analytical foundation for evaluating both the potential and the constraints of intelligent learning ecosystems.
2.2 Search Strategy
To ensure a comprehensive coverage of the literature, searches were conducted in two major scientific databases Scopus and Web of Science because of their extensive indexing of peer-reviewed journals across education, computer science, and social-science domains. The search queries combined Boolean operators and relevant keywords such as “smart learning”, “Internet of Things”, “Big Data”, and “Artificial Intelligence”. A systematic filtering process was applied to refine the results and identify publications that explicitly addressed our research questions [10][11]. Each retrieved article was screened for methodological soundness and thematic relevance before inclusion.
2.3 Keywords
Keywords were grouped into three conceptual categories reflecting the scope of this review:
1.
Learning Context: learning, personalized learning, e-learning
2.
Technological Dimension: Artificial Intelligence, Internet of Things, Big Data
3.
Analytical Method: SWOT analysis, systematic review
This taxonomy ensured that the retrieved studies represented both pedagogical and technological perspectives, as well as methodological rigor.
2.4 Selection and Quality Assessment
To guarantee validity and consistency, specific inclusion and exclusion criteria were applied.
A.
Inclusion criteria:
Articles published between January 2020 and March 2025.
Peer-reviewed articles listed in Scopus or Web of Science.
Publications written in English.
Studies describing or applying AI, IoT, Big Data systems in education.
B.
Exclusion criteria:
Articles outside the defined time frame (2020–2025).
Studies unrelated to our research questions.
Duplicate publications.
Conference papers or short communications lacking methodological clarity.
Review papers and purely theoretical contributions.
The selection process involved three stages:
Initial screening of titles and abstracts to remove irrelevant papers.
Full-text assessment by two independent reviewers.
Quality evaluation using a checklist examining clarity of objectives, methodological robustness, and coherence of findings.
Discrepancies between reviewers were resolved through discussion with a third author until consensus was reached. Only studies that met all quality criteria were retained. After de-duplication and screening, 24 studies satisfied the inclusion conditions. Metadata such as publication year, journal, study design, AI/IoT techniques, data sources, and key findings were extracted and summarized in Table 1. This dataset formed the empirical basis for the SWOT analysis conducted later in the study.
2.5 Data Extraction and Synthesis
Each selected article was read in full to extract detailed information on:
Research objectives and methodological approach.
Technologies and algorithms used.
Educational context and learner population.
Reported outcomes, limitations, and recommendations.
The extracted information was organized into a comparative synthesis matrix to facilitate cross-study analysis.
From this synthesis, four overarching dimensions were identified Strengths, Weaknesses, Opportunities, and Threats forming the conceptual basis of the SWOT framework presented in the next section.
Source | Publication Year | Article Title | Publication Source | References |
|---|---|---|---|---|
Journal Article | 2022 | An efficient framework for intelligent learning based on artificial intelligence and IOT | International journal of emerging technologies in learning | [12] |
Journal Article | 2024 | Greening smart learning environments with artificial intelligence of things | Internet of things | [13] |
Journal Article | 2022 | A study on mobile resources for language education of preschool children based on wireless network technology in artificial intelligence context | Computational and mathematical methods in medicine | [14] |
Journal Article | 2020 | Artificial intelligence based efficient smart learning framework for education platform | Inteligencia artificial-iberoamerical journal of artificial intelligence | [15] |
Journal Article | 2022 | Smart educational learning strategy with the internet of things in higher education system | International journal on artificial intelligence tools | [16] |
Journal Article | 2022 | Designing an extended smart classroom: an approach to game-based learning for IOT | Computer applications in engineering education | [17] |
Journal Article | 2023 | Generating an environmental awareness system for learning using IOT technology | Internet of things | [18] |
Journal Article | 2021 | A smart learning ecosystem design for delivering data-driven thinking in stem education | Smart learning environments | [19] |
Journal Article | 2021 | Investigating the impact of the internet of things in higher education environment | IEEE access | [20] |
Journal Article | 2023 | Empowering learning process in secondary education using pervasive technologies | Interactive learning environments | [21] |
Journal Article | 2023 | Smart education system to improve the learning system with cbr based recommendation system using IOT | Heliyon | [22] |
Journal Article | 2021 | Internet of things for education: a smart and secure system for schools monitoring and alerting | Computers & electrical engineering | [23] |
Journal Article | 2021 | A smart learning assistance tool for inclusive education | Journal of intelligent & fuzzy systems | [24] |
Journal Article | 2022 | A framework for designing applications to support knowledge construction on learning ecosystems | Interaction design and architectures | [25] |
Journal Article | 2023 | Towards intelligent e-learning systems | Education and information technologies | [26] |
Journal Article | 2023 | An evolutive knowledge base for ``askbot'' toward inclusive and smart learning-based NLP techniques | International journal of advanced computer science and applications | [27] |
Journal Article | 2023 | Personality-based tailored explainable recommendation for trustworthy smart learning system in the age of artificial intelligence | Smart learning environments | [28] |
Journal Article | 2023 | Developing a personalized e-learning and MOOC recommender system in IOT- enabled smart education | IEEE access | [29] |
Journal Article | 2024 | Learn with m.e.-let us boost personalized learning in k-12 math education! | Education sciences | [30] |
Journal Article | 2021 | Personalized smart learning recommendation system for arabic users in smart campus | International journal of web-based learning and teaching technologies | [31] |
Journal Article | 2025 | Developing a smart learning system for large enterprises based on intelligent augmented reality | Journal of organizational and end user computing | [32] |
Journal Article | 2024 | Smart evaluation: a new approach improving the assessment management process through cloud and IOT technologies | International journal of information and education technology | [33] |
Journal Article | 2021 | Study on learning analytics data collection model using edge computing | International journal of engineering trends and technology | [34] |
Journal Article | 2024 | Development of an intelligent learning evaluation system based on big data | Data and metadata | [35] |
In total, 24 studies were selected for inclusion (see Fig. 1). These contributions offer empirical insights, often through case studies or experimental data, on how AI, IoT, and Big Data technologies are applied in personalized education. The selection process was inspired by the PRISMA (Preferred Reporting Items for Systematic Reviews and Meta-Analyses) framework and aimed to build a high-quality and representative corpus.
Fig. 1
: PRISMA Flow Diagram
In the second phase, each selected publication was read in full to systematically extract key information, including research objectives, methodological approach, technologies used, study context or sample, main findings, identified limitations, and proposed recommendations. Based on this extracted data, a comparative synthesis table was constructed to facilitate cross-study analysis. This synthesis formed the basis for a SWOT analysis, aimed at identifying overarching themes and distilling the strengths, weaknesses, opportunities, and threats related to the integration of AI, IoT, and Big Data in personalized learning environments.
This SWOT analysis highlights that the joint integration of AI, IoT and Big Data represents a powerful strategic lever for Smart Learning, provided that the challenges of ethics, data governance, digital inclusion and social acceptability are met.
Figure 2 charts the annual output of selected studies, revealing an initial surge from a single article in 2020 to six in 2021, as the COVID-19 pandemic accelerated interest in remote and adaptive learning technologies. Growth continued with five publications in 2022, culminating in a peak of seven studies in 2023, likely driven by advances in real-time data analytics and widespread deployment of IoT sensors in educational settings. In 2024, the count dipped slightly to four articles, possibly reflecting the transition from exploratory prototypes to larger-scale deployments that often appear in conference proceedings rather than journals. Only one study was published by March 2025, suggesting a shift toward longitudinal evaluations and integrative reviews in the field.
Fig. 2
Annual distribution of selected published articles between 2020 and 2025
Our analysis of publication venues (Fig. 4) shows that two journals—Smart Learning Environments and IEEE Access, each published two of the 24 articles, indicating their central role in disseminating interdisciplinary research on intelligent learning technologies. The remaining articles are spread across 16 different journals, including Internet of Things, International Journal on Artificial Intelligence Tools and Computational and Mathematical Methods in Medicine, each contributing one study. This wide dispersion underscores the multifaceted nature of the domain, drawing contributions from educational technology, computer engineering, data science and applied AI communities. The breadth of journals also highlights the need for cross-disciplinary collaboration to address challenges such as data privacy, system interoperability and pedagogical integration.
A
Fig. 3
Number of articles per journal
3. Results of the Systematic Review
As part of this systematic review, a detailed SWOT analysis was performed to identify the strengths, weaknesses, opportunities, and threats associated with the combined use of Artificial Intelligence (AI), the Internet of Things (IoT), and Big Data in personalized learning environments. The analysis is based on 24 peer-reviewed studies published between 2020 and 2025 and indexed in Scopus and Web of Science.
The findings were organized according to the traditional SWOT framework, originally developed in strategic management studies, to facilitate a comparative and strategic understanding of the current research landscape.
Overall, the analysis reveals a distinct pattern: Strengths emerge from AI-driven personalization and real-time monitoring, while Weaknesses stem from practical constraints such as cost, data complexity, and insufficient teacher training. Opportunities highlight collaborative and scalable possibilities for future education, whereas Threats underscore the risks of privacy breaches, algorithmic bias, and unequal access to technology.
Fig. 4
SWOT Analysis of Personalized Learning Systems
Strengths: AI and IoT-based solutions enable a level of personalization and real-time adaptability that traditional learning methods cannot achieve. Several studies demonstrate how AI algorithms recommend tailored learning content, evaluate student performance automatically, and adjust pedagogical paths based on learner data [12][17]. IoT devices such as wearables and classroom sensors facilitate continuous monitoring of student attention, stress levels, and participation, thereby enhancing engagement and allowing timely teacher intervention [13][16][19]. Big Data dashboards and analytics offer educators intuitive visual summaries of student progress, providing valuable insights for decision-making and personalized feedback [18]. Collectively, these technologies strengthen educational efficiency and make learning processes more data-driven and student-centered.
Weaknesses: Despite these advantages, the integration of AI, IoT, and Big Data is hampered by several notable weaknesses. High infrastructure and maintenance costs represent a significant barrier to adoption, especially for resource-limited institutions [13][16]. The heterogeneity and volume of data collected from multiple sources create technical challenges in terms of storage, processing, and standardization [14]. In addition, many teachers lack the training and digital competence necessary to interpret and utilize these tools effectively [21]. Data quality issues such as noise, bias, or missing values further limit the reliability of AI-based recommendations and predictive models [24][26]. These limitations highlight the need for greater institutional investment in capacity building and technical support.
Opportunities: Despite existing barriers, the integration of AI, IoT, and Big Data offers substantial opportunities for innovation in education. Inter-institutional collaboration and data sharing can foster more robust predictive models and reduce implementation costs [14][18]. The emergence of cloud-based platforms and open-source algorithms enables scalable and modular deployment of smart learning systems [23][25]. Furthermore, AI-assisted content generation tools are transforming pedagogical design by allowing the rapid creation of adaptive learning materials and automatic feedback mechanisms [30]. Policy developments such as the General Data Protection Regulation (GDPR) can serve as a foundation for ethical governance, encouraging trust and responsible use of educational data among students, teachers, and institutions. These opportunities point toward a future where technology supports personalization without sacrificing equity and ethics.
Threats: On the other hand, the analysis identifies serious threats that could undermine these technological advances. Data privacy and security risks are among the most critical concerns, especially when biometric and behavioral data are collected through IoT devices [13][16][19]. The potential for algorithmic bias and lack of transparency can reinforce existing inequalities, particularly in systems that analyze socio-economic or linguistic factors [12][26]. Institutional resistance to technological change, combined with ethical apprehensions about constant monitoring, can hinder the adoption of AI-enabled learning systems [19][21]. Finally, unequal access to digital infrastructure between urban and rural areas may deepen the digital divide, limiting the equitable benefits of these technologies. Addressing these threats requires a proactive and ethical governance framework to ensure data protection, algorithmic accountability, and social acceptability.
3.1 Synthesis of Findings
The findings from the SWOT analysis indicate that the integration of AI, IoT, and Big Data in education is a double-edged sword. While these technologies can enhance personalization, efficiency, and inclusivity, they also pose risks related to privacy, ethics, and educational equity. Figure 4 and Table 2 (see original document) summarize the distribution of studies across years and journals and outline the key themes emerging from each approach. The steady increase in publications since 2020 reflects growing interest in AI-driven learning and its potential to transform pedagogy. However, the literature also emphasizes the importance of teacher training, cross-disciplinary collaboration, and clear data governance policies to translate technological innovation into sustainable educational practice.
3.2 SWOT Analysis Table
The following table provides a synthesized overview of the Strengths, Weaknesses, Opportunities, and Threats for each of the 24 selected approaches. Inspired by the traditional SWOT analysis framework, this presentation highlights the strategic dimensions of each solution and facilitates comparative evaluation across different implementations.
Ref | Approach | Strengths | Weaknesses | Opportunities | Threats |
|---|---|---|---|---|---|
[12] | AI Recommendation System | Fine content personalization; Partial automation of assessment | High infrastructure cost; Requires large, relevant training datasets | Scalability to other education levels; Partnerships with publishers | Socio-cultural bias risk; Dependency on AI vendors |
[13] | IoT Platform for Smart Classrooms | Real-time engagement tracking; Access to raw data (sensors, logs) | Complex hardware setup; Frequent and costly maintenance | Scalability to other classrooms/schools; Research program synergy | Sensor hacking; Access inequality for underfunded schools |
[14] | Big Data Predictive Analytics | Early detection of at-risk students; Institutional decision support | Heterogeneous data hard to standardize; Info overload for teachers | Platform sharing across schools; Use of open-source algorithms | Sensitive data leaks; Bias from unrepresentative datasets |
[15] | Conversational AI Chatbot | 24/7 support for FAQs; Personalized interaction with students | Variable quality; Can be overwhelmed during peak times | ML-based improvement; Multilingual support potential | Inappropriate responses; Loss of trust due to errors |
[16] | Class IoT & Biometric Sensors | Accurate stress and attention measurement; Real-time feedback | Perceived as intrusive; Requires strict ethical protocols | Learning environment optimization; Interdisciplinary collaboration | Excessive monitoring concerns; Real-time data vulnerability |
[17] | Adaptive AI for Assessment | Automatic correction of simple tasks; Reduced grading time | Hard to assess complex tasks; Model inertia if poorly trained | Multi-subject deployment; Longitudinal cohort comparison | Assessment bias; Limited teacher acceptance |
[18] | Big Data Dashboards | Intuitive progress visualization; Key performance indicators for teachers | Lack of interpretation skills; Risk of misleading metrics | Content adjustment based on metrics; Inter-school experience sharing | Overfocus on performance; Metric manipulation risks |
[19] | IoT Wearables in Class | Individualized tracking of movement/posture; Gamified engagement | Cost and hardware replacement; Noisy or inaccurate data | Physical + learning integration; Integration with school sports | Geolocation overexposure; Parental resistance |
[20] | AI-based Emotion Detection | Interest/boredom feedback; Real-time course adaptation | Risk of misinterpretation; Sensitive emotional content | Emotion-driven pedagogy; Improved group dynamics | Ethical risks (emotional profiling); Reduced classroom spontaneity |
[21] | Longitudinal Data Mining | Long-term learning path analysis; Success modeling | Expensive infrastructure; Complex historical data cleaning | Academic planning support; Education-career path correlation | Long-term data consistency issues; Risk of misuse for control |
[22] | Collaborative AI for Projects | Enhanced group work; AI provides collective suggestions | Bias in favor of certain projects; Requires tutor training | Interdisciplinary project expansion; Soft skills development | Over-reliance on AI; Traditional teacher disengagement |
[23] | IoT & Analytics in Immersive Environments | Virtual class with sensor tracking; Multimodal interaction | Connectivity dependency; High ecosystem cost | Immersive labs for science; Remote practical work collaboration | Network security risks; Unsuitability for large classes |
[24] | AI for Language Correction | Fast feedback on writing; Customization by language level | Limited to major languages; Hard to detect cultural nuances | Extension to more languages; Support for multilingual learners | Language standardization risks; Dialect bias |
[25] | Modular IoT Approach | Gradual sensor installation; Progressive data fusion | Complex setup; Multiple supplier coordination | Phased experimentation; Practice sharing among schools | Slow scalability; Equipment obsolescence |
[26] | Big Data: Socioeconomic Correlation | Fine-grained inequality identification; Support for policy decisions | Risk of stigmatization; Hard to anonymize all variables | Targeted remediation strategies; Collaboration with social services | Public distrust due to profiling; Public funding dependency |
[27] | Oral Assessment AI (NLP) | Automatic transcription and feedback on oral presentations | Speech recognition inaccuracies; Lacks context awareness | Support for shy learners; Expansion to MOOCs/distance learning | Linguistic bias; Limited use in oral-heavy subjects |
[28] | Outdoor IoT for Field Trips | On-site data collection (GPS, weather); Increased motivation | Logistical management; Maintenance costs | Geography/biology field projects; Citizenship education integration | Poor network coverage; Surveillance resistance |
[29] | Big Data & AI for Cheating Detection | Enhanced fraud detection; Automatic alerts for teachers | False positives; Needs constant dataset updates | Enhanced academic quality; Standardized validation protocols | Permanent surveillance fear; Tech circumvention |
[30] | Generative AI for Content Creation | Auto-creation of learning materials; Time saving | Possible factual errors; Requires human validation | Quick content personalization; Pedagogical design innovation | IP issues; Quality control concerns |
[31] | Collaborative IoT in Science Labs | Real-time experimental data sharing; Accurate tracking of manipulations | Team coordination complexity; Hardware reliability dependence | Research skills development; University ties | Sabotage risks; Sensor obsolescence |
[32] | Governmental Big Data Platform | Public policy based on indicators; National-level impact evaluation | Hard inter-region interoperability; Risk of program standardization | State-level resource sharing; Support for rural schools | Centralized control concerns; Local resistance |
[33] | Integration of smart technologies into e-learning platforms | Automation of evaluation processes and time-saving for instructors | Requires powerful servers and advanced technical resources; Risk of misclassification by automated systems | Expansion to other smart modules; Resource sharing among institutions via cloud infrastructure | Unequal access to stable internet connections; Increased dependency on digital infrastructure and cloud providers |
[34] | IoT in Hands-on Workshops | Detailed manual skills measurement; Precise feedback | Learner misunderstanding; Multi-sensor analysis complexity | Professional skill valorization; Trial/error learning approach | Indicator obsession; Technician shortage |
[35] | Global Adaptive Learning with AI & Big Data | Holistic approach (grades, logs, IoT interactions); High predictive intelligence | Costly and heavy configurations; Long calibration time | Longitudinal curriculum studies; Research-school collaboration | Learner overprofiling; Data/privacy concerns |
4. Discussion and synthesis
The SWOT analysis provides a multidimensional picture of how Artificial Intelligence (AI), the Internet of Things (IoT), and Big Data technologies are reshaping personalized learning. Three key themes emerge from the analysis: (1) the transformative power and ethical tensions of data-driven personalization, (2) the practical barriers that complicate implementation in real educational contexts, and (3) the strategic opportunities that can guide future integration.
4.1 Analysis of Key Strategic Dimensions
4.1.1 The Promise and Peril of Data-Driven Personalization
Across the reviewed studies, AI appears as the central enabler of personalized learning. Machine learning algorithms analyze students behaviour and performance data to deliver adaptive content, recommend resources, and provide instant feedback [12][17]. For instance, AI-based recommendation systems support individualized learning trajectories, while adaptive assessment tools automate grading and detect learning gaps in real time [15][17]. Conversational agents extend tutor availability beyond class hours and help students with specific questions or revisions [15].
IoT technologies add a complementary dimension by capturing real-time physical and behavioural signals through wearables, classroom sensors, and connected devices. This data, when combined with Big Data analytics, creates a feedback loop where teachers and students receive timely insights into engagement, motivation, and performance [13][19]. Yet these same mechanisms raise ethical questions concerning the extent to which continuous monitoring can intrude on students’ privacy and autonomy. A central tension thus emerges between the precision offered by data-driven systems and the human need for trust, freedom, and emotional space in learning environments [20].
Algorithmic bias represents another risk. If training datasets reflect existing inequalities or cultural biases, AI systems may inadvertently reproduce them, leading to unfair recommendations or assessments [24][26]. Addressing these concerns requires transparency in model development and continuous ethical oversight.
4.1.2 Implementation Barriers: Cost, Complexity, and Human Factors
While the technical potential of AI, IoT, and Big Data is clear, their adoption in educational practice faces multiple barriers. Financially, the installation and maintenance of IoT infrastructure remain cost-intensive [13][16]. Pedagogically, the lack of teacher training and institutional support limits effective use [21]. From a technical perspective, the integration of heterogeneous data from various devices and platforms poses challenges of standardization and interoperability [14].
Moreover, resistance to change is common in schools and universities. Teachers may fear being replaced by automated systems or feel uncomfortable with the monitoring aspects of IoT technologies [19][21]. Such perceptions can undermine the adoption of otherwise beneficial tools. To overcome this, training programs must emphasize how AI and IoT can augment rather than replace human teaching. Empowering educators to co-design digital solutions would increase their sense of ownership and trust.
Finally, the digital divide remains a systemic barrier. Institutions with limited resources often lack the necessary infrastructure to deploy advanced learning technologies. Without targeted policies to address these disparities, the benefits of smart learning will remain unequally distributed [18][22].
4.1.3 Strategic Opportunities and Future Directions
Despite the challenges, the integration of AI, IoT, and Big Data presents remarkable opportunities for educational innovation. Collaborative data-sharing initiatives between institutions could improve predictive accuracy and reduce implementation costs [14][18]. The increasing availability of open-source platforms and modular IoT solutions allows schools to experiment progressively with digital transformation [25]. At the same time, emerging AI applications in content generation and emotion-aware teaching hold promise for more context-sensitive and inclusive pedagogies [20][30].
Policy and governance frameworks are equally crucial. Legal standards such as the GDPR have begun to shape ethical boundaries around data use, but education-specific guidelines are still evolving. Establishing clear protocols for data ownership, consent, and algorithmic transparency will be essential to build public trust and long-term sustainability of these systems [32].
Looking forward, hybrid learning models that combine AI-driven analytics, IoT feedback, and Big Data insights offer the greatest promise for transformative education. However, this integration must be accompanied by a human-centered approach that values creativity, empathy, and critical thinking alongside technological efficiency.
4.1.4 Synthesis and Future Outlook
The collective evidence suggests that the integrated use of AI, IoT, and Big Data can significantly enhance personalization, engagement, and assessment in education. Advanced solutions such as Adaptive AI for Assessment [17], AI-based Emotion Detection [20], and Big Data: Socioeconomic Correlation [26] outline a comprehensive, data-driven model where instructional decisions are informed by real-time analytics and predictive indicators.
Nevertheless, this raises critical concerns. Ethically, it is essential to balance innovation with privacy and autonomy, avoiding intrusive surveillance or over-standardization. Professionally, the evolving role of educators must be addressed. Rather than reducing teachers to system operators, these technologies should empower them to focus on human-centered mentoring. Studies like those on Conversational AI Chatbot [15] and Longitudinal Data Mining [21] stress the irreplaceable value of interpersonal relationships.
On a systemic level, the emergence of interoperability standards [12][16][35] could accelerate the adoption of hybrid learning systems that combine e-learning platforms, IoT sensors, and data analytics. National platforms such as Governmental Big Data Platform [32] already explore centralized approaches to early risk detection and evidence-based policy design. If adequately governed, these systems could redefine educational management and promote interdisciplinary collaboration among technologists, educators, sociologists, and legal scholars.
To succeed, such integration demands a clear strategic vision. Interoperability standards would facilitate multi-approach deployment; cross-disciplinary collaboration would ensure alignment between innovation and equity; and public institutions must play a central role in funding infrastructure, research support, and data protection.
The SWOT synthesis highlights that, across varied implementations, AI-driven personalization, IoT-enabled real-time feedback and Big Data-powered analytics each bring distinct advantages, and also shared challenges around data privacy, system interoperability and user training. Moving forward, hybrid frameworks that leverage these three pillars in concert will deliver the greatest impact, provided institutions pair technological investment with robust risk-mitigation strategies and comprehensive professional development for educators.
5. Conclusion
The integration of Artificial Intelligence (AI), the Internet of Things (IoT), and Big Data marks a defining moment in the evolution of modern education. As evidenced by this systematic review of twenty-four peer-reviewed studies (2020–2025), these technologies have the potential to fundamentally reshape how learning is personalized, assessed, and managed. Through AI-driven analytics and IoT-enabled monitoring, educators can detect learning difficulties earlier, tailor interventions in real time, and promote more efficient use of educational resources. However, the review also reveals that this technological convergence introduces a new set of challenges. Ethical and privacy issues remain paramount, particularly when IoT devices capture physiological or behavioural data from students. Questions concerning data ownership, algorithmic fairness, and transparency must be addressed before these systems can be trusted and scaled responsibly. Moreover, the persistent lack of digital competence among teachers and unequal access to technological infrastructure threaten to widen, rather than bridge, existing educational divides.
To ensure that innovation leads to inclusion rather than inequality, a holistic vision of educational transformation is required one that aligns technological progress with social responsibility. Policymakers must develop robust regulatory frameworks for data governance; universities and training institutions should invest in continuous teacher education; and developers must design AI systems that respect ethical boundaries and cultural diversity. Only through such an integrated approach can the benefits of AI, IoT, and Big Data be equitably shared, ensuring that technology empowers rather than replaces the human dimensions of learning. Ultimately, the success of smart and personalized learning will depend not merely on the sophistication of algorithms or devices but on the collective commitment to build ethical, inclusive, and human-centered educational ecosystems. Future research should therefore focus on longitudinal assessments, cross-cultural comparisons, and real-world implementations that evaluate both academic outcomes and social impacts.
A
Author Contribution
AMK (Al Mahdi Khaddar) and YS (Youssef Said) conceptualized the main idea of the study, developed the methodological framework, and wrote the main text of the manuscript.AB (Abdellah Bakhouyi) supervised the research, validated the methodology, and contributed to the scientific revision of the content.AD (Amine Dehdi) collected and analyzed the data, prepared the tables and figures, and participated in the synthesis of the results.TC (Tarik Chafiq) provided critical revision of the manuscript and ensured compliance with publication standards.All authors read, commented on, and approved the final version of the manuscript.
6. References
A
Dehbi A, Bakhouyi A, Khaddar AM, Talea M (2025) Education and smart technologies: Towards a new pedagogical paradigm. Int J Evaluation Res Educ 14(1):297–309A
Rane N, Choudhary S, Rane J (2023) Education 4.0 and 5.0: Integrating artificial intelligence (AI) for personalized and adaptive learning. https://doi.org/10.61577/jaiar.2024.100006A
Dehbi A, Dehbi R, Bakhouyi A, Talea M (2023) Survey analysis of students and teachers’ perceptions of e-learning and m-learning in Morocco. iJIM 17(03):103. https://doi.org/10.3991/ijim.v17i03.36325A
Bakhouyi A, Dehbi A, Amhaimar L, Broumi S, Talea M, Khalidi A (2025) Improving SMART Learning: Course Completion via AI-Driven Hybrid System Integration in Big Data. Telematics Inf Rep 100199. https://doi.org/10.1016/j.teler.2025.100199Dehbi A, Bakhouyi A, Dehbi R, Talea M (2025) Towards Technology-Enhanced Learning: A Novel Machine Learning Approach in Education 4.0. Telematics Inf Rep 18:100194. https://doi.org/10.1016/j.teler.2025.100194
A
Berson IR, Berson MJ, Luo W (2025) Innovating responsibly: ethical considerations for AI in early childhood education. AI Brain Child 1(1):2. https://doi.org/10.1007/s44436-025-00003-5A
Dehbi A, Bakhouyi A, Dehbi R, Talea M (2024), December Towards a Smart Learning Environment: A Model for Smart Education using IoT and AI. In 2024 6th International Symposium on Advanced Electrical and Communication Technologies (ISAECT) (pp. 1–5). IEEE. https://doi.org/10.1109/ISAECT64333.2024.10799743A
Pati D, Lorusso LN (2018) How to write a systematic review of the literature. HERD: Health Environ Res Des J 11(1):15–30. https://doi.org/10.1177/1937586717747384A
Kitchenham B, Brereton P (2013) A systematic review of systematic review process research in software engineering. Inf Softw Technol 55(12):2049–2075. https://doi.org/10.1016/j.infsof.2013.07.010A
Khaddar AM, Dehdi A, Alali A, Bakhouyi A, Talea M (2025) Emerging Technologies in Learning: A Bibliometric Analysis of Technology Integration and Applications. Int J Interact Mob Technol 19(5). https://doi.org/10.3991/ijim.v19i05.51267A
DEHBI A, DEHBI R, BAKHOUYI, A., TALEA M (2022) Interoperability in smart Education: A systemic review based on bibliometric and content analysis methods. J Theoretical Appl Inform Technol 100(24):7211–7221A
Alanezi MA (2022) An efficient framework for intelligent learning based on artificial intelligence and IoT. Int J Emerg Technol Learn (IJET) 17(7):112–124. https://doi.org/10.3991/ijet.v17i07.27851A
Tabuenca B, Uche-Soria M, Greller W, Hernández-Leo D, Balcells-Falgueras P, Gloor P, Garbajosa J (2024) Greening smart learning environments with Artificial Intelligence of Things. Internet Things 25:101051. https://doi.org/10.1016/j.iot.2023.101051Li Q (2022) A study on mobile resources for language education of preschool children based on wireless network technology in artificial intelligence context. Comput Math Methods Med 2022(1):6206394. https://doi.org/10.1155/2022/6206394
A
Cao W, Wang Q, Sbeih A, Shibly FHA (2020) Artificial intelligence based efficient smart learning framework for education platform. Inteligencia Artif 23(66):112–123. https://doi.org/10.4114/intartf.vol20iss59pp123-127A
Wang J, Yu Z (2022) Smart educational learning strategy with the internet of things in higher education system. Int J Artif Intell Tools 31(05):2140101. https://doi.org/10.1142/S0218213021401011A
Petrović L, Stojanović D, Mitrović S, Barać D, Bogdanović Z (2022) Designing an extended smart classroom: An approach to game-based learning for IoT. Comput Appl Eng Educ 30(1):117–132. https://doi.org/10.1002/cae.22446A
Tabuenca B, Moreno-Sancho JL, Arquero-Gallego J, Greller W, Hernández-Leo D (2023) Generating an environmental awareness system for learning using IoT technology. Internet Things 22:100756. https://doi.org/10.1016/j.iot.2023.100756A
Benita F, Virupaksha D, Wilhelm E, Tunçer B (2021) A smart learning ecosystem design for delivering Data-driven Thinking in STEM education. Smart Learn Environ 8:1–20. https://doi.org/10.1186/s40561-021-00153-yA
Mircea M, Stoica M, Ghilic-Micu B (2021) Investigating the impact of the internet of things in higher education environment. IEEE Access 9:33396–33409. https://doi.org/10.1109/ACCESS.2021.3060964A
Stojanović D, Bogdanović Z, Petrović L, Mitrović S, Labus A (2023) Empowering learning process in secondary education using pervasive technologies. Interact Learn Environ 31(2):779–792. https://doi.org/10.1080/10494820.2020.1806886A
Veeramanickam MRM, Dabade MS, Borhade RR, Barekar SS, Navarro C, Roman-Concha U, Rodriguez C (2023) Smart education system to improve the learning system with CBR based recommendation system using IoT. Heliyon 9(7). https://doi.org/10.1016/j.heliyon.2023.e17863A
Qureshi KN, Naveed A, Kashif Y, Jeon G (2021) Internet of Things for education: A smart and secure system for schools monitoring and alerting. Comput Electr Eng 93:107275. https://doi.org/10.1016/j.compeleceng.2021.107275A
Srivastava S, Varshney A, Katyal S, Kaur R, Gaur V (2021) A smart learning assistance tool for inclusive education. J Intell Fuzzy Syst 40(6):11981–11994. https://doi.org/10.3233/JIFS-21007A
Silveira PDN, Cury D, de Menezes CS (2022) A framework for designing applications to support knowledge construction on learning ecosystems. IxD&A 52:158–178. https://doi.org/10.55612/s-5002-052-009A
Liu M, Yu D (2023) Towards intelligent E-learning systems. Educ Inform Technol 28(7):7845–7876. https://doi.org/10.1007/s10639-022-11479-6A
El Azhari K, Hilal I, Daoudi N, Ajhoun R, Belgas I (2023) An evolutive knowledge base for AskBot toward inclusive and smart learning-based NLP techniques. Int J Adv Comput Sci Appl 14(5). https://doi.org/10.14569/IJACSA.2023.0140544A
Takami K, Flanagan B, Dai Y, Ogata H (2023) Personality-based tailored explainable recommendation for trustworthy smart learning system in the age of artificial intelligence. Smart Learn Environ 10(1):65. https://doi.org/10.1186/s40561-023-00282-6A
Amin S, Uddin MI, Mashwani WK, Alarood AA, Alzahrani A, Alzahrani AO (2023) Developing a personalized E-learning and MOOC recommender system in IoT-enabled smart education. IEEE Access 11:136437–136455. https://doi.org/10.1109/ACCESS.2023.3336676A
Annuš N, Kmeť T (2024) Learn with ME—Let Us Boost Personalized Learning in K-12 Math Education! Educ Sci 14(7):773. https://doi.org/10.3390/educsci14070773A
Meddeb O, Maraoui M, Zrigui M (2021) Personalized smart learning recommendation system for Arabic users in smart campus. Int J Web-Based Learn Teach Technol (IJWLTT) 16(6):1–21. https://doi.org/10.4018/IJWLTT.20211101.oa9Wu HT (2025) Developing a Smart Learning System for Large Enterprises Based on Intelligent Augmented Reality. J Organizational End User Comput (JOEUC) 37(1):1–14. https://doi.org/10.4018/JOEUC.367726
Dehbi A, Bakhouyi A, Dehbi R, Talea M (2024) Smart Evaluation: A New Approach Improving the Assessment Management Process through Cloud and IoT Technologies. Int J Inform Educ Technol 14(1):107–118. https://doi.org/10.18178/ijiet.2024.14.1.2030
A
Lee M-S, Lee J-H Ju-Geon Pak. Study on Learning Analytics Data Collection Model using Edge Computing. Int J Eng Trends Technol, 69(4),142–145. https://doi.org/10.14445/22315381/IJETT-V69I4P221A
Ridhani D, Krismadinata, Novaliendry D,. A, Effendi H (2024) Development of An Intelligent Learning Evaluation System Based on Big Data. Data Metadata 3:569. https://doi.org/10.56294/dm2024.569