Copyright © 2020 for this paper by its authors. Use permitted under Creative Commons License Attribution 4.0 International (CC BY 4.0) Adaptive and Personalized e/m-Learning : Approaches and Techniques Ouissem Benmesbah Mahnane Lamia Mohamed Hafidi LRS Laboratory LRS Laboratory LRS Laboratory Badji Mokhtar University Badji Mokhtar University Badji Mokhtar University Annaba-Algeria Annaba-Algeria Annaba-Algeria ouissa2007@yahoo.fr mahnane_lamia@yahoo.fr mhafidi@yahoo.fr Abstract— Adaptive and personalized approaches within learning path. Finally, Section 4 presents the conclusion of e/m learning systems enable adapting learning Objects (LOs) the work. and process to the different needs and contexts, to help the learners in improving their knowledge or skills. In this paper, II. OVERVIEW OF THE LEARNING ADAPTATION we review the recent research on learning adaptation to pursue CLASSIFICATION two goals: First is to unify the classification of adaptation The first research question can be answered by presenting types; the second is to study the different approaches and a general classification of adaptation types used in the field techniques used to implement the learning adaptation in its two main types : adaptation of the LOs selection and adaptation of of e/m-learning. Several researchers have addressed the the LOs sequencing. adaptation type classification applied in the field of adaptive learning, but these classifications are slightly different. In Keywords—Adaptive e/m Learning, Learning path this section we will study them and propose a new adaptation, Adaptation by selection classification. I. INTRODUCTION Sampson [2] identifies three main categories of adaptation related to educational resources within adaptive With the rapid development of information technology in and personalized learning systems: -Selection Adaptation: education and learning field, researchers have created myriad This type of adaptation deals with selecting appropriate learning resources. It has been a difficult task for learners to learning objects LO based on different selection criteria find suitable learning resources from the Internet. Without derived from learners’ contextual elements. -Presentation effective adaptation, irrelevant resources will lead to Adaptation : considers that LOs is adaptively structured for learners’ cognitive overload and affect learning outcomes. access via mobile devices by taking into account parameters Therefore, learning systems need to be adapted to the related to the learners’ type of mobile device, the learner’s learners' context and needs. profile (including learner’s preferences and learning style), An adaptive learning environment provides personalized Parameters related with learner’s location, physical learning resources and processes to the learner through self- conditions and learner’s temporal information. -Sequencing directed study. An adaptive learning model can be Adaptation : This type of adaptation rearranges or reorders subdivided basically into a learner model, domain model, the navigation and sequencing possibilities of different LOs and adaptive engine. In such environment, the adaptive e/m that are linked to each other towards creating personalized learning system should adapt its services to a learner’s needs learning paths by taking into account different criteria and context. The purpose of adaptation is to optimize the derived from learners’ contextual elements (previous relationship between the learner context and learning knowledge, availability and current location, Time,…). content; hence, the learning outcome could be obtained with Premlatha defines two adaptation types: adaptive minimum time and interaction and could also increase the presentation at the content level and adaptive navigation learner satisfaction [1]. support at the link level [1]. Even though academic research on adaptive learning El jenati [3] adopts in his work three types of adaptation : environments has increased, the field lacks a comprehensive -Adaptive content: The adaptation of the content is based on literature analysis of the classification of the adaptation the selection of the adequate pedagogical content which take types, and the most used approaches and algorithms used to into account the learner’s context. Some learners may wish implement every type of adaptation. to get a simple version of the content and others may wish to This paper presents a study of learning adaptation in e/m get a detailed version. - Adaptive navigation: The Adaptive learning systems from 2008 up to 2019. It aims to specify: navigation allows to learners to find their paths by adapting the presentation of links to the objectives, knowledge and the Rq1- What should be adapted? preferences of the learner. -Adaptive presentation: The Rq2- How it should be adapted? adaptive presentation is to adapt the visual presentation to the preferences and needs of the learner. Some learner can easily The rest of this paper is structured as follows: Section 2 read the presented music score and will know how it sounds, describes the main types of learning adaptation. Section 3 but others will want an audio version. collects the most used approaches and algorithms used to implement the adaptation of the learning content and the The work presented in [4] focused on various e-learning problems, from these problems, we can extract a set of adaptation types: -Learning path generation (LPG): focused called a context model, this model must integrate information on providing a sequence of learning object materials to the such as the location of the learner, time, the environmental learners -Object recommendation (OR): Allows an adaptive characteristics (noise level, lighting level), device selection of LOs and -Personalization of content (POC): characteristics…etc [5-7-8-10-11-12-15] . specifies what learning objects are needed for the course established for a specific learner requiring a specific subject. Arriving to the second research question:“How the learning content should be adapted’. The answer to this Based on the works presented above, we can define two question consists to find the most applied approaches and main classes of adaptation: algorithms in the learning adaptation field.  Adaptation related to the content: It can be divided into As can be seen from Table 1, the majority of the works two sub categories: depends on Ontologies enriched with Semantic Web Rule Language (SWRL). Ontologies as a key and important - Adaptation of the content selection: Proposes a set of component of semantic web technologies are used to LOs adapted to the learner’s needs and contexts. represent knowledge about e-learning domain. SWRL is a -Adaptation of the content presentation: Different strong mechanism for inferring new relations and knowledge presentation forms of educational resources include [2]: which cannot be reached using ontologies [5-6-3-7-8-9-10- Changing the format for the same type of educational 11]. resource (e.g. wav files to mp3 files), Changing the type of Another category of works is based on ontology the educational resource (e.g. -Changing the dimensions of modeling of the context and the Leaning domain but it is not the educational resource (e.g. scaling down or scaling the based on rules but rather on Algorithms [12-13-14-15]: dimensions of the educational resource). Erazo-Garzón in [12] used semantic search (keywords)  Adaptation related to the learning path: It aims to find the and route algorithms applied to ontological models due to learners’ paths while learning in e-learning system by their expressiveness and extensible architecture, to determine adapting the learner preferences, learning styles, and with precision the concepts and semantic relations that exist other characteristics of an individual user. among academic and contextual information. In [13] and In this work, we have excluded the adaptation of the [14], Semantic learning objects search is proposed, it is based Content presentation, that’s because we are convinced that on the query expansion of the user query and uses the the adaptation of the format and the scale can be avoided and semantic similarity to retrieve semantic matched learning replaced by a selection adaptation which will include the objects. In [15], a novel context-aware mobile learning Content selection with the most appropriate format and scale application is proposed to encourage and promote Hadith according to the learner's learning style and / or learning, three dimensions (location, time and profile) of user characteristics of his mobile while being based on the density context are implemented, context-filtering based regular and high availability of resources pedagogical models in the expression and ontology matching-recommending techniques LO Repositories, which store several format of the same are used to match appropriate hadith on learner’s context. content. So in our study, we will focus only on two major The third category of works applied Evolutionary types of adaptation that are: adaptation by selection and computing algorithms (EC) to implement the adaptation by adaptation of the learning path. selection: In the following sections, we will move to the second Latha in [16] presents an evolutionary approach for research question which is "How pedagogical content should tuning the parameters required for personalizing the learning be adapted? And this, by studying the different approaches, content delivery. The compatibility level of the LO are tuned techniques or algorithms used to achieve each type of with respect to the learning style of the learner; the adaptation. complexity levels of the learner are tuned based on the III. OVERVIEW OF THE LEARNING ADAPTATION APPROCHES feedbacks from similar learners and the knowledge levels of the learners are tuned with respect to the complexity level of AND ALGORITHMS the learning objects. The interactivity levels of the learners A. Adaptation of LOs Selection: are tuned based on the behavior of the learners during the Table 1 presents a collection of works that propose an learning process. For that purpose Compatible Genetic adaptation by selection, these works can be divided into two Algorithm (CGA) is applied. categories: the first implements only the adaptation by Yang [17] in his work, proposed an Attributes-based Ant selection, which consists of selecting LOs adapted to the Colony Optimisation System (AACS) to help learners find needs and/or context of the user. The second category aims at an adaptive learning object more effectively by considering performing adaptation by selection as the first phase the relationship between learner attributes (e.g. learning followed by a second phase which is the adaptation of the style, domain knowledge) and LO’s attributes. For that LO s sequencing. In this second case, we are limited to study AACS algorithm is proposed, it is derived from an extension the technique or the algorithm used in the first phase of the of the Ant colony system that updates the trails’ pheromones adaptation. from different knowledge levels and different styles of a Among 16 works collected in Table 1, 9 works perform group’s learners to create a powerful and dynamic learning the adaptation by selection based on the learner model [6-3- object search mechanism. 9-13-14-16-17-18-19]. This model includes information such In Dwivedi [18], the author develops learning path as learner profile, learning needs or objective, learning style, recommendation framework by employing course knowledge level, ..., this means that these works have not generator’s advice and evolutionary approach namely a integrated the notion of the context which constitutes what is variable length genetic algorithm (VLGA) after generating [14] Semantic Ontology - Semantic Learning style, learner profiles through registration process. Algorithm search (Query expension teaching methods, learning activities Discrete Particle Swarm Optimization (PSO) was [15] Semantic Ontology -context- Location-Time- employed by Wang and Tsai [19] to choose the material Algorithm filtering based regular Profile suitable for a review course based on the material relevance expression + Ontology degree, difficulty level and the number of available learning matching resources. [16] Evolutionar GA - Compatible Learning style y Algorithm Genetic Algorithm Knowledge level Based on the related works analysis presented above, we (CGA) Feedback notice that the adaptation algorithms used to select a set of [17] Evolutionar ACO - attribute-based Learning style- LOs adapted to the profile/Context of the learner can be y Algorithm ant colony system Knowledge level classified into two main categories (see Fig. 1): (AACS) [18] Evolutionar GA - Variable lentgh Learning Style –  Rule-based techniques: y Algorithm genetic algo Knowledge level – Goal Ontology combined with SWRL adaptation rules is the [19] Evolutionar PSO for selecting LO Difficultly level, most used Rule-based technique. Ontology helps in y Algorithm relevance of material improving adaptive learning by providing a suitable vocabulary for learners to describe the course materials and represent the learner’s belief, expectation and context being used for the recommendation of LOs which are dependent on the domain ontology [4].  Algorithms-based techniques : - Semantic Algorithms: based on Ontology modeling and reasoning based on different types of algorithms - Evolutionary computing Algorithms that includes genetic algorithm (GA), Swarm optimization techniques (PSO and ACO techniques). Fig. 1. Approaches and techniques to implement Adaptation by Selection TABLE I. ALGORITHMS USED TO IMPLEMENT THE ADAPTATION BY LOS SELECTION B. Adaptation of LOs sequencing (Learning path Work Approach Technique Learner’s context dimensions adaptation): [05] Rule-based Ontology + SWRL Rules Network and Battery Curriculum sequencing, learning path adaptation, level, learner profile adaptive learning path generation and adaptive learning (Age,Knowledge level) schema generation, designate all the same purpose, which is the personalization and the adaptation of the learning [06] Rule-based Ontology + SWRL Rules Learning Style – material sequence called learning path. Providing an optimal Knowledge level- learning path tailoring to the context of the learners is a [03] Rule-based Ontology + SWRL Rules Learner profile crucial issue in online learning adaptation. An optimal Learning Style – learning path could reduce the student’s cognitive overload Knowledge level- Language and disorientation; consequently, this process would improve preferences) the learner’s learning outcome and efficiency of the adaptation in the online learning systems [5]. [07] Rule-based Ontology + SWRL Rules Location, time [08] Rule-based Ontology + SWRL Rules Location time This section presents a survey on learning path adaptation kewords knowledge efforts in the m/e learning environment from 2008 up to level device 2019. The survey highlights two points: [09] Rule-based Ontology + SWRL + LO Pre-requisite-  The different approaches to formulate the learning weighting Algo) Knowledge level- LO history- Social path problem. relation  The algorithms applied to solve the learning path [10] Rule-based Ontology + SWRL Rules Learning Style- adaptation problem. Location- TimeTechnology- 1) Approaches (Problem formulation): objective- mobility The main objective in learning path adaptation is to [11] Rule-based Ontology + SWRL rules Location, time, + Greedy algo device, profile minimize the path or route of individual learning. According to [18], to formulate the issue of learning path adaptation, [12] Semantic Ontology – Route Context model various approaches have emerged. Among these approaches: Algorithm algo+Semantic search (Location, time) (Keywords)  Constraints Satisfaction Problem (CSP): It’s a single [13] Semantic Ontology - Semantic LO Learner profile objective with several constraints. The problem in CSP Algorithm search (Query expansion is defined as the state of the variable definition. The and Semantic similarity solution space of the problem comprises all possible Algo) sequences and the objective function is to minimize or maximized the penalty function designed to evaluate evolutionary Optimizer for optimal search, is used for the sequencing [5-20]. Many works in the literature mathematical optimization problems in the field of Artificial proposed evolutionary algorithm, such as Genetic Intelligence. Given a large set of inputs and a good heuristic Algorithm (GA) [28-29], Particle Swarm Optimization function, it tries to find a sufficiently good solution to the (PSO) [34] or Ant Colony Optimization (ACO) [32] problem. This solution may not be the global optimal and scheduling and planning Problem [38]. maximum.  Multi-Objective Optimization Problem (MOOP): The According to [4], Meta-heuristic algorithms like concern in multi-objective optimization problem is to Evolutionary computation approaches (EC), have great satisfy multiple objectives simultaneously [5]. In impact in the solution of the learning path adaptation MOOP approach several algorithms are used, among problem by providing appropriate learning paths to learners. them GA[25-27], PSO[35], Heuristic algorithms [25- Genetic algorithms [25-27-28-29], Ant colony algorithms 26] and Planning and Scheduling Technique [39]. [30-31-32-33] and Particle swarm optimization [34-35] are widely used techniques in the construction of learning path  Domain modeling: This approach is implanted Through: sequence. directed graph, concept map and ontology [3-9-21-22- 11], however, there has been no formal model for Machine learning techniques are widely present in the discussing learning path problems based on Domain learning path adaptation field; Through the use of modeling []. Reinforcement learning (RL) and Bayesian Network (BN). RL is an area of Machine Learning. It is about taking suitable 2) Techniques: action to maximize reward in a particular situation. It is Once the learning path model has been formulated, the employed to find the best possible behavior or path it should methods to build the approach are chosen according to the take in a specific situation. In the field of adaptive learning, problem [5]. Table 2 presents the most used Algorithms in RL is used in [36]. The proposed approach consists of the the field of learning path adaptation. following steps. Firstly, the learner’s state is determined. Secondly, a learning material or path is suggested through a Reasoning based On ontology is present in several works set of actions. Thirdly, based on RL, the learner state is [3-9-21-22-11]. In such approach, reasoning techniques are updated, in addition, the rewards received by recommended usually applied on metadata derived from an ontology model learning paths or materials are updated. [8]. The reasoning is performed in terms of SWRL (Semantic Web Rule Language) rules that are applied on knowledge Bayesian Network BN (also known as Bayesian represented in the OWL-DL (Description Logics) ontology. probability theory) is also used for finding the adaptive Problem of concern with this approach is their learning path [37]. BN is a directed graph whose nodes inappropriateness to reasoning with uncertainty. It should be represent the uncertain variables of interest and edges are noted that some of the context elements are quantized with influential links between the variables. Node probability uncertainty leading to certain ambiguity while defining and table contains conditional probability (CP) values which are reasoning with context, [8]. This problem can be dealt with assigned on the basis of the level of expertise, learning style by integrating various reasoning models that may combine and learning pace of the learner. In the second step, BN is probabilistic, Fuzzy reasoning techniques [8]. For that, we constructed to calculate CP value for each knowledge unit in highlight in this survey other works, which propose a hybrid the learning path. Finally, the shortest path is selected to solution by combining Ontology-based reasoning with: provide appropriate learning path for the learner [4]. Fuzzy logic [9] and with a Greedy algo [11]. Planning& scheduling techniques [38-39-40], as an Case based reasoning technique CBR as another Artificial intelligence (AI) techniques, are also proposed to reasoning technique [23-24] is also used to implement the generate sequences of e-learning routes which are tailored to learning path adaptation problem. CBR is the process of the students’ profiles. solving new problems based on the solutions of similar past Employing Data mining (DM) in intelligent learning problems. CBR has both the capacity to represent knowledge systems has become a trend in developing learning systems, and to reason about it. However, CBR suffers from the which makes educational data mining the focus of a new and inexistence of genericity in knowledge representation; growing research community. Such a technique has the specific requirements for CBR are usually processed as they following strengths: (1) it reduces the constraints on the scale come. There are also some limitations like as knowledge of the database quality and the variable types; (2) it can acquisition problems for unavailable or limited cases, analyze both a continuous variable and discontinuous Inference efficiency is not always good as desired, straight variables efficiently; and (3) its results with graphical or rule forward provision of explanation is missing [5]. expressions can be understood easily and can be explained. Since Learning path problem is NP-Hard problem, Lin in [41] suggested that learning materials based on the heuristic and meta-heuristics are usually used to approximate tree mechanism can meet individual requirements and can its solutions. Heuristic search optimization algorithms are enhance learning efficiency in a learning environment. used to implement a solution for the learning path adaptation; Other algorithms derived from the graph theory approach among these algorithms we have Greedy algorithm and Hill are used for the same purpose, among these algorithms we Clumbing algorithm. Greedy algorithm [25, 20, 11] is an can mention: the first-search depth (DFS) [42], binary integer algorithmic paradigm that builds up a solution piece by programming and Adaptive Shortest path algorithm [44] piece, always choosing the next piece that offers the most (See Fig. 2). obvious and immediate benefit. So the problem where choosing locally optimal also leads to global solution are best fit for Greedy. Hill Climbing algo [26], which is another [27] Evolutionary GA Context model (Learner Computing motivation) [25] Evolutionary GA Knolwledge level, Computing [29] Evolutionary GA Knowledge level, Computing Concept, learning objective, Time [36] Machine RL Context model learning (environmental context, social, cognitive, Feedback) [37] Machine BN preferences learning [38] Planning & Case based Learner model Scheduling planning Fig. 2. Approaches and techniques to implement Learning Path Adaptation [39] Planning & Planning Learner model Scheduling technique + ontology + TABLE II. ALGORITHMS USED TO IMPLEMENT THE LEARNING PATH ADAPTATION Intelligent agents (Hybrid Work Approach Technique Learner’s context technique) dimensions [40] Planning & AI Planning and Knowledge level, [3] Ontology Ontology + Learner context Scheduling scheduling Metadata LO, learning SWRL Rules technique style, learning objective [9] Ontology Ontology + Location time kewords Time SWRL Rules+ knowledge level device [44] Graph based Adaptive Learning objective, Fuzzy Logic Shortest path Learning background, [21] Ontology Ontology + Knowledge level- algo Preferences SWRL Rules Prerequisites -satisfaction Prior knowledge, [22] Ontology Ontology + Objective, personality learning style SWRL Rules type, Academic state [42] Graph based the depth First- Previous knowledge, search (DFS) time restriction [11] Ontology Ontology + Context model (Location, Rules + Time, device, Greedy environment) algorithm IV. SUMMARY AND CONCLUSION [24] CBR CBR + Not Mentioned Ontology for Learning adaptation and personalization is an important indexing cases research field in e/m-learning environment. It is quite [23] Context model (Location, necessary to discover the most efficient approach to realize CBR CBR + Distance heuclédienne time, technology) it. This paper presents a literature review of personalized and adaptive learning algorithms from the two sides of [25] Heuristic based Greedy Prequisites and approach competencies adaptation: the adaptation of the LOs selection and the Algorithm for shortest path adaptation of the learning path or the LOs sequencing. [20] Heuristic based Greedy Learner preferences, approach Algorithm previous knowledge, Through the statistical analysis of the current availibility of time individualized learning algorithms, the different approaches that are applied to construct them vary between semantic [26] Heuristic based hill-climbing Learner model approach algorithms based on the ontology modeling of the domain heuristic learning path) model and the learner’s context model and evolutionary [30] Evolutionary Shortest path Context (Mobility, computing techniques that includes genetic algorithms and Computing Ant Colony Luminosity level, noise swarm optimization techniques. An additional set of Optimization level algorithms are used for the purpose of the learning path (ACO) Leraning style, knowlege adaptation, in this category we can find machine learning- algorithms level) Evolutionary based algorithms like BN and RL, graph-based algorithms [31] ACO Learner Computing profile(Knowledge level) and planning and scheduling techniques. [32] Evolutionary ACO Learner model (Cognitive In future work we will focus on the adaptation within m- Computing level s, LO complexity Learning environment, which offer adapted learning level metadata) services in mobility, according to the nature of this kind of [33] Evolutionary ACO from data Knowledge level – learning systems, we need to integrate on the one hand more Computing mining based prerequisites frequent pattern learners contextual data like, location, time, mobility state, graph model device characteristics, environment characteristics…and in the other hand information that characterizes the learning [34] Evolutionary PSO Learning objective, Computing previous knowledge task like learner’s Learning style, knowledge level, preferences,… As adaptation type, we are interested in the [35] Evolutionary PSO Learning objective, well-known learning path adaptation problem which plays a Computing learner knowledge level, central role in intelligent learning systems and it is time for learning, weights of concepts considered as one of the most challenging problems. Since this problem is seen as a combinatorial optimization problem, we are going to study the effects of the application and Internet Technology (pp. 126-136). Springer, Berlin, Heidelberg,2013. of computational evolutionary algorithms, which is still a [21] Y.L. Chi, T.Y. Chen and W.T. Tsai, “Creating individualized learning hot research field. paths for self-regulated online learners: An ontology-driven approach.”, In International Conference on Cross-Cultural Design REFERENCES (pp. 546-555). Springer, Cham, 2014. [1] K.R. Premlatha,T.V Geetha, “Learning content design and learner [22] O. Iatrellis, A. Kameas and P. Fitsilis, “EDUC8 ontology: semantic adaptation for adaptive e-learning environment: a survey”, Artificial modeling of multi-facet learning pathways. Education and Intelligence Review, 44(4), pp.443-465, 2015. Information Technologies, pp.1-20, 2019. [2] D. G. Sampson, P. Zervas, “Context-aware adaptive and personalized [23] H. Supic, “Case-based Reasoning Model for Personalized Learning mobile learning systems. In Ubiquitous and mobile learning in the Path Recommendation in Example-based Learning Activities”. In digital age”, pp. 3-17. Springer, New York, NY, 2013. 2018 IEEE 27th International Conference on Enabling Technologies: [3] K. Jetinai, “Rule-based reasoning for resource recommendation in Infrastructure for Collaborative Enterprises (WETICE) (pp. 175-178). personalized e-learning”, In 2018 International Conference on IEEE, 2018. Information and Computer Technologies (ICICT), pp. 150-154, IEEE, [24] D. Mansouri, A. Mille and A. Hamdi-Cherif, “Adaptive delivery of 2018. trainings using ontologies and case-based reasoning”, Arabian Journal [4] A. Khamparia and B. Pandey, B., “Knowledge and intelligent for Science and Engineering, 39(3), pp.1849-1861, 2014. computing methods in e-learning”, International Journal of [25] G. Durand, N. Belacel and F. LaPlante, “Graph theory based model technology enhanced learning, 7(3), pp.221-242, 2015. for learning path recommendation”, Information Sciences, 251, pp.10- [5] A. Muhammad, Q. Zhou, G. Beydoun,D. Xu and J. Shen, “Learning 21, 2013. path adaptation in online learning systems”, In 2016 IEEE 20th [26] V. Tam, E.Y. Lam and S.T. Fung, “A new framework of concept International Conference on Computer Supported Cooperative Work clustering and learning path optimization to develop the next- in Design (CSCWD), pp. 421-426. IEEE., 2016. generation e-learning systems”, journal of computers in education, [6] S. Ouf, M.A. Ellatif, S.E. Salama, and Y. Helmy, “A proposed 1(4), pp.335-352, 2014. paradigm for smart learning environment based on semantic web”, [27] N.C. Benabdellah, M. Gharbi and M. Bellafkih, “Units’ Computers in Human Behavior, 72, pp.796-818, 2016. Categorization Model: The Adapted Genetic Algorithm for a [7] B. Zhang, C. Yin, B. David, Z. Xiong and W. Niu, “Facilitating Personalized E-Content”, In Europe and MENA Cooperation professionals' work-based learning with context-aware mobile Advances in Information and Communication Technologies (pp. 149- system”, Science of Computer Programming, 129, pp.3-19, 2016. 158). Springer, Cham, 2017.. [8] R. Benlamri and X. Zhang, “Context-aware recommender for mobile [28] T.Y. Chang and Y.R. Ke, “A personalized e-course composition learners”, Human-centric Computing and Information Sciences, 4(1), based on a genetic algorithm with forcing legality in an adaptive p.12, 2014. learning system. Journal of Network and Computer Applications, 36(1), pp.533-542, 2013. [9] B. Bouihi M. Bahaj, “Ontology and Rule-Based Recommender System for E-learning Applications”, International Journal of [29] I. El Guabassi, M. Al Achhab, I. Jellouli and B.E.E. Mohajir, Emerging Technologies in Learning, 14(15), 2019. “Personalized Ubiquitous Learning via an Adaptive Engine.”, [10] M. Abech, C.A. Costa, J.L. Barbosa, S.J. Rigo and R. Rosa Righi, “A International Journal of Emerging Technologies in Learning (iJET), 13(12), pp.177-190, 2018. model for learning objects adaptation in light of mobile and context- aware computing”, Personal and Ubiquitous Computing, 20(2), [30] N. Benabdellah, M. Gharbi and M. Bellafkih, “Toward E-Content pp.167-184, 2016. Adaptation: Units’ Sequence and Adapted Ant Colony Algorithm.,” [11] F. Soualah-Alila, C. Nicolle and F. Mendes, “Context-Aware Information, 6(3), pp.564-575, 2015. Adaptive System For M-Learning Personalization”, 2014. [31] A.P. Dharshini, S. Chandrakumarmangalam and G. Arthi, “Ant colony optimization for competency based learning objects [12] L. Erazo-Garzón, A. Patiño, P. Cedillo and A. Bermeo, “ CALMS: A sequencing in e-learning,” Applied Mathematics and Computation, Context-Aware Learning Mobile System Based on Ontologies”, In 263, pp.332-341, 2015. 2019 Sixth International Conference on eDemocracy & eGovernment (ICEDEG) , pp. 84-91. IEEE, 2019. [32] S. Sengupta, S. Sahu and R. Dasgupta, “Construction of learning path using ant colony optimization from a frequent pattern graph,”, arXiv [13] K.M. Fouad, M.A. Nofal, H.M. Harb and N.M. Nagdy, “Using preprint arXiv:1201.3976, 2012. semantic web to support advanced web-based environment”, 2011. [33] L. de Marcos, J.J. Martínez and J.A. Gutiérrez,” Swarm intelligence [14] E. Kurilovas, S. Kubilinskiene and V. Dagiene, “Web 3.0–Based in e-learning: a learning object sequencing agent based on personalisation of learning objects in virtual learning environments”, competencies”, In Proceedings of the 10th annual conference on Computers in Human Behavior, 30, pp.654-662, 2014. Genetic and evolutionary computation (pp. 17-24). ACM, 2008. [15] AZ. Sevkli, L. Motiwalla and H.F. Abdulkarem, “The design and [34] C.P. Chu, Y.C. Chang and C.C. Tsai, “PC 2 PSO: personalized e- implementation of a context-aware mobile hadith learning system”, course composition based on Particle Swarm Optimization”, Applied International Journal of Mobile Learning and Organisation, 11(4), Intelligence, 34(1), pp.141-154, 2011. pp.295-313, 2017. [16] B.C.L. Christudas, E. Kirubakaran and P.R.J. Thangaiah, “An [35] D. Shawky and A. Badawi, “February. A reinforcement learning- evolutionary approach for personalization of content delivery in e- based adaptive learning system”, In International Conference on learning systems based on learner behavior forcing compatibility of Advanced Machine Learning Technologies and Applications (pp. learning materials”, Telematics and Informatics, 35(3), pp.520-533, 221-231). Springer, Cham, 2018. 2018. [36] N.V. Anh, N.V., Ha and H.S. Dam, “Constructing a Bayesian belief [17] Y.J. Yang and C. Wu, “An attribute-based ant colony system for network to generate learning path in adaptive hypermedia system” adaptive learning object recommendation”. Expert Systems with Journal of Computer Science and Cybernetics, 24(1), pp.12-19, 2008. Applications, 36(2), pp.3034-3047, 2009. [37] A. Garrido, L. Morales and I. Serina, “On the use of case-based [18] P. Dwivedi, V. Kant and K.K. Bharadwaj, “Learning path planning for e-learning personalization”, Expert Systems with recommendation based on modified variable length genetic Applications, 60, pp.1-15, 2016. algorithm”, Education and Information Technologies, 23(2), pp.819- [38] B. Fernández-Reuter, E. Durán and A. Amandi, “Designing a hybrid 836, 2018. method for personalized ubiquitous learning paths generation”, [19] T.I. Wang and K.H. Tsai, “Interactive and dynamic review course In 2017 36th International Conference of the Chilean Computer Science Society (SCCC) (pp. 1-9). IEEE, 2017. composition system utilizing contextual semantic expansion and discrete particle swarm optimization”, Expert Systems with [39] A. Garrido and E. Onaindia, “Assembling learning objects for Applications, 36(6), pp.9663-9673, 2009. personalized learning: An AI planning perspective”, IEEE Intelligent [20] P. Basu, S. Bhattacharya and S. Roy, “February. Online Systems, 28(2), pp.64-73, 2011. recommendation of learning path for an e-learner under virtual university”, In International Conference on Distributed Computing [40] C.F. Lin, Y.C. Yeh, Y.H. Hung and R.I. Chang, “Data mining for providing a personalized learning path in creativity: An application of decision trees”, Computers & Education, 68, pp.199-210,2013. [41] A.H. Nabizadeh, A. Mário Jorge and J. Paulo Leal, “July. Rutico: Recommending successful learning paths under time constraints”, In Adjunct Publication of the 25th Conference on User Modeling, Adaptation and Personalization (pp. 153-158). ACM, 2017. [42] N. Belacel, G. Durand and F. Laplante, “A Binary Integer Programming Model for Global Optimization of Learning Path Discovery”. In EDM (Workshops), 2014. [43] I.A. Alshalabi, S.E. Hamada, K. Elleithy, I. Badara and S. Moslehpour, “Automated Adaptive Mobile Learning System using Shortest Path Algorithm and Learning Style”, International Journal of Interactive Mobile Technologies, 12(5), 2018.