The Role of Learner Characteristics in the Adaptive Educational Hypermedia Systems: The Case of the MATHEMA

Published Online October 2017 in MECS DOI: 10.5815/ijmecs.2017.10.07

The rapid development of the Internet and the Web over recent years has led to an increasing interest in creating Web-based learning tools and learning environments, such as the Adaptive Educational Hypermedia Systems (AEHSs).

Hypermedia consists of different media and integration, such as text, graphic, animation, audio, etc. not only have various media and their integration greatly enriched the learning environment, but also the production of multimedia teaching material. Learners and hypermedia systems can freely realize man-machine interaction [25].

The term adaptation in e-Learning systems involves the selection and manner of presentation of each learning activity as a function that examines the entity of knowledge, skills and other information given by each subject taught [71].

AEHSs combine ideas from hypermedia and intelligent tutoring systems (ITS) in order to produce applications whose the content, link structures, and other features are dynamically adapted to the learner’s characteristics, such as the learning goal, current task, knowledge level, performance, background-experience, interests, preferences, stereotypes, cognitive preferences, learning or cognitive style, personal data, abilities/disabilities, social group, working environment, demographic data, and situation variables. All these characteristics stored in the learner’s model, play an important role both in the development and the functionality of the AEHSs.

According to [11], AEHSs could be considered as the solution to the problems of "traditional" online educational hypermedia systems, due mainly to their static content, the "lost in hypermedia" syndrome and the "one-size-fits-all" approach. Furthermore, AEHSs, increase the functionality of the conventional hypermedia by combining free browsing with personalization and can support all the continuum of the learning model, from the pure system-controlled to the fully learner-controlled.

In the Web-based AEHSs, several adaptive and intelligent techniques have been applied to introduce adaptation, such as [11]:

• (a)    Curriculum Sequencing : It helps the learner to follow an optimal path through the learning material.

• (b)    Adaptive Presentation : It adapts the content presented in each hypermedia node according to specific characteristics of the learner.

• (c)    Adaptive Navigation Support : It adapts the link structure in such a way to guide the learner towards interesting and relevant information, kept away from non-relevant information either by suggesting the most

relevant links follow or by providing adaptive comments with visible links.

• (d)    Meta-adaptive Navigation Support : It selects or suggests the most appropriate adaptive navigation technique that suits the given learner best relative to the given context, either by observing and evaluating the success of each technique in different contexts and the resulting learning from these observations, or by assisting the learner in selecting the navigation technique that best suits to him/her.

• (e)    Interactive Problem-Solving Support : It provides the learner with intelligent help on every step of problem-solving, by giving a hint to executing the next step for the learner.

• (f)    Intelligent Analysis of Learner's Solutions : It uses intelligent analysers that not only tell the learner, whether the solution is correct, but also it tells him or her what exactly is wrong or incomplete.

• (g)    Example-based Problem-Solving Support : It helps the learners in solving new problems, not by articulating their errors, but by suggesting them relevant successful problem-solving cases, chosen from their earlier experience.

• (h)    Adaptive Collaboration Support; Adaptive Group Formation and/or Peer Help : These techniques support the collaboration process either just like the interactive problem-solving support systems help an individual learner in solving a problem, or they use knowledge about possible collaborating peers to form a matching group relative to the kind of the collaborative task.

According to [11], the adaptive navigation support is now a popular research area while the adaptive presentation is the most popular and the most studied method of hypermedia adaptation.

Two significant terms using in the AEHSs are the adaptability and adaptivity. According to [6], the adaptability refers to the capacity of adaptive learning systems to automatically adapt the learning process to the specific requirements and preferences of a particular learner. According to [62], the adaptivity could be defined as the capability of an adaptive learning system to alter its behaviour according to learner needs and other characteristics.

First generation systems provided limited adaptability through stereotype-based user models and limited functionality adaptation techniques, such as direct guidance, stretch text, hiding, and primitive link annotation. An example of such a system is the ISIS-Tutor. Second generation systems have improved upon the functionality of first generation systems, introduced new capabilities, such as adaptive multimedia presentation, map adaptation, and link sorting. Examples of such systems are the ELM-ART, AHA!, and CS383. Third generation adaptive hypermedia systems support multidimensional user models that improve their functionality and enhance the support offered to learners. INSPIRE and TANGOW are examples of such systems.

In this paper, the aim is to investigate the AEHSs developed so far in order to find out which of the learner characteristics used by them for various kinds of adaptation to compare with the characteristics that the MATHEMA uses for various kinds of its adaptation support so that to draw conclusions about the relationship of student characteristics with adaptive techniques (rationale for selecting students' characteristics for adaptation).

The rest of the paper is organized as follows: In Section II, we present related works on the learner characteristics used for adaptation in AEHSs. In Section III, we refer to the MATHEMA and the learner characteristics that it uses in its adaptation techniques. In Section IV, we present the evaluation of the MATHEMA about its functions. In Section V, we summarize the most significant points of our work and we refer to our future plans.

• II.    Related Works

The bibliographic method used in this research is the state-of-the-art review about AEHSs and specifically for the learner’s characteristics used for their adaptation techniques. For this aim, we made an extended research to draw conclusions about the selection of a concrete learner's characteristic for a concrete adaptation technique.

Educators know that individual learner characteristics play a huge role in how fast and in how well overall learning occurs. According to [11], an AEHS is based upon the assumption that each learner has different characteristics and that different educational settings could be more suitable for one type of learner than for another. Whenever, course content could be provided in a flexible way, adapted to individual learners’ characteristics through the e-learning system, the system can deliver the course content so that it capitalizes on the learner’s characteristics to optimize the learning outcome.

According to [2], an adaptive e-learning system based on the learner knowledge and learning style has a higher level of perceived usability than a non-adaptive elearning system. As usability influencing on the learner’s satisfaction, engagement, and motivation when using elearning systems, learning enhancement expected when the system is highly usable.

According to [69], the learning process in an actual AEHS environment is complex and influenced by many characteristics of the learner. It is therefore important to consider accommodating as many of these characteristics as possible into the learner model to generate an exact adaptation. However, many learner characteristics have been identified in the literature; it is, therefore, important to select for use in the learner model only those characteristics that directly influence learner achievement in the specific learning process, otherwise, the design of the learner model will become unnecessarily complex. Thus, the [69] suggest that the learner characteristics being considered in an AEHS are knowledge level, learning styles, experience, background, and preference.

Other characteristics, such as age, sex/gender, race/ethnicity, demographic data, interests, etc. are not taken into account as they are considered much less influencing the learner’s achievement.

According to [24], an important issue in AEHSs is to investigate the characteristics of the learner, to specify on which ones the educational process should be adapted. The goals of the learner, but also their background [11], knowledge level [28], experience and learning style [69] perceived as characteristics that differentiate the users of a system and considered very important on the influence they may have on the learning process. According to [73], a profile could be considered complete when it incorporates the users’ perceptual preference characteristics that mostly deal with the intrinsic parameters.

A research of [49] showed that learning systems’ adaptation is highly successful when one or more of the following learner characteristics adapted: learning styles, cognitive styles, background knowledge, preferences (for particular types of learning materials), and motivation.

Also, research of [3] indicated that the adaptation based on the combination of the information perception, learning style, and knowledge level yields a much better learning outcome (both in the short- and long-term) and learner satisfaction than adaptation based on either of these learner characteristics alone; this combination is also marked by a much higher level of perceived usability compared to a non-adaptive version of the elearning system.

A research of [22] on learners with differing knowledge and motivation indicated that low prior knowledge learners benefited more than high prior knowledge learners. A research of [60] indicated that the learners with more favourable characteristics (i.e., higher prior knowledge, more complex epistemological beliefs, more positive attitudes towards mathematics, better cognitive and meta-cognitive strategy use) tended to show a more adaptive example utilization behaviour, reported less cognitive load, and solved more problems correctly than learners with less favourable characteristics.

According to [17], many studies have found that learners with different levels of prior knowledge benefit differently in hypermedia learning systems, with experts and novices showing different preferences to the use of hypermedia learning systems and requiring different levels of navigational support.

According to [26], in Web-based educational systems that consider either only learning styles or only cognitive traits, the relationship leads to more information. This additional information can be used to provide better adaptivity, for combined learning styles and cognitive traits instead of only for one of them. In systems that incorporate learning styles as well as cognitive traits, the interaction can be used to improve the detection process of the counterpart.