With advances in artificial intelligence (AI), adaptive hypermedia research, and the growth of big data, personalized training is getting a big boost with the rise of adaptive learning. This new avenue aims to generate, for every learner, in real time, the learning path most likely to enable him/her to achieve every learning objective. There are many spotlights on this approach that brings forward many opportunities!
Personalized learning has taken several forms over time: private lessons – whose first traces date back to the 19th century, computerized adaptive tests and differentiated pedagogy. Therefore, it is not recently that we recognize the benefits of a customized approach compared to a teaching method that addressed everybody in an undifferentiated way. In the 1980s, Benjamin Bloom – to whom we owe the well-known “Bloom Taxonomy” pedagogical model – had noticed that learners who received tutoring (alone with a tutor) were more positive and had better results in exams, while studying, in addition, over shorter periods (Bloom, 1984).
The “first generation” of personalized training focused on what the learner had to learn. Today’s adaptive learning is more about helping the learner achieve his or her goals by focusing on areas they master less. In its most modern version, this second-generation personalized learning is inseparably linked to computer technologies. It resorts to these, not only to transmit the contents and evaluate the learning outcomes but also to adapt to the learner according to his/her objectives. It can take into account the learner’s knowledge, preferences, as well as his/her performance and how he/she interacts with the system. It should be noted that this customized training is suitable for a wide range of learning situations, in both academic and professional settings.
… to adaptive and intelligent systems
Before going any further, it should be mentioned that the terminology surrounding adaptive learning can be confusing. This is because this approach is based on principles that have been used for a long time, but whose distribution channels have technologically multiplied and refined in recent years. Therefore, those who today evoke the concept of adaptive learning do not always speak of the same reality. To avoid getting lost, but above all to follow the version with the most promising future, let us focus on the one that is embodied in “adaptive and intelligent web-based learning systems”, such as define Peter Brusilovsky, a pioneer in the field of adaptive learning, and Christoph Peylo in Adaptative and Intelligent Web-based Educational Systems.
These systems include “adaptive web-based learning systems” and “intelligent web-based learning systems.” In the first case, these are systems that “attempt to be different for different students and groups of students by taking into account information accumulated in the individual or group student models.” In the second case, we are talking about systems that “apply techniques from the field of Artificial Intelligence (AI).”
By “adaptive” and “intelligent” technologies, Brusilovsky and Peylo essentially refer to “different ways to add adaptive or intelligent functionality to an educational system.” They also note that several systems fall into both categories, being both adaptive and intelligent. Added to this, according to the authors, the line between a system called “intelligent” or “non-intelligent” is sometimes thin… In the light of these observations and considering that in the near future the AI should articulate all these types of personalized learning, we will follow their development and will address them as “intelligent adaptive learning.”
The components of intelligent adaptive learning
Let’s take a closer look at the conditions that led to the emergence of intelligent adaptive learning and that are still necessary for its development.
Nourished with big data and algorithms
Big data is the information we all produce using new digital technologies. In adaptive learning, this data comes from a variety of sources: from the learner, who generates it on the training platform; other learners who follow the same training (but not the same path); and any other relevant source on the web or not that can feed and guide the training.
As for the algorithms, they are, in a way, the instructions that a system must follow to reach a specific result or to solve a problem. These adaptive learning systems can be integrated into various platforms that generate this data, whether it is online learning platforms, augmented or virtual reality, etc.
Powered by Web 2.0 and hypermedia
The rise of Web 2.0 has contributed to the development of adaptive learning by expanding its possibilities for interaction and personalization, including online training. The web allows it to include a much wider variety of technologies derived from artificial intelligence.
The development of hypermedia paved the way for research on adaptive learning. Hypermedia is an approach in presenting information that allows the activation of links between textual, sound and visual elements (texts, videos, graphics, etc.). Hypermedia is based on models (components) that take into account user preferences. After classical linear hypermedia, where all users were offered the same links, we developed the adaptive hypermedia, which offers each user links according to some of their characteristics. Used in the field of learning, it is called “adaptive educational hypermedia.” The latest of these systems is the dynamic adaptive hypermedia, an adaptive hypermedia whose link construction is not predefined but is done as the execution progresses.
Articulated by Artificial Intelligence (AI)
It is artificial intelligence, especially deep learning, that can handle big data. Deep learning is that branch of AI that lets the computer find for itself the best way to solve a problem from data and indications about the expected outcome.
In the field of education, this technology is embodied mainly in intelligent tutorial systems (ITS), which are computer programs that are part of the expert systems. ITS can decode learners’ mental trajectories from the way in which they solve problems, in order to give them real-time explanations, advice, and exercises that will benefit them the most. It is not only the content of a course that can “intelligently” adapt to each learner, but also its presentation and navigation.
Structured by model components
The architecture of an intelligent, adaptive learning system includes components called “models,” which make it possible to structure the operation of a set of computer programs.
- Learner model: it personalizes the learning taking into account the particularities of the learner
- Pedagogical (or learning) model: it determines how information should be taught
- Expert model: it represents what needs to be taught
- Model of the interface (or environment): it constitutes the communication layer (the interactions) between the learner and the system
A super learning
Being able to follow a personalized learning path, in real time, has several benefits, including the following:
Benefits for the learner
- Increases motivation
By being exposed to content that fits his/her profile and targets the knowledge he/she lacks, while using an optimal pedagogical approach, the learner is more likely to perceive the training as relevant and challenging. The weaker and the stronger are less likely to drop out, as it’s the case in the form of learning that is aimed at the “average” learner. Motivation is an essential factor to complete and succeed in training.
- Optimizes learning time
The learner will not waste his/her time on notions that he/she already masters and will be able to devote all his/her energy to the appropriate content. This is without counting that once the learner has assimilated new knowledge, the system will detect it and adjust the rest of the content accordingly.
- Reinforces the learning process
The fact that the content as well as the presentation and the navigation are adjusted to the individual’s profile, and backed up by the findings of the research in neuroscience, maximizes the chances of learner’s success in each phase of the learning (assimilation, consolidation, and application).
The benefits for the company
- Determines training needs
An intelligent adaptive learning system provides valuable information that, when used appropriately, can enable the company to identify what knowledge and skills it would benefit from developing among its employees.
- Used to evaluate the impact of its investments in training
Once again, thanks to the data it collects on the learning path of its employees, this system allows the company to know if the efforts and money invested in training report as desired, or if it must adjust its aim.
- Consolidate a culture of development
Since this training system is at the cutting edge of technology, allowing employees to acquire new skills, and without wasting their valuable time, it has everything it needs to seduce and encourage adherence to a culture of continuous learning. Moreover, such a system reinforces the image of an innovative employer that gives its team members the means to make the best of themselves.
Looking towards the future
Who does not want to follow the “right” educational path? This approach gives learners maximum chances of achieving their learning goals by reducing wasted time… and increasing motivation. Also, which company does not want to offer its employees innovative and tailor-made training, which, also, provides it with valuable information to better guide its training initiatives. While some aspects need to be refined and not yet widely available, intelligent adaptive learning is likely to be one of the most promising approaches to deliver a broad range of learning to all learners, whatever their profile and goals!
The intelligent adaptive
To tailor training to a learner, or group of learners, an intelligent, adaptive learning system performs the following functions:
It covers these 3 phases of the learning process:
- Knowledge assimilation
- Knowledge consolidation
- Knowledge application
It includes 5 processes:
- Identification of the learner’s profile
- Adaptation of training according to learner’s performance
- Gradual refinement of the profile
- Recurrent evaluations
- State of acquired skills
It can adapt to the training on 3 levels:
Based on neuroscience
Like any serious learning approach, intelligent adaptive learning is built around proven pedagogical principles. For the learner to assimilate, consolidate and then apply his/her new knowledge, adaptive learning builds on the discoveries of cognitive sciences, particularly those in the neurosciences, which revealed us how the brain learns and which teaching methods should be used.
Neuroscience researchers rely on sophisticated techniques such as functional magnetic resonance imaging (fMRI), eye tracking systems and electroencephalographs.