Dr. Penteado,

Notwithstanding that there can be many different interpretations and opinions, I think the major difference between ITSs and AHSs can be captured based on the models they are based on, or operate with. There seems to be a growing agreement, even a general consensus, amongst researchers that intelligent tutoring systems operationalize four basic models, namely: (i) application domain model, (ii) learner model (iii) active tutoring model, and (iv) user interface model. At the same time, in addition, adaptive educational hypermedia systems also involve: (v) embedding environment model, (vi) learning context model, and (vii) media integration model. In the practice AHSs typically provide more immersive environment for learning, but target a lower level of intelligence on the side of the system. ITSs originally pursued a lower level of intervention from human educators and providing immediate and customized instruction or feedback to learners. ITSs are more influenced by the rapid change of AI and smart agent technologies, while AHSs are more influenced by the emerging VR, game and communication technologies. Originally, ITSs were focused on learning contents, learning paths, learner interactions, automated assessments and feedbacks, while AHSs on rich personal experiences, adaptive stimulation, and monitoring engagement levels. My impression is that, based on the current functionalities, the scopes and objectives of these systems cannot be sharply demarcated. Nowadays, there is a kind of convergence, and as a consequence both of them target emergent collective behaviors, social interactions, measurable role of emotions in learning, and service oriented computing. They follow both the applicative approaches and the experimental approaches.

Kind regards, I.H.

Adaptive E-learning Environments, supports individuals through content adaptation, and learning path adaptation.  Different Adaptive e-learning system can use a variety of different learning strategies and technologies to predict and recommend the preferred learning material (Ghali & Cristea 2009) . This can be achieved by recommending and adapting the appearance of hyperlinks or simply by recommending actions and resources.  Depending on what adaptive system you will look at, some might only focus on learner characteristics traits (Yarandi et al.,  2013).  There is also Adaptive Intelligent Retrieval (AIR) & Adaptive Educational Hypermedia 

E-learning environments (Adaptive Systems) enable learners to identify existing educational paths that pervious students have taken before, giving them guidance. Adaptive Systems try to anticipate the needs and desires of the user.

Intelligent Tutor Systems (ITS)

Intelligent Tutor Systems is used within learning to adapt, not just to the pedagogical activities but also the individual (student/learner) needs. So, the (ITS) try to identify some student’s characteristics through (User Modelling)  obtaining and extracting features that can suggest certain activities, as well as know how to react to some real time exercises.  (ITS) depending on model of choice can focus on the Pedagogical Layer, Learner-Centricity, Learner-Support, Tutor Support, etc.. There are many different literature supporting layers in which ITS can be used or where it has been used,

Ghali et al., (2009)  introduced an adaptive Web 2.0 e-learning tool called My Online Teacher (MOT), which was developed to support a variety of features and facilities like: collaborative authoring; group-learning-paradigm; social annotation (group rating, feedback, etc..);

References:

Ghali, F., & Cristea, I. A., (2009). MOT 2.0: A Case Study on the Usefuleness of Social Modeling for Personalised E-Learning Systems. In The 14th Int. Conference of Artificial Intelligence in Education (AIED‘09) (2009), IOS Press

Yarandi, M., Jahankhani, H., & Addel-Rahman H. T., (2013). A Personalised Adaptive E-learning approach based on Semantic web technology WEbology, Volume 10, Number 2, December 2013

Martins, A. C., Faria, L., Vaz de Carvalho, C., & Carrapatoso, E. (2008). User Modeling in Adaptive Hypermedia Educational Systems. Educational Technology & Society, 11 (1), 194-207.

Martins, C., Faria, L., & Carrapatoso, E., (2008)*. Constructivist Approach for an Educational Adaptive Hypermedia Tool, The 8th IEEE International Conference on Advanced Learning Technologies, Santander, Spain, IEEE Computer Society

Klašnja-Milićević, A., Vesin, B., Ivanović, M., & Budimac, Z., (2010). Integration of Recommendations and Adaptive Hypermedia into Java Tutoring System [Accessed July 2014] [On-line] Available:http://www.doiserbia.nb.rs/img/doi/1820-0214/2010%20OnLine-First/1820-02141000021K.pdf

Cristea, I A., & Ghali, F., (2010). Towards Adaptation in E-Learning 2.0 [Accessed, July 2014] [On-line] Available: http://www.dcs.warwick.ac.uk/~acristea/Journals/NRHM10/NRHM-aic-2010.pdf

Costello, R., & Mundy, D.P., (2009). The Adaptive Intelligent Personalised Learning Environment, icalt, pp.606-610, 2009 Ninth IEEE International Conference on Advanced Learning Technologies, 2009

Canavan, J., (2004)., Personalised E-learning Through Learning Style Aware Adaptive Systems [Accessed, July 2014] [On-line] Available: http://www.scss.tcd.ie/publications/tech-reports/reports.05/TCD-CS-2005-08.pdf

Keles, A., Ocak, R., Keles, A., & Gulcu, A., (2010). ZOSMAT: Web-based intelligent tutoring system for teaching–learning process Expert Systems with Applications, Volume 36, Issue 2, Part 1, March 2009, Pages 1229-1239

AEHS is a educational hypermedia system that adapts to the learner (knowledge, goals, traits, etc.) usually by guiding the learner through link adaptation or adapting content presentation. It should have learning content represented in hypertext/hypermedia form to operate (although nearly anything now qualifies in the Web age).

Now, the difference with ITS depends on how you teat it. In a narrow sense, an ITS is an AI-based system that supports tutoring, i.e., assisting a learner in solving a problem either helping on any step of solution, or examining partial/full solutions. ITS identify problems, offer hints and help. and support learning by doing. In this strict sense, ITS and AEHS are basically two smart technologies that supports two kinds of learning, both being important. While most reported ITS/AEHS has only one of these components, there are a number of educational systems that have both components - like ELM-ART.

Nowadays, by ITS people frequently mean any educational system that uses AI (the correct term is IES, intelligent educational system). In this sense, most AEHS would be a kind of IES since most are based on AI technologies for knowledge representation and knowledge based algorithms of navigation support and presentation. Yet, there is a good share of AEHS, especially among early work that do not really use AI techniques, so can't be justly called ITS/IES. So, in this broader sense, the difference is between "adaptive" and "intelligent" systems. A truly adaptive system should adapt to the user using whole information it has, not just react to the last answer/click/step. An intelligent system should use AI for its work. A good share of IES/ITS systems are both adaptive and intelligent (good adaptation needs intelligence and adaptation is one of the natural benefit to add when using AI), but, as I said, there are a number of simple adaptive systems that would be hard to claim intelligent. And quite a number IES and ITS are not adaptive, although the react smartly to each user action, this reaction is defined by the last action and doesn't depend on the broader history.

The following two papers could help to clear that difference:

Weber, G. and Brusilovsky, P. (2001) ELM-ART: An adaptive versatile system for Web-based instruction. International Journal of Artificial Intelligence in Education 12 (4), 351-384.

Brusilovsky, P. (2001) Adaptive hypermedia. User Modeling and User Adapted Interaction 11 (1/2), 87-110.

Brusilovsky, P. and Peylo, C. (2003) Adaptive and intelligent Web-based educational systems. International Journal of Artificial Intelligence in Education 13 (2-4), 159-172.

Thanks everyone for the answers!

But, by standard definition, both have an UI model, a Domain Model (like a knowledge map, with concepts and relations), a Student Model (overlaid in knowledge space, with rules to estimate proficiency). AI is used at least in knowledge representation and proficiency estimation.

However, I see a difference in the Tutor Model, both being considered as tutoring practices:

• In ITS, it works by simply offering tips, following the map path or ensuring a mastery learning approach, blocking from moving on.
• In AEHS, it offers supplementary material (content presentation), the next topic to learn (curriculum sequencing), in a given order (link ordering) 

I see Khan Academy as an example. It looks like an ITS (what topic to work next, tips for problem resolution) and an AEHS (the next modules are chosen based on the previous proficiency along the knowledge map).

Another difficult example is ASSISTment: I see no clear AI involved, but it is regularly classified as an ITS in the literature.

By the time of these papers, both did not exist.

So, I see 2 differences: