Intelligent e-Learning with New Web Technologies

Technology innovations supporting Web 4.0 and Web 5.0 influence the quality of student learning and performance in blended and distant learning. Development of web technology extends the capabilities and varieties of e-learning. The Covid-19 pandemic drastically accelerated the changes in education and transformed the mode of teaching and learning. This paper analyzes the changes that have occured in e-learning following the evolution of World Wide Web. It also focuses on Web 3.0, Web 4.0 and the trends in Web 5.0 and Web 6.0 to outline the new features of e-learning. The symbiotic relationship between Artificial Intelligence (AI), Big Data, Linked Data, Cloud Computing, and Data-Driven Science is also focused on. "The Semantic Web" connecting Web's data performs context- based search and research. The Internet of Things lets Web-connected machines of all kinds to communicate with each other and with us, creating a rich flow of data about their location and status. In the near future Web 5.0 and Web 6.0 probably will completely alter our way of communication and acquiring knowledge. The simulating models ofvirtual and augmented reality with the power of AI will reshape all spheres of our social life, for good or bad.

towards the "Intelligent Web". According to Harshal and Hayatnarkar (2007) in Web 3.5 there are fully pervasive services based on matured and embraced semantic technologies, to be upgraded to the next level of sophistication. Latest technologies of Web 3.0 are included in the Web 3.5 and will be fully matured within Web 4.0.Each new period in the evolution of the World Wide Web has transformed the way business is conducted and companies themselves (Kambil, 2008). Main characteristics of web 4.0 important for e-learning are analyzed by (Nedeva&Dineva, 2012).
The Web 4.0 (The Intelligent Web) -services will be autonomous, proactive, and content exploring, self-learning, collaborative, and content-generating agents based on fully matured semantic and reasoning technologies as well as AI. Examples might be services interacting with sensors and implants, natural language services, or virtual reality services. Web 4.0 is known as symbiotic web, meaning interaction between humans and machines in symbiosis (Hemnath, 2010).Web 4.0 is called also "Mobile Web",a space of interconnected web pages, web apps, videos, photos, and interactive content (http://www.evolutionoftheweb.com/). However, until now, there is no exact idea about web 4.0 and its technologies, but the web is moving toward using artificial intelligence to become as an intelligent web (Aghaei et.al, 2012).Actually, Web 4.0 connects all devices in the real and virtual world in real-time (https://flatworldbusiness. wordpress.com/flat-education/previously/web-1-0-vs-web-2-0-vs-web-3-0-a-bird-eye-on-thedefinition/).Web 4.0 based on wireless communication occupies the fourth step in the evolutionary process. For example, the GPS that guides cars and now helps drivers to improve the planned route or save fuel will shortly save them from having to handle it. This 4.0 or mobile version is ready to take off, with an apparently remote Web 5.0, the ''sensitive'' Web, hard on its heels (Kambil, 2008). Web 5.0, the sensory and emotive Web, is designed to develop computers that interact with human beings. Although at the moment the Web is ''emotionally'' neutral, that is, it does not perceive what users feel and although emotions are still difficult to map. In order to manage in this new scenario (although onlypotential as yet), teachers will have to master new competences and skills (Benito-Osorio, et al., 2013).Currently the Web is "emotionally" neutral: do not feel the user perceives. The company Emotive Systems has created, neuro technology through headphones that allow users to interact with content that meets their emotions or change in real time facial expression an "avatar" (Patel, 2013).

E-learning evolution
E-learning is an open system that blends access to information and purposeful communication into a dynamic and intellectually challenging learning community. E-learning fully integrate the benefits of personal freedom with connectivity (Garrison&Randy, 2011). Table 1presents new set of combined Semantic Web and E-learning characteristics for a holistic 3.0 E-learning model (Issa et al., 2015). New characteristic of e-learning is adaptability offered by e-learning systems. Adaptability refers to the idea that technology-enhanced learning environments can automatically adapt needs and preferences of the learner. Such systems are usually referred to as Adaptive Educational Hypermedia Systems (Brusilovsky, 2001) or when accessible on the Web-Adaptive Web-based Educational Systems (Brusilovsky&Peylo, 2003).Adaptive educational systems use a model of individual user's characteristics -user model. (Brusilovsky, 2001). The user model is a representation of information about an individual user that is essential for an adaptive system to provide the adaptation effect, i.e. to behave differently for different users (Triantafillou et al., 2003).
Adaptive educational systems are able to perform several adaptive procedures. They are able to perform adaptive presentation; the system adapts the content according to the user's model; provide adaptive navigation support (Brusilovsky&Millan, 2007).Many types of intelligent learning systems are available, but five key components are common in most systems namely:the student model, the expert model, the pedagogical module, the domain knowledge module, and the communication model (Ma, 2006).
Recent development in computer-based educational systems resulting in a new generation of system encompassing intelligence, to increase their effectiveness; they are called Intelligent Educational Systems(IESs). Intelligent Tutoring Systems (ITSs) constitute a popular type of Intelligent Educational Systems (Brusilovsky, 1999). Adaptive Educational Hypermedia System (Brusilovsky,1998) are another type of educational system, specifically developed for hypertext environments such as WWW. Some authorssuggest two type of intelligent and adaptive Web tutors, called Intelligent Web-Enabled Tutors (Ma,2006). The first tutor supported students to think critically and suggest hypotheses, observations, and data while solving a case. The second tutor customized Web content based on student learning needs. The intelligent agent are able to initiate actions in order to achieve goals, thus being able to behave autonomously in an environment. They are several type of agents: Cognitive Agents, Reflective Agent, Pedagogical Agent, Expert Agent, and Communication Agent.
Cognitive Agent in REAL are software entities that carry out some set of operations on behalf of a User in a learning environment with some degree of autonomy, and in so doing, employ some knowledge or representation of the users it represents. Reflective Agent represents the mental states of users. Build in Design Mode by the user, this agent combines the knowledge representations for the entity agents with specific rules to guide its actions. Like the entity agents, Reflective agent behaviors are constructed in the form of promotional networks, production rules and mental images. Instructional designers design the strategies in Pedagogical Agent. They may learn appropriate pedagogical practices for REAL applications in the design team as well as though pilot studies with some students. An Expert Agent exhibits mastery of knowledge in a domain and can perform a task in an optimized way. Discrepancies between the behaviors of the reflective and expert agents can be seen as missed concepts or misconceptions in student understanding and may lead to interventions within the game to help the user overcome his or her conceptual difficulties. A Communication Agent acts as a collaborator that facilitates user-computer interaction. It can be designed using Microsoft Agent, utilizing configurable featuressuch as speech, tonality, gesture, facial expression, gender, and screen navigation that can be combined together to emulate a human face-to-face communication act (Bai&Black, 2010).
Summarizing analyzed literature, we can review that the Intelligence and Adaptivity are essential features of e-learning and its Sub-Characteristics are: Adaptive presentation; Adaptive navigation; Stimulating critically thinking and task solving; Intelligent analyzing and interactive student help.
The Internet links more than 10 billion pages, creating an opportunity to adapt millions of instructional resources for individual learners by Web technologies. Three components drive this educational inflection point. They are Artificial Intelligence (AI), cognitive science, and the Internet(Woolf, Beverly Park, 2010).AI techniques contribute to self-improving tutors, in which tutors evaluate their own teaching (Issa&Isaas,2015). Most of AI's success so far has been primarily in 'restricted' domains where rules, settings and objectives are well defined, e.g. chess.
In more open-ended domains such as education, the success of AI has been limited. This limitation primarily comes from the fact that open-ended domains are inherently more complex and therefore an AI system needs to contain many parameters, which in turn require many data for estimation and as a result require significant amounts of computational power (Rubens et al., 2014).Crucial components needed for the AI to succeed in more general open-ended domains starting to fall in place. There is a vast amount of data of available; importantly many of this data is "open" to a wide audience (Big Data). No matter how vast the dataset is it tends to provide a limited view on the problem. New technologies are allowing to establish links between these datasets as to obtain a more complete picture (Linked Data). The significant infrastructure needed to store and intelligently process this data is now becoming easily accessible and affordable (Cloud Computing). The new scientific framework is becoming available for supporting AI in the process of scientific discovery (Data-driven Science) (Rubens et al., 2014).
AI -Big data.Web data contains a precious resource -intelligence and is therefore often referred to as "Web Intelligence" (Zhong et al., 2000). This intelligence needs to be extracted and utilized, and AI is a perfect tool for accomplishing this objective. We consider that the role of Web 2.0 was to enable data production, and the role of Web 3.0 will be to enable utilization of this data.
AI -Linked data. However, we along with many others (Marshall&Shipman, 2003)believe that semantic linking is overly ambitious and is yet hard to achieve on the wide and general scale due to inherent ambiguity of natural language. However, this does not mean that the data could not be linked and utilized. In order to widen the linking objectives the concept of "Linked Data" has been recently developed (Fischetti, 2010).There has been a number of success of using AI to produce the needed links that even captures some of the semantics e.g. folksonomy (Halpin et al., 2007).
AI -Cloud computing.Processing and analyzing large quantities of data requires significant computational resources as well as frameworks to make these resources easily accessible. A variety of competitively priced cloud computing services are becoming available, e.g. Amazon's AWS, Google's App Engine, Microsoft's Azure to name a few. In addition, a number of supporting frameworks has been developed that made the power of computational clouds easily accessible, e.g. a widely adopted Hadoop/MapReduce, and a more specialized ones such as Mahout, Hive, Pig, Oozie, and Rhipe (Rubens et al., 2014).
AI -Data driven science.Recently large number of datasets have become available for little or no cost (Rubens et al., 2011). Data-driven science is starting to gain a foothold in the education, as indicated by the rapid development and increasing applications in the new areas of educational data mining (EDM) (Baker&Yacef,2009),and learning analytics (Siemens, 2010).
Analyzed characteristics of e-learning related to the development of Web technologies can be presented in summary form in the following table (Table 2):