CHAPTER ONE
Introduction
1.1 Background to the study
The growing number and sophistication of smart phones and tablet PCs are radically changing the way we access the almost near infinite volume of information on the World Wide Web. Today almost every mobile device has the capability to access the internet while on the go. The social networking media make it possible for news, events and information to be disseminated in real time thus, indeed making the world a global village. On the other hand, while universities are still grappling with providing adequate infrastructure to provide quality education to their students, an unprecedented amount of lessons are being accessed on daily basis from Youtube, and other online university sites with mobile devices. It is unfortunate that traditional schools and university systems are still struggling to leverage the many opportunities for innovation in this area. One of these instant areas of learning gaining ground is Mobile learning (or commonly called M-learning). Mobile learning is gradually gaining recognition that UNESCO recently began to take it seriously and is now organizing workshops from university to university in order to fashion out a framework for its adaptation.
Learning is an active process of building knowledge and skills through practice within a supportive community. It comprises not only a process of continual personal development and enrichment, but also the possibility of rapid and radical conceptual change [1]. A first step in postulating the theory of M-learning is to distinguish what is special about mobile learning compared to other types of learning activities. An obvious, yet essential, difference is that it starts from the assumption that learners are continually on the move. We learn across space as we take ideas and learning resources gained in one location and apply or develop them in another. We learn across time, by revisiting knowledge that was gained earlier in a different context, and more broadly, through ideas and strategies gained in earlier years providing a framework for a life time of learning. We move from one to another, managing a range of personal learning projects, rather than following a single curriculum. We also move in and out of engagement with technology, for example, as you enter and leave cell phone coverage.
The need to re-conceptualize learning in the mobile age has given rise to intensive research work on Mobile learning. In recent times, this need has also propelled researchers to recognize the essential role of mobility and communication in the process of learning. That is to say, the role communication and interaction play in the learning process is a critical success factor. Within this context M-learning can contribute to the overall quality and accessibility of learning.
Unlike the traditional teaching and learning environments, in which learners follow a fixed sequence to instructional resources, such as textbooks in classroom settings, M-learning as an extension of E-learning has the potential to make learning even widely available and more accessible than we are used to in the existing E-learning environments. While mobile devices are approaching ubiquity today, the M-learning industry is still in its infancy [2]. M-learning has a number of common deficiencies, such as slow access to course materials, courseware not being able to adapt to individual students, real time interaction between student and system being difficult to achieve because of connection unreliability and bandwidth limitations. In order to understand this new concept called M-learning, the following subsections provide some background information.
1.2 M-learning vs E-learning
Since the past decade we have come to embrace the concept of e-learning, now M-learning is emerging. Is there any difference between electronic learning (E-learning) and mobile learning (M-learning) [3]?
E-learning covers a wide variety of applications and processes including computer-based learning, Web-based learning, virtual classrooms and digital collaboration [4]. E-learning is defined as the delivery of content [and interaction] via all electronic media, including the internet, intranets, extranets, satellite broadcast, audio/video tape, interactive TV, and CD-ROM. E-learning can also be defined more narrowly as distance learning, which would include text-based learning and courses conducted via written correspondence.’
M-learning is a subset of E-learning. E-learning is the macro concept that includes online and mobile learning environments. In this regard the following simple definition is very useful: ‘M-learning is E-learning through mobile computational devices: Palms, Windows CE machines, even a digital cell phone’ [5]. Figure 1.1 shows the subsets of flexible learning.
Figure 1.1: The subsets of flexible learning (courtesy, T.H Brown)
1.3 Agents
The past decade has experienced a significant interest in agent-oriented technology and a distinct trend has evolved to the research and development work on intelligent agents. This trend relates to the diversification in the types of agents being investigated and most popular types include user interface agents, information agents, multi agents system, mobile agents, and so on. Mobile intelligent agents have the potential to address some of the deficiencies inherent in mobile learning. This research work also demonstrates how intelligent mobile agents can handle the problems that limit the potential M-learning environment development.
An agent is anything that can be viewed as perceiving (percepts) its environment through sensors and acting (action) upon that environment through actuators [6]. A robot agent might have cameras and infra red range finders for sensors and various motors for actuators. A software agent (softbot or soft robot) receives keystrokes from the keyboard, file contents and network packets as sensory inputs and acts on the environment by displaying on the screen, writing files, and sending network packets.
The past decade has experienced a significant interest in agent-oriented technology and a distinct trend has evolved to the research and development work on intelligent agents. This trend relates to the diversification in the types of agents being investigated and most popular types include user interface agents, information agents, multi agents system, mobile agents, and so on. Mobile intelligent agents have the potential to address some of the deficiencies inherent in mobile learning. This research work also demonstrates how intelligent mobile agents can handle the problems that limit the potential M-learning environment development.
An agent is anything that can be viewed as perceiving (percepts) its environment through sensors and acting (action) upon that environment through actuators [6]. A robot agent might have cameras and infra red range finders for sensors and various motors for actuators. A software agent (softbot or soft robot) receives keystrokes from the keyboard, file contents and network packets as sensory inputs and acts on the environment by displaying on the screen, writing files, and sending network packets.
1.4 The structure of agents
The job of artificial intelligence (AI) is to design the agent program that implements the agent function, mapping percepts to actions. We assume this program will run on some sort of computing device with physical sensors and actuators-we call this the architecture:
agent = architecture + program .
The architecture might be just an ordinary PC, or it might be a robotic car with several onboard computers, cameras, and other sensors. In general, the architecture makes the percepts from the sensors available to the program, runs the program, and feeds the program’s action choices to the actuators as they are generated. The difference between the agent program and the function is that, the agent program takes the current percept as input, while the agent function takes the entire percept history. The agent program takes just the current percept as input because nothing more is available from the environment; if the agent’s actions depend on the entire percept sequence, the agent will have to remember the percepts.
Russell and Norvig [6] stated that Alan Turing proposed the method to build learning machines and then to teach them. In many areas of AI, this is now the preferred method for creating state-of-the-art systems. Learning has another advantage, it allows the agent: to operate in initially unknown environments and to become more competent than its initial knowledge alone might allow.
1.5 Components of a Learning Agent
A learning agent can be divided into four conceptual components, as shown in the figure 1.2 below [6].
The most important distinction is between the learning element, which is responsible for making improvements, and the performance element, which is responsible for selecting external actions. The performance element is what has previously been considered to be the entire agent: it takes in percepts and decides on actions. The learning element uses feedback from the critic on how the agent is doing and determines how the performance element should be modified to do better in the future. The design of the learning element depends very much on the design of the performance element.
Figure 1.2: A general model of learning agents [6].
When trying to design an agent that learns a certain capability, the first question is not “How am I going to get it to learn this?” but “what kind of performance element will my agent need to do this once it has learned how?’ [6]. Given an agent design, learning mechanisms can be constructed to improve every part of the agent. The critic tells the learning element how well the agent is doing with respect to a fixed performance standard. The critic is necessary because the percepts themselves provide no indication of the agent’s success. For example, a chess program could receive a percept indicating that it has checkmated its opponent, but it needs a performance standard to know that this is a good thing; the percept itself does not say so. It is important that the performance standard be fixed. Conceptually, one should think of it as being outside the agent altogether, because the agent must not modify it to fit its own behavior. The last component of the learning agent is the problem generator. It is responsible for suggesting actions that will lead to new and informative experiences. The point is that if the performance element had its way, it would keep doing the actions that are best, given what it knows. But if the agent is willing to explore a little, and do some perhaps suboptimal actions in the short run, it might discover much better actions for the long run. The problem generator’s job is to suggest these exploratory actions. This is what scientists do when they carry out experiments [6].
1.6 Intelligent Mobile Agents
Mobile agents are autonomous intelligent programs moving around the network on behalf of the user helping the user to search for and interact with services on his/her behalf. These systems use specialized servers to interpret the agent’s behaviour and communicate with other servers. A Mobile Agent has inherent navigational autonomy and can ask to be sent to some other nodes.
Mobile Agents should be able to execute on every machine in a network and the agent code should not have to be installed on every machine the agent could visit. Therefore Mobile Agents use mobile code systems like Java and the Java virtual machine where classes can be loaded at runtime over the network
1.7 Research Problem
Both Silander’s AEFIRIP model for mobile learning [7], and Kazi’s model of an ideal Intelligent Tutoring System (ITS) [2][8] described in sections 2.3.4 and 2.3.6, respectively, are good ideally for M-learning in environments where there is adequate infrastructure. M-learning requires real-time access, a form of consistent connection and provision of adequate bandwidth if the goals of ‘learning anywhere, anytime’ must be met. In developing countries, internet infrastructures are either not affordable or not accessible to many students in semi-urban and rural areas; bandwidth cost is prohibitive especially when one has to pay in Dollars. Power is also a very critical factor, because E-learning and its associated technologies are dependent on constant provision of electricity. These factors make E-learning inadequate for most students in developing countries. By the end of 2008 there were over 4.1 billion mobile phone users worldwide according to ITU, [9], out of which two-thirds of the users are in developing countries. In Africa, Nigeria and South Africa had the highest number of mobile users, 26% and 19%, respectively, according to ITU [10]. On the other hand, while fixed broadband lines increased to almost 20 percent in rich countries, only just over 1 in 20 have access to fast Internet connections at home in developing countries. Again, while Asia and Europe had 42.4% and 23.6% of the global internet usage, respectively, Africa and the Middle East had 4.8% and 3.2%, respectively (http://www.internetworldstats.com/stats.htm) by the year 2009. Nigeria had the highest internet usage of 27.8%, followed by Egypt with 19.3% of the total internet usage in Africa. Statistics has shown that the growth of mobile phone usage is higher in developing countries than in the developed ones. Mobile connectivity has tremendously increased in Africa, from 600,000 subscriptions in 1995 to 735 million in 2012. The National Communications Commission puts the teledensity in Nigeria to be 85.25% in April, 2013 [66].
As at mid 2012, ITU [26] reported that there were about 6 billion mobile phone users, almost equaling the world population. The countries with the highest share of mobile traffic as part of total web traffic are India at 48.87 percent, Zambia at 47.09 percent, Sudan at 44.95 percent, Uzbekistan at 42.36 percent, Nigeria at 40.65 percent, Zimbabwe at 37.95 percent, Laos at 35.46 percent, Brunei at 34.66 percent, Ethiopia at 31.79 percent and Kenya at 29.2 percent.
The data reveals that Africa and Asia split the list between them. Africa amassed six countries, which left Asia with four. The first European country is the United Kingdom with 10.71 percent, and the U.S. showed 8.61 percent mobile web traffic as share of all web traffic. These statistics show that because of infrastructural challenges people in developing countries find it easier to surf the web with mobile devices, the major reason being that mobile communication operators have already provided the platform and infrastructures at much cheaper prices. Most of these mobile service providers offer 3G data services at more affordable rates than the traditional internet service providers, who deliver E-learning.
From education perspective, UNESCO offers the following statistics in 2010 [67]. That 10.5% and 30.6% of children in Nigeria and sub-Saharan Africa, respectively are out of school. While adult and youth literacy in Nigeria stands at 61% and 72%, respectively. The same figure is also applicable in sub-Saharan Africa. 35 million adults in the aforementioned regions cannot read or write, but can use the mobile phone to communication. Most rural dwellers in Nigeria have at least, one mobile phone.
Mobile phones and Personal Digital Assistants (PDAs) are low-powered devices, whose batteries, if fully charged can last at least, 48 hours. More also the prices are very affordable compared to the prices of PCs. Mobile learning is a technology not regulated to labs but facilitate personalized learning. It also removes barriers to technology use. Mobile devices are increasingly ubiquitous and powerful, and its expanding applicability to teaching and learning increases the potential to benefit learners everywhere. These premises make M-learning the nearest alternative to E-learning and distance learning education. We can leverage on these to implement M-learning, especially in countries where academic calendar is not consistent, and most of the illiterate groups dwell in remote locations where they may not be reached by conventional teaching and learning techniques.
This research, proposes a model of M-learning that uses a modified Silander’s model [7] in section 2.3.4 and Kazi’s models [2][8] in section 2.3.6. Unlike Kinshuk’s Bee-gent framework [4] in section 2.3.7, the proposed model will be implemented with JADE (Java Agent Development Environment) framework. Bee-gent is said to support agent-based inter-application communication, facilitating co-operation and problem-solving [11].
The research concludes with a software implementation of intelligent mobile learning system using the JADE (Java Agent Development Environment) platform, which is a multi-agent framework, instead of the Bee-gent framework used in the Kinshuk’s model. The application is developed using Java Server Pages (JSP) and Servlet technologies, as well as core Java classes and XML with associated web services such as the Web Service Integration Gateway (WSIG) for JADE.
1.8 Research Questions
The following research questions are formulated and answered in this thesis.
- Can learning be done effectively using mobile devices?
- Can a mobile learning system that overcomes the bandwidth limitations inherent in wireless and ad-hoc communications systems be implemented?
1.9 Research Objectives
The education industry needs new models and fresh frameworks to avoid losing touch with the radically evolving needs of its many current and potential new constituencies. The objectives of this research work are to propose and formulate a model for the design and implementation of effective M-learning system that will be able to adapt to the fast changing technologically mediated academic curricula which adapts to the peculiar challenges encountered in developing countries. This work will also demonstrate how one can use intelligent multi agents to address the problems that limit the potential of mobile learning environment development. This work will be able to amplify the advantages of M-learning systems while eliminating their drawbacks.
1.10 Research Methodology
There will be extensive study and evaluation of other existing M-learning models elsewhere in the world, with a view to proposing an effective model. Mathematical and other empirical, as well as software models may be used in this study where necessary. There may also be need to use network simulation and/or monitoring tools to demonstrate network traffic flow. Sample data, like network bandwidth, data size, network speed, etc may be gathered and analyzed as a basis for proposing acceptable standard in the M-learning industry.
1.11 Research Contributions
Contribution to Practice: It is expected that at the end of this research work a model of M-learning system that overcomes the major common deficiencies, such as slow access to course materials, lack of real time interaction, and the system not being able to adapt to individual students will emerge. Also, bandwidth limitations are expected to be overcome in the proposed model.
Contribution to Learning: This work will also form a basis for future implementations for mobile learning projects. Educators will be able to prepare their course modules using this model, as will be demonstrated in the sample application that accompany this work. Students also will be able to access course materials from their mobile phones anywhere, any time.
1.12 Scope of the Research
This thesis covers a review of previous work that done in the field of mobile learning. Detailed study of multi-agent technology is undertaken, a brief overview of learning theories, as well as the design and implementation of M-learning systems using artificial intelligence techniques.
Do you need help? Talk to us right now: (+234) 08060082010, 08107932631 (Call/WhatsApp). Email: [email protected].
IF YOU CAN'T FIND YOUR TOPIC, CLICK HERE TO HIRE A WRITER»
