This essay has been submitted by a student. This is not an example of the work written by our professional essay writers.
Abstract: An adaptive e-learning system provides specific learning content to the specific learner according to their knowledge level, preferences, Prerequisite knowledge and learning style. Individualization is very important to improve learning experiences. The aim of e-learning is to improve the learner's learning and performance levels during online learning. Each student may have different learning styles, preferences and knowledge levels. This is the reason why there is the need for adaptability of learning environment. The adaptive intelligent system helps students to achieve their learning goals effectively by delivering knowledge in an adaptive way through online learning settings. So, there is an aim to provide education to students or learners in different places around the whole world at any time, where no teacher is available for face-to-face help. In this, Semantic Web technologies can also used to have changed the direction of E-learning systems from task-based approaches to knowledge-intensive ones. Semantic web overcome limitations of the current web. It supports the implementation of intelligent agents, provides features of adaptability, using search engines and inference. In this paper we have proposed new strategies for constructing Adapting system in e-learning using the semantic web where we can search the content according to individual learner. Using semantic search, it can improve the process of searching .It analyzes the learners model with the help of ontology. After the implementation of this e-learning system, the learner will get a new positive educational experience
Keywords: Adaptive learning, semantic web, e-Learning, Learner Modeling.
Adaptability is the main issues in today's online e-learning era. The e-learning systems provide educational services to a wide range of students and they can help students to achieve their learning goals by delivering knowledge in personalized way.The learners have different knowledge level ,mental level and personalities. The aim of e-learning is to improve the learner's learning and performance levels during online learning. Adaptive learning is a type of e-learning technology which works wisely to understand the learner's needs before presenting the content.
Each student may have different learning styles, preferences and knowledge levels. This is the reason why there is the need for adaptability of learning environment. E- Learning systems must be flexible so that it could be suitable for different type of students or learner. Teaching at a distance includes the use of variety of skills for the instructors comparing to the ones used in traditional classrooms. The adaptability involves identifying and monitoring the student's learning activities according to his respective profile. The adaptive intelligent system helps students to achieve their learning goals effectively by delivering knowledge in an adaptive way through online learning settings. So, there is an aim to provide education to students or learners in different places around the whole world at any time, where no teacher is available for face-to-face help. That is the reason why to support e-learning systems. It would provide individualized help just as human tutor does. .In this, Semantic Web technologies can also used to have change the direction of E-learning systems from task-based approaches to knowledge-intensive ones. Semantic web overcome limitations of the current web. It supports the implementation of intelligent agents, provides features of adaptability, using search engines and inference. 
In this paper we have suggested new strategies for constructing Adapting system in e-learning using the semantic web where we can search the content according to individual learner. Using semantic search, it can improve the process of searching .It analyzes the learners model with the help of ontology. After the implementation of this e-learning system, the learner will get a new positive educational experience.
Adaptive systems are capable of adapting the content according to the profile of a particular learner. Thus, using adaptive learning learners with different learning goals, knowledge, preferences, knowledge level and learning styles can access different contents/data with different presentation formats. In order to ensure that information to be learnt more accurately ,the content presented and the detailing leaner should be given more in Semantically and systematic way. Semantic web technology organizes knowledge in more structure and more meaningful way. The most common adaptive learning systems consists of ontology whose major function is to provide a domain ontology and user ontology in order to provide adaptive support. Adapted system provide learning content to learner based on their prerequisites, of knowledge, preferences, learning styles and learning histories, as well as their characteristics. In  paper discusses about some fundamental Semantic Web technologies and applications of these technologies in e-learning but it does not explain the any algorithm for practically understand the concept. The work described in  utilizes semantic web with personalization web service and related work is done in  and .So the traditional methods and papers neither were providing accuracy nor effective methods and algorithm to dynamically collect the data according to individual learners. Therefore, in this paper, we have adopted the architecture, algorithm and work flow diagram as one the well known source of information for personalization.
3. Proposed Method for Adaptive System Based on semantic web:
3.1 Architecture of proposed Adaptive System:
Personalized system is essential for improving the learner's experience. It provides the content according to individual learner means according to each learner's level of knowledge, preferences, and other considerations. Here we have proposed an adaptive learning system which is based on semantic web technology that can provide highly personalized learning content. The architecture of the proposed semantic web based adaptive learning system consists of two main ontology as shown in Figure 1.
C:\Documents and Settings\another\Desktop\architecture.bmp
Figure 1: Architecture of the proposed ontology-based adaptive learning system
Ontology is used in knowledge-based systems as conceptual frameworks for providing, accessing, and structuring information in comprehensive manner. The ontology contains nodes, edges and relationship between them. We have constructed the ontology using the tool Neo4j.
Domain Ontology: Domain Ontology is a representation of domain knowledge which is known as domain Ontology. It describes the complete structure of domain. It is information content of the application which contains the concepts and concept relationships. The domain model of the system is totally based on the notion of learning aim.
C:\Documents and Settings\hdff\Desktop\domain.bmp
Figure 2: Classes and Object Properties of the Domain Ontology.
As it is shown in Fig. 2, a JAVA Tutorial consists of four main classes i.e. Core Java, JSP, Servlet and Struts. Servlet have some sub classes like ServletRequest, Servlet Collaboration, ServletConfig, Basics of Servlet, Session Tracking and ServletContext. Similarly JSP, CoreJava and Struts also have some sub classes. All the sub classes are connected with its main classes with a relationship.
Learning content management module: It is responsible for managing the content of knowledge base or Domain ontology. It inserts, deletes a new learning data into the domain Ontology or manipulates existing learning objects.
Learner model Ontology: This Ontology stores all learner-related data, i.e. the learner' profiles, including the behavior, characteristics, knowledge level, history, prerequisite knowledge, performance and deficiency record.In Fig. 3, Learner model Ontology consists of five main classes i.e. Learner Performance, learning style, learning time, prerequisite knowledge , preferences and learning history. The class "Preference" formally representing a learner's preference likes Language and Author. The sub classes are connected with main class preferences with relationship hasLanguage and hasAuthor.
C:\Documents and Settings\hdff\Desktop\muaahh.bmp
Figure 3: Classes and Object Properties of the Learner model Ontology.
The content will be presented according to learner's choice (author and language). Each Learner is assigned performance-related data that presented in the form of "Learner Performance" class. It has four sub-classes. "Deficiency Record" describes the mistakes which learner has made during the test and gives the solution of that questions. "Progress Report" contains the marks which learner has obtained after test. "Coding knowledge" defines the coding knowledge (java) of learner. "Date-Time Records" stores the date and time when the learning process took place. The class "History" stores all the content covered by learner. Class "prerequisite knowledge" stores the previous knowledge records of the learner.
Learner Model Management Module: It updates any changes in learner model ontology like updates data of history, performance and deficiency record.
Dynamic Content Collection Module: This module dynamically generates personalized learning content for a specific learner. This module is totally capable to combine available content (obtained from Learning Content Management Module) to form a coherent learning content that suits a particular learner.
Adaptive Presentation Module: The adaptive presentation module is responsible for presenting individualized learning content to the learner based on results from the dynamic content collection module. It presents adaptive e-learning content to the learner using link-hiding techniques. It hides the entire advance topic from the user and only shows the current content or previous covered content of the same topic. Green bullet shows a recommended content means the concept that the learner has not learned yet, but has knowledge about all previous topics. Yellow bullet shows the topic which is covered topic.
User Interface: It provides the interaction between the system and the learner.
3.2 Workflow of Adaptive System:
Under the proposed architecture, the adaptive system allows learners to get specific information about the learning content for the specific learner.
C:\Documents and Settings\another\Desktop\final flow.bmp
Figure 4: A Description Scenario of the workflow of Adaptive e-learning system.
A Scenario Example: John is a new learner; he wants to get the information on core java. So first of all he will register himself by giving some personal information. After that the system will ask some question from learner and make a Learner/student profile based upon his behavior, characteristics, knowledge level, history and prerequisite knowledge. Now system will dynamically collect the content according to user profile and shows it to him. When the learner will read all the content then system will take the evaluation. The question paper will contain all the questions related to the currently covered topic. Now system will update the learner information like learner's performance, history etc. In case learner needs information on other topic then again loop will go to the learner's profile and collect the user prerequisite information regarding the topic, learner's style, performance and history etc and dynamically collect the content for the specific learner.
Figure 5 is a screenshot of the Registration page when a learner registers with the system during the first
session. The learner will fill the basic information and create the username and password. After that next page will get open.
C:\Documents and Settings\another\Desktop\reg frm.bmp
Figure 5: Registration form for E-learning.
As shown in the figure 6, learner will fill all his learning style. Using the radio button learner will choose the one option from given three. There are basically four types of learning style i.e. visual-verbal, sensing-intuitive, sequential-global and inductive-deductive. The each learning style is explained in the table given below:
Pictures & Demonstrations
Words & Explanation
Patient with details, Careful but may be slow, Senses, facts and experimentation.
Quick but may be careless, principles and theories.
Steady progression and Convergent thinking and analysis.
Jumping directly and Divergent thinking.
ability to figure out the
Rules/theories/principle from observed instances of an event.
Moving from specific observations to broader generalizations and theories.
TABLE I: Different Learning style dimensions
C:\Documents and Settings\another\Desktop\learning style.bmp
Figure 6: screenshot for determining students learning style.
After learning style page, self evaluation form will be presented by adaptive system to learner. As for initial determination of the learner's knowledge about the domain, the system relies on the learner's Self evaluation. In particular, the learner is presented with the following set of options: 'No knowledge at all',
'Having slight knowledge', 'Familiar', 'Well enlighten' and 'Demand advanced topics', and 'High level topics required' as shown in Figure 7.Adaptive system will convert the learner's selection for each topic into its numerical value (0, 0.1, 0.2, 0.3 or 0.4 respectively). These numerical values will later help the system to find the learner's initial position in the Java domain. It will provide him proper guidance and
C:\Documents and Settings\another\Desktop\learning style.bmp
Figure 7: screenshot for calculating the Prerequisite knowledge of Learner.
4. Proposed algorithms:
The proposed first algorithm is showing how the Learner's content history is writing and reading and the second algorithm help the system to find the learner's initial position in the Java domain
4.1. Algorithm for History of Learner:
Writing LearnerContent: If the Learner is logged in, then system will check the learner id and store all the content which Adaptive system will recommend or present.
if(learner logged in)
Reading LearnerContent : First of all if the learner is logged in with some account and he wants to check his previous covered topics then he can check it by clicking the History button. The system will check the learner id of the learner and fetch the learner previous topics which he has covered earlier.
if(learner logged in)
//fetch past information from historical which is saved in //Learner Model Ontology.
// display recommended Content
4.2 Algorithm to find the learner's initial position (Prerequisite Knowledge) in the Java domain:
The algorithm is numerically evaluating the learner's Prerequisite Knowledge. So that according to learner's knowledge the content can be presented by the adaptive system.
getValue():At first learner gives the answers to all the five question which system asks. Now the PrerequisiteKnowledgeOfLearner() function is called to numerically evaluate the user knowledge.
//To get the initial determination of the learner's knowledge
//about the domain, the adaptive system will present five //questions.
// Enter the query (Q).
//learner is having good knowledge,
// show the content for which the learner is asking
//Learner's basic concept or previous concept is not clear
//show the basic content to the learner to clear the concept first.
PrerequisiteKnowledgeOfLearner(): When the learner types the query then Adaptive system first of all checks his Prerequisite Knowledge for example: suppose query is " AWT ".Now system will go to the getValue() to get the user input and calculate the learner Prerequisite Knowledge using function PrerequisiteKnowledgeOfLearner().Adaptive system converts the learner's selection for each topic into its numerical value (0, 0.1, 0.2, 0.3 or 0.4 respectively).Now system checks learner's self- evaluation form to know the that how much learner knows the topic which comes before "AWT" .If the learner selected "No knowledge at all', 'Having slight knowledge' or 'Familiar' for the previous topics then system will present the content of basic java and after that system will show the content of "AWT".
System will check if learner's knowledge for the previous topics is equals to or greater than 0.3 then system will show the content of "AWT".
In this paper considerable efforts have been made in the adaptive learning using semantic web searching to standardize knowledge learning building blocks. Adaptive system in e-learning presents the data according to the learner's knowledge level, preferences, Prerequisite knowledge and learning style. This paper attempts to respond to the demand for self-adaptive learning systems by providing an architecture, algorithm and data flow diagram. The suggested algorithm is guiding learners towards a customized learning route. The implementation of the presented Adapting system in e-learning environment framework further provides an open learning environment to offer enriching learning experiences.