Improved Content Personalization in E-Learning using MQACP

Improved Content Personalization in E-Learning using MQACP

1. Vijayalakshmi I. Shamili D. Sundaranarayana

   Assistant Professor, Dept of CSE, Final Year M.Tech, Dept of CSE, Final Year M.Tech, Dept of CSE,

   Christ College of Engg. & Tech, Pondicherry, India.

   Christ College of Engg. & Tech, Pondicherry, India.

   Christ College of Engg. & Tech, Pondicherry, India.

   AbstractData Mining is a process of identifying invisible patterns and relationship within data that can help for decision making. E-learning is a tool which supports teaching and learning using a computer web technology. The main aim of the E-learning is to provide exact content to the well known students profiles and preferences and also to specific the better content for each student. In this paper we proposed the comparison of techniques and algorithms which used in the system to improve the student knowledge level and academic progress, but the quality of e-learning service is low and so there is a lack of interaction among e-learners. The proposed System deals with a comprehensive metadata quality pledge method, specifically the (MQACP) Metadata Quality Assurance Certification Process. The basic of this method is to propose a set of methods that can be deployed throughout the lifecycle of a repository to ensure that metadata generated from content providers are of high quality. This system deals with Metadata quality assurance certification process to improve the high quality of learning object in digital repository and proper identification of metadata labeling. We measured the resulting metadata quality to certify that the anticipated improvement in metadata quality actually took place.

   KeywordContent Personalization; E-learning Planning; Intelligent E-Learning System; Educational technology.

   1. INTRODUCTION

      E-learning is the use of electronic instructive technology in knowledge and education. The development of e-learning systems and online courses has made instructive objects broadly available by students of changing ability, backgrounds and manners. An important necessity of the current e-learning systems is the personalization that is a role able to adapt the e-earning content or serves to the customer profile. The personalization

      Includes how to find and filter the studying information that suits the user preferences and wants a most important part of instructional plan and e-learning growth is to know the behavior of learners. This way, an e-learning experience can be created that offers the most benefit for the learners and ensure that every component of the e- learning course is helping them to reaching their objectives and goals.

      Personalized content in E-learning can particularly improve the learning routes personalization in an almost transparent way to the user. The fundamental thought is to build a powerfully-related and planned leaning route capable of fulfilling each students profile. The proposed System deals with a comprehensive metadata quality pledge method, specifically the (MQACP) Metadata Quality Assurance Certification Process. A digital repository is a method for supervising and storing digital content. Repositories can be focus or institutional in their centre of attention. Placing content into an institutional repository enables employees and institutions to supervise and maintain it, and thus gain maximum value from it.

      A repository can support explore, learning, and executive processes. Repositories use open principles to certify that the content they contain is available in that it can be sought and recalled for later use. The use of these approved international standards permits mechanisms to be place high which classify, store, retrieve, import and export the digital content inside the repository.The basic idea of this method is to propose a set of methods that can be deployed throughout the lifecycle of a repository to ensure that metadata generated from content providers are of high quality. We supported the metadata creation process throughout the different phases of the repositories involved by setting up the MQACP.

      During each level, we evaluated the ensuing metadata quality to attest that the expected improvement in metadata quality essentially took Place. The results showed that the MQACP was successfully adapted to the latest contexts, with smallest amount adjustment desired, through related outcome formed and also with comparable costs. We

      Can also determine high Quality assurance methods

   2. LITERATURE SURVEY

      For checking metadata documentation quality that will have an optimistic outcome on the resulting quality of the metadata records of a given group of repositories. Our Proposed Metadata Quality Assurance Certification Process that can be applies in several domains with proven effectiveness.

      PAPER NAME	YEAR	EXISTING	DISADVANTAGE	PROPOSED	ADVANTAGE	TECHNIQUE	ALGORITHMS
      E-Learning and	Feb 2014	1.Adjacency	To Monitor and Adapt	The myPTutor	The planning	Case-Based	–
      Intelligent		Matrices	the learning object of	approach	Techniques helps	Planning
      planning:		2.Integer	each learning route		to bridge the gap	Technique
      Improving		Programming	against unexpected		between the e-
      Content		Constraints	contingencies		learning
      Personalization		Satisfaction			necessities and
      		Models			the student
      		3.Neural			Content
      		Network			Adaptation
      		4.Soft
      		Computing
      		Methods
      A Generic	July 2013	One Size-Fits	Low Quality of e-	Personalized	1.Improve the	Intelligent e-	Inherited	Adaptive
      Model of		All	learning Service	search engine	Quality of e-	learning System	Object-Oriented
      student-based					learning Service		Structure	of the
      Adaptive					2.To Achieve the		Course Material
      Intelligent Web-					Student learning
      Based learning					goals effectively
      Environment
      E-Learning	2013	Educational	Lack of Interest in	Data Mining in	1.Improve	Visualization	–
      Using Data		Data Mining	Education Data Mining	E-learning	corporate the	Technique
      Mining		(EDM)	System		learning task
      					2.Successfully
      					incorporate to e-
      					learning
      					environment
      Quality	June	Numerous	Low Quality of e-	QA	Increasing the	Quality Assurance	Grade
      Assurance of e-	2011	initiative have	learning	Methodology	Quality of e-		Correspondence
      learning		been developed			learning		cluster analysis
      Processes		on QA in
      		Education
      Applying case-	2011	AI Planning	Sequence of learning	Approach to	Allow the best	Case-Based plan	Learning	object
      Based Planning			Objects (LO)	Automatically	learning routes for	Merging technique	Algorithm
      to personalized				extract	each student
      e-learning				information	profile and course
      				from the LO	objectives
      Planning and	2011	Adjacency	Execution of the learning	Adaptation	Reuse the original	LPG-	–
      Execution in		Matrics, Integer	routes, check its progress	Approach (Plan	Route as much as	ADAPT,LPG &
      personalized E-		Programming	and act when	Stability)	possible and adapt	SGPLAN6
      learning Setting		Model,	discrepancies appears		/replan
      		Neural Network
      		and AI Planning
      Data Mining	2010	Educational	Resulted in a high Drop	Dynamically	1.During the	Cluster difficulty	K-means
      For		Data Mining	out of Participants	tailored hints	challenge in the	ranking	algorithms
      Individualized		(EDM)		for users	form of hints and
      Hints in e-					easier access to
      learning					notes
      					2.hinting system
      					were evaluated
      					through a large
      					scale participants
      Discovering	2007	Filter or	Not paying too much	Adaptive	User can make a	Stretch-Text	IB algorithm
      student		Sort	attention to student	Machine	better use of it	techniques
      preferences in			preferences	learning system
      e-learning
      E-Learning	2007	Electronic	Lack on Interaction of	Learners	Improving the	Data driven	Frequent	Episode
      Process		Learning	Learners	Profile	Flexibility and	Approach	Discovery
      Characterization		Environment		Characterization	Authenticity of		Algorithm(FED)
      using Data					the Learners and
      driven					cost Benefits
      Approaches					Ratio

      In intelligent planning Antonio Garrido and Lluvia Morales rescues a victorious technique called myPTutor is a combined effort to take improvement of Artificial Intelligence planning techniques in the adaptation of sequences of Learning Objects to pedagogic and students' necessities[1]. In [5, 6] Ivan Serina prove that the imminent is a real valuable to increase the stability of the education course as well as to increase the performance and quality of the education routes. The most important intention of intelligent e-learning environments is to get better students learning progression by generous higher educational techniques, so that intelligent learning environments are supported on different pedagogical approach and theories that have been originated in the education field [2, 3] Nedhal A. M. Al Saiyd, Intisar A. M. Al-Sayed and Shimaa Abd Elkader Abd Elaal recommended the suitable architectural design for the growth modules of intelligent and contented with the intelligent Web-based learning environment about the effectiveness and competence. Zbigniew Mrozek [4] a quality assurance system (QA) had better to make sure the Necessitates is satisfied. Similarly the significance of fundamental guarantee plus learner perceptions, Increase in Educational facility, and also improves efficiency of E- learning system. In [7] Anna Katrina Dominguez, Kalina Yacef, James R. Curran well again Higher average mark achieved from user who be permissible with preventative hints to put off future mistakes. In [8] Cristina Carmona, Gladys Castillo, Eva Millán brought in an adaptive learning user model aimed at

      discovery out the students preferences about the educational materials terminate Time, such model is actually appropriate in e-learning systems so as to require to filtering the large intensities o information available, and so users can build a better use of it. Silvia Rita Viola

      [9] recommended the approaches used for learners profiles characterization. significant the value of the learning system used by learners in Learners profiles characterization in both the area by means of deference to different ways of non linear navigation. By comparing all techniques and algorithms is shown in survey table [10] those techniques and algorithms which used in the system to recover the student knowledge level and academic progress, although the quality of e-learning examine is low and so there is a lack of interaction among e-learners.

   3. EXISTING SYSTEM

      Data mining extract the data as of the e-learning content in order to adjust, stores, reprocess and to select the righter learning method according to the students profile and preference. By applying the myPTutor techniques [12] they have suggested how to monitor and adjust the studying objectives of every education routes against surprising contingencies.

      The techniques comprise of 4 steps. The First one is Knowledge Extraction consist Automated techniques of information technology to mine their necessary features. Second one is the Solve Problem It involves the usage of a

      smart planner to discover a plan or education path. Third is Execute & Monitor the implementation of the education path inside a LMS that supports supervising.

      We assure whether the actual state meets the awaited state. If a disagreement is noticed, that is if changeability happens on the route implementation, we apply plan- validation technique to assure if the path is quiet possible. The last and final one is Repair/Adapt the plan it requires fixing or adjusting the plan to make it executable again. The drawbacks found in the existing system is the explanation of the contents is not sufficient and the quality of the produced metadata records is not in high quality.

   4. PROPOSED SYSTEM

      The Present System proposing a complete metadata quality assurance technique, namely the Metadata Quality Assurance Certification Process (MQACP). The fundamental estimate of this technique is to propose a set of methods that can be deployed during the lifecycle of a repository to ensure that metadata generated from content providers are of high quality. This method handles with Metadata quality assurance certification process to improve the high quality of learning object in digital repository and proper recognition of metadata labeling. We evaluated the resulting metadata quality to certify that the excepted improvement in metadata quality really took place. The main advantages of proposed system it gives Accuracy, Correctness, Completeness, and Appropriateness of Metadata labeling. The automated mechanisms of measuring information quality that can easily be scaled up and maintained. The metadata quality is an integral part of any digital repository project and a critical success factor.

   5. OVERALL ARCHITECTURE

      Fig 1: System Architecture

   6. ALGORITHMS

      Fig 2: K-Means Clustering Algorithms Flowchart

   7. MODULES DESCRIPTION

      In [11] MQACP method personalized to the contents, with lowest adaptations required, and moreover with equivalent expenses. Metadata lies at the spirit of each digital repository plan in the logic that it determines and causes the explanation of digital content stored in the repositories. Metadata permits content to be successfully stored, managed and retrieved but also conserved in the long-term. Although the large significance of metadata in digital repositories, solitary that is broadly established, examines specify that what is defined as metadata quality is comparatively low in nearly events of digital repositories [13]. Metadata superiority is freely defined as suitability for purpose meaning that low quality of metadata means that metadata cannot complete its principle which is to permit for the retrieval of resources, management and successful storage. In present low metadata quality conducts to useless searches for content, no more resources which make them unseen to the aimed user, which is the customer of each digital repository.

      The present thesis approaches such problem by suggesting a broad metadata quality pledge method, specifically the (MQACP) Metadata Quality Assurance Certification Process. The basic idea of this thesis is to suggest a set of methods that can be deployed all over the development of a repository to assure that metadata generated from content providers are of high quality. These methods have to be direct, simple to apply with important outcome. They also have to be adaptable with minimum effort so that they can be used in different contexts easily. This set of methods was described systematically, taking into account the actors needed to apply them, drawing the tools required and significant the expected outcomes.

   8. CONCUSLION

Modern courses in remoteness education center of attention on exhibiting digital contents on LMSs and on covering such contents as learning objects labeled according to SCORM standards. The aim of these standards is perhaps to sell, exchange among platforms and/or schools, or else just entrée the contents distantly and all of these by using LMS-based environments. subsequent this thread, in this paper we have presented a general system implemented on top of Moodle that facilitates the use of generic GUIs to build up, incorporate and manage the alteration system based on intelligent planning. The preparation tools do not permit us to personalize the learning paths, other than in addition to carry out, observe their advance and adapt them when unexpected contingencies are found.

REFERENCES

1. Antonio Garrido and Lluvia Morales, E-Learning and Intelligent Planning: Improving Content Personalization, IEEE Revista Iberoamericande Technologies Del Aprendizaje, Vol.9, No.1, Feb.2014.

2. Nedhal A.M.AI Saiyd, And Intisar A.M.AI -Sayed , A Generic Model Of Student-Based Adaptive Intelligent Web-Based Learning Environment, Proceedings Of The World Congress On Engineering 2013 Vol II, WCE 2013, July 3-5, 2013, London, U.K.

3. Shimaa Abd Elkader Abd Elaal, E-Learning Using Data Mining, Chinese-Egyptian Research Journal Helwan University 2013.

4. Zbigniew Mrozek, Quality Assurance Of E-Learning, Presented On 22-Nd EAEEIE Annual Conference, EAEEIE 2011, Maribor, Slovenia, and JUNE 13-15, 2011.

5. Antonio Garrido, Lluvia Morales and Ivan Serina, Applying Case-Base Planning To Personalize E-Learning, Proceedings/ Dms11/ DET/_Antonio_ GARRIDO.PDF 2011.

6. L. Morales, A. Garrido, and I. Serina, Planning and Execution in A Personalized E-Learning Setting, In Advances in Artificial Intelligence (Lecture Notes In Computer Science). New York, NY, USA: Springer- Verlag, 2011, Pp. 232242.

7. Anna Katrina Dominguez, Kalina Yacef, James R. Data Mining for Individualized Hints in E-learning, University Of Sydney, Australia 2010.

8. Cristina Carmona1, Gladys Castillo2 and Eva Millán. Discovering Student Preferences In E-Learning, International Workshop On Applying Data Mining In E-Learning (ADML07) As Part Of The Second European Conference On Technology Enhanced Learning (EC-TEL07) 2007.

9. Silvia Rita Viola, E-Learning Process Characterization Using Data Driven Approaches, International Workshop On Applying Data Mining In E-Learning (ADML07) As Part Of The Second European Conference On Technology Enhanced Learning (EC-TEL07) 2007.

10. V.vijayalakshmi, I.shamili, K.kaarguzhali, Survey on E-learning: Comparison Techniques and Algorithms International Conference on Evolving Trends in Engineering & Technology (ICETET 2015).

11. Palavitsinis. N, Manouselis. N, and Sanchez. S, (2011). "Metadata quality in learning repositories: Issues and considerations, in Proc. of the World Conference on Educational Multimedia, Hypermedia & Telecomunications (ED-MEDIA 2011), Lisbon, Portugal

12. Phivos Mylonas, Paraskevi Tzouveli and Stefanos Kollias, E- learning and intelligent content adaptation: an integrated approach, Int. J. Cont. Engineering Education and Life-Long Learning, Vol. x,

    No. x, 200x 273 2007

13. Sarah Currier, Jane Barton, Rónán OBeirne & Ben Ryan,

Quality assurance for digital learning object repositories: issues for the metadata creation process, ALT-J, Research in Learning Technology Vol.12, No. 1, March 2004