This paper propose an enhanced e-learning Social Network Exploiting Approach focused around chart model and clustering algorithm. Through closeness revelation, trust weights overhaul and potential companions change, the algorithm actualized a programmed adjusted trust association with progressively upgraded fulfillments#. It is an imuted e!erience on the grounds that eole investigate new zones of learning together in such a ath as to ma"e a tyical center and ideas.
Descrição original:
Título original
Solving The Problem of Adaptive E-Learning By Using Social Networks
This paper propose an enhanced e-learning Social Network Exploiting Approach focused around chart model and clustering algorithm. Through closeness revelation, trust weights overhaul and potential companions change, the algorithm actualized a programmed adjusted trust association with progressively upgraded fulfillments#. It is an imuted e!erience on the grounds that eole investigate new zones of learning together in such a ath as to ma"e a tyical center and ideas.
This paper propose an enhanced e-learning Social Network Exploiting Approach focused around chart model and clustering algorithm. Through closeness revelation, trust weights overhaul and potential companions change, the algorithm actualized a programmed adjusted trust association with progressively upgraded fulfillments#. It is an imuted e!erience on the grounds that eole investigate new zones of learning together in such a ath as to ma"e a tyical center and ideas.
International Journal of Advance Foundation and Research in Computer (IJAFRC)
Volume 1, Issue 8, August!1"# I$$% &"8 ' "8(&
88 ) * !1", IJAFRC All Rights Reserved +++#i,afrc#org $olving -he .ro/lem of Adaptive 0'1earning 23 4sing $ocial %et+or5s Mohammed Yaqub Anbar, Iraq yaqubmoha@gmail.com
A 2 $ - R A C - This paper propose an enhanced E-Learning Social Network Exploiting Approach focused around chart model and clustering algorithm, which can consequently gathering dispersed e- learners with comparative premiums and make fitting suggestions, which can at last upgrade the collective learning among comparable e-learners. Through closeness revelation, trust weights overhaul and potential companions change, the algorithm actualized a programmed adjusted trust association with progressively upgraded fulfillments# Inde6 -erms7 Relations, Adaptive E-Learning, Clustering, Social Network , E-learning, Collaborative Learning
I# I%-R894C-I8%
Throughout the last few decades, the nature of training in numerous instructive organizations has declined romting inadmissible low scholastic e!ecution, large amount of useful lac" of education, develoing oulace of unemloyable graduates. Adating never was limited to classrooms. The changing sociotechnical connection offers a guarantee of new oen doors, and the feeling that by one means or another things may be distinctive. #tilization of the Internet and other develoing innovations is sreading in recurrence, time and sace. Individuals and associations wish to utilize innovation to hel learning loo" for seculations to casing their comrehension and their advancement.
$earning is a dynamic transaction between individuals as one individual teaches and an alternate learns . It is an imuted e!erience on the grounds that eole investigate new zones of learning together in such a ath as to ma"e a tyical center and ideas. Additionally, it is a tyical encounter as understudy acquire the same educated viewoints of certain learning ranges.
%&$earning which brea"s the traditional classroom based learning mode enables distributed e&learners to access various learning resources much more convenient and fle!ible. 'owever, it also brings disadvantages due to distributed learning environment. Thus, how to rovide ersonalized learning content is of high riority for e&learning alications. An effective way is to grou learners with similar interests into the same community ()*.
In the early beginnings of the Adative systems develoment this new aroach of software individualization romised to imrove human&comuter interaction considerably. Many frequently occurring roblems of usability and learn ability seemed to be easily solved. 'owever, even today, after elaborating significantly the modeling techniques, it is not obvious whether this romise has been "et, because only few emirical evidence e!ists that suorts this claim(+*. Adative e&$earning is a new aroach that can ma"e an %&$earning system more effective by adating the resentation of information and overall lin"age structure to individual users in accordance with their "nowledge and behavior(,*. International Journal of Advance Foundation and Research in Computer (IJAFRC) Volume 1, Issue 8, August!1"# I$$% &"8 ' "8(&
8: ) * !1", IJAFRC All Rights Reserved +++#i,afrc#org
A social networ" is a structure formed by eole and by connections between eole, with the connections enabling interactions and e!change of information and influence (-*(.*,/ocio technical systems recognized many years ago that organizations functioned most effectively when their social and technological networ"s were comatible. This is the case e!actly with e&$earning systems (0*. /ocial networ" analysis is a research area that has received significant attention, esecially in the last several years. The corresondence between real&world social relations and social media ties has been confirmed on several occasions (1& )2*.
In other hand, costs of digital devices, storage and bandwidth have fallen, audio, video and te!t rocessing has come to be used daily by eole at home, at wor" and in formal education. Internet usage has became commonlace in the develoed world and is growing fast in develoing countries see 34igure ) , 4igure +5 36Internet #sage /tatistics,6 +2)) and +2)+5. There are still significant discreancies in Internet enetration rates but Africa is en7oying the greatest growth rate in users +222&+2)+ 3from a low base5, and Asia already has the largest number of Internet users. This widesread and growing number of Internet users offers the ossibility of local and global dialogue and sharing of resources, sub7ect to linguistic and socio&cultural constraints.
Figure 1# Internet 4sage $tatistics for 3ear !11
Figure # Internet 4sage $tatistics for 3ear !1 #
II# R01A-09 ;8R<
Adative e&learning is often meant to be new or in an early develoment stage())*.8omuter&based learning system are critized by many researchers for their adatability of teaching actions comared to rich tactic and strategies emloyed by human e!ert teachers and limited ranges ()+* (),*. International Journal of Advance Foundation and Research in Computer (IJAFRC) Volume 1, Issue 8, August!1"# I$$% &"8 ' "8(&
:! ) * !1", IJAFRC All Rights Reserved +++#i,afrc#org
There are different tyes of online learning that is individualized, but an adative e&learning system is regarded as one of the most recent models. This system tries to adat to each individual by resenting learning materials in accordance with one or several characteristics of the student. The adative e& learning system needs to be develoed by considering that the conventional e&learning systems are not caable of roviding an individualized learning9 they resent the same material for all students ()-*. :eneral Architecture of adative e&learning , in this art, we resent various diagrams of alication design for an adative e&learning shown in 34igure ,5. The ob7ective is to conceive a system which can model the descrition of edagogic resources and guide the learner in his formation according to his assets and to the edagogic ob7ective that is defined by the trainer. This edagogic ob7ective resents the caacities that the learner must have acquired at the end of the formation activity. The %!ert /ace which relates to the monetization of edagogic resources to reare them to be used by the adater. In %!ert /ace, nominally the teacher or the e!ert, who see"s to integrate new resources in the base, describes them by filling a form ,The $earner /ace erforms the following 7obs; it accommodates the identifiers of a learner, selects his rofile from the $earners database and returns it to the adater as well as the goal of this formation. The adater 3adatation rocess5 uses otimization algorithms to see" the otimal strategy while selecting the courses in the resources base and rovides them to the user interface. The $earner database contains the identifiers of the learner and his "nowledge or asset. As a result, the system rovides an otimal courses list to achieve the current goal by alying the genetic algorithms to see" the intermediate states ().*.
Figure &# $tructure of the adaptive e'learning s3stem
Learners Data Base Request = profile+ goal Result =list
Adaptation process Resource + form Modelisation Process
Learner space Resources Data Base Expert Space Learner Interface Interface International Journal of Advance Foundation and Research in Computer (IJAFRC) Volume 1, Issue 8, August!1"# I$$% &"8 ' "8(&
:1 ) * !1", IJAFRC All Rights Reserved +++#i,afrc#org <irtual $earning %nvironments sometimes called learning management systems or managed learning environment, create a controlled virtual learning sace where teachers and students interact, ost and submit homewor", give and receive feedbac" from their eers, and lin" to course resources and information. 8ombining technology and education, <$%s rovide comrehensive course management tools and are often used in distance education. Traditionally, <$%s were simlistic and contained a lot of te!t, simle grahics, and included the lecturer=s notes in an e>oo" format. ?Today, we have the ability to create very sohisticated and comle! interactive virtual environments,@ that include sound, video, and animation. These <$%s hel students and teachers interact and communicate in ways reviously not ossible 34igure -5 shows >lac"board e&learning framewor" which has many limitation in social interaction and ersonalization ()0*.
Figure "# 2lac5/oard 0'learning Frame+or5
Adative&based learning The ob7ective of adative based learning is to conceive a system which can model the descrition of edagogic resources and guide the learner in his formation according to his assets and to the edagogic ob7ective that is defined by the trainer, This edagogy ob7ective resents the caacities that the learner must have acquired at the end of the formation activity ()1*()A*. Adative e& $earning systems would be a good solution for better e&$earning()B*./ystem such Moodle use adaptive e- learning which creates the best ossible learning e!erience for students by emulating the talents of great educators. This is achieved by using technologies that adat and shae teaching to the needs of the individual student.
The rocess of alying rule mining over the Moodle data consists of the same four stes as the general data mining rocess (+2*;
Collect data. The $M/ system is used by students and the usage and interaction information is stored in the database. Ce are going to use the students= usage data of the Moodle system.
International Journal of Advance Foundation and Research in Computer (IJAFRC) Volume 1, Issue 8, August!1"# I$$% &"8 ' "8(&
: ) * !1", IJAFRC All Rights Reserved +++#i,afrc#org .reprocess the data. The data are cleaned and transformed into a mineable format. In order to rerocess the Moodle data we used the My/D$ /ystem Tray Monitor and Administrator tools (+)* and the Een F> Grerocess tas" in the Ce"a %!lorer (++*.
Appl3 association rule mining. The data mining algorithms are alied to discover and summarize "nowledge of interest to the teacher.
Interpret, evaluate and deplo3 the results. The obtained results or model are interreted and used by the teacher for further actions. The teacher can use the discovered information for ma"ing decision about the students and the Moodle activities of thecourse in order to imrove the students= learning(+,*.
Gro7ect&>ased $earning a strategy that encourages students to e!lore a roblem through driving questions and authentic e!eriences, can be an e!cellent means to a social constructivist end. G>$ hels teachers blend technology with hands&on materials to create imaginative and challenging student activities (+-*.
Management Information /ystem %&$earning tends to revolutionize and manage the learning rocess (+.*and is divides in two tyes #niversity Management Information /ystem and $earning Management /ystem 3$M/5 is the software that automates the administration of education. $M/ registers students, trac"s courses in a catalogue, records data from learners, and rovides reorts to management. An $M/ is tyically designed to handle courses by multile ublishers and roviders. It usually doesnHt include its own authoring caabilities9 instead, it focuses on managing courses created by a variety of other sources (+0*.
:lobal 8ollaboration ta"es such student engagement and learning outside the wallsIor virtual wallsIof the classroom and into the greater world. $earning communities, such as those established by communities or universities, rovide a lace for learners to ursue their interests, solve a roblem, rovide suort, get feedbac", or otherwise connect with eole who share similar assions. /uch constructivist interaction can be scaffold in a classroom environment to allow students to wor" more indeendently and even hel someone else solve a roblem (+1*.
III# .R8.8$09 FRA=;8R< A# .roposed Frame+or5 In order to overcome the roblem of traditional In order to overcome the roblem of traditional e& learning or adative e&learning, we roosed half breed framewor" that satisfies the social e&learning framewor" however with some new features or adative e&learning, we roosed mi!ture framewor" that satisfies the social e&learning framewor" yet with some new features. 8ollaborative feature, each user 3student, teacher5 has own sharing and chatting tool which introduce availability. /emantic /uort feature, each student or teacher has suorted with intelligent rocess which suggest the closest friends. Agent feature, each agent in the community holds a set of resources such 3Grofiles, 4riendshi, 8ourses and %!ams5 which are rated by roosed algorithm. 34igure .5 shows roosed social e&learning framewor". International Journal of Advance Foundation and Research in Computer (IJAFRC) Volume 1, Issue 8, August!1"# I$$% &"8 ' "8(&
:& ) * !1", IJAFRC All Rights Reserved +++#i,afrc#org To measure the nature of framewor" must first ensure that framewor" satisfied ,th rinciles of connectives. Grinciles of connectives are the accomanying; J Gersonality of connections with abnormal amount of security. J 'uge 8aacity is needed for successful social learning networ". J Fecision ma"ing is itself a learning rocess. Gic"ing what you learn and who you learn.
Figure (# .roposed $ocial 0'learning Frame+or5 International Journal of Advance Foundation and Research in Computer (IJAFRC) Volume 1, Issue 8, August!1"# I$$% &"8 ' "8(&
:" ) * !1", IJAFRC All Rights Reserved +++#i,afrc#org 2# .roposed Algorithms Kodes which control our framewor" are actors such 3/tudent, Teacher and Gosts or 8ourses5. %very node has one or more relations with other node. The strength of relation is calculated by grah classification techniques. 34igure 05 shows social e&learning concet grah. Ce suose case study of - students, - courses and - teachers which I means number as 3),+,..n5 and cursors means relations and nodes / means /tudent, 8 means 8ourses and T means Teachers.
Figure ># -he ?raph of $ocial 0'learning Concept <ery Lelationshi between different nodes has strength number come from matri! of this relationshi. Lelations are studentHs friends and favorite courses. Cith these relations, system can be determined which friend or course must be suggested first. 4rom 34igure 05, we can demonstrate the following matrices of relationshis in order to classified friends or courses and therefore enhance e&learning.
4irst, the friend relationshi matri! M Then from this matri!, we can conclude that /) is friend of student /+ in our case study and /+ is friend of student /,, and /, will be the first suggested friend to /-because they subscribers in friendshi of /,.
4irst, the favorite courses relationshi matri! M Then from this matri!, we can conclude that /-has not any courses in our case study and will be the first suggested courses is 8+ because 8+subscribers in friendshi of /+ and /,.
C1 C2 C3 C4 S1 S2 S3 S4 S 1
S 2
S 3
C 1
C 2
C 3
T 1
T 2
T3 S4 T4 C4 S1 S2 S3 S4
S
Si1 S2 S3 S4
International Journal of Advance Foundation and Research in Computer (IJAFRC) Volume 1, Issue 8, August!1"# I$$% &"8 ' "8(&
:( ) * !1", IJAFRC All Rights Reserved +++#i,afrc#org IV# R0F0R0%C0
()* L. /hen, 4. Yang, and G. 'an, A dynamic self&organizing e&learner communities with imroved multi&agent matchma"ing algorithm. The )0th Australian Noint 8onference on Artificial Intelligence +22,, . .B2I022.
(+* 8hen 'uandong, Cu /hulei , /ong 8hunhui. Analysis and rosect on the %&$earning /ystem instructional design. +22A International /ymosium on 8omuter /cience and %ngineering, I/I/%+22A. I%%% 8EMG#T%L /E8IT%TY;+22A, ,2,&,20.
(,* L. /hen, 4. Yang, and G. 'an, A dynamic self&organizing e&learner communities with imroved multi&agent matchma"ing algorithm. The )0th Australian Noint 8onference on Artificial Intelligence +22,, . .B2I022.
(.* Yu, A. Y., Tian, /. C., <ogel, F., P 8hi&Cai Owo", L., 68an learning be virtually boostedQ An investigation of online social networ"ing imacts6, 8omuters P %ducation, ..3-5, )-B-& ).2,,+2)2.
(0* F. G. da /ilva, L. <. Furm, %. Fuval, and '. EliviR,?8oncets and documents for adative educational hyermedia; A model and a rototye,@ resented at the +nd Cor"sho on Adative 'yerte!t and 'yermedia 'YG%LT%STHBA, Gittsburgh,)BBA.
(1* F. Mac$ean, /.'angal, /. O. Teh ,M. /. $am ,N. 'eer ,6:rous Cithout Tears; Mining /ocial Toologies from %mail6, 8omuter /cience Feartment, /tanford #niversity,+2)).
(A* O. Fzieciolows"i, 6Is Fata Mining Fifficult to $earnQ And If so, Chat can Ce do to Ma"e it %asierQ6,IAFI/ International 8onference Intelligent /ystems and Agents 3I/A5 And IAFI/ %uroean 8onference on Fata Mining,+2)+.
(B* L.L."horasgani ,68ommunity Mining And its Alications in %ducational %nvironment6, Thesis, 8omuter /cience Feartment, #niversity of Alberta,+2)2.
()2* '.N. Cu, I&'sien Ting, and '.Tsan 8hang, 6Intergrating /KA and FM Technology into 'L Gractice and Lesearch ; layoff Grediction Model 6,Feartment of Information Management, Kational #niversity of Oaohsiung, Taiwan,+2)2.
())* Fietinger, T,6 Asects of %&$earning %nvironments. Fissertation6, :raz #niversity of Technology, . +.,+22,.
()+* 4eli! Modritscher , <ictor Manuel :arcia&>arrios , 8hristian :Ttl , 6The Gast, the Gresent, and the 4uture of adative %&$earning6, Groceedings of the International 8onference on Interactive 8omuter Aided $earning 3I8$+22-5.
(),* 'aitham A. %$&:hareeb , Thesis 6Etimizing /ervice Eriented Architecture To /uort %&$earning Cith Adative And Intelligent 4eatures6,+2)).
International Journal of Advance Foundation and Research in Computer (IJAFRC) Volume 1, Issue 8, August!1"# I$$% &"8 ' "8(&
:> ) * !1", IJAFRC All Rights Reserved +++#i,afrc#org ()-* 'erman Fwi /ur7ono ,6 The Fesign of Adative %&$earning /ystem based on /tudent=s $earning /tyles6, International Nournal of 8omuter /cience and Information Technologies, <ol. + 3.5 ,.+,.2.,+22).
().* Azough /., >ellaf"ih M., >ouya"hf %l '. 3+2)25 Adative %&learning using :enetic Algorithms; IN8/K/ International Nournal of 8omuter /cience and Ketwor" /ecurity, <E$.)2 Ko.1, Nuly +2)2.
()0* 8lar", /., P Maher, M. $. ,6The role of lace in designing a learner centred virtual learning environment6,In >. F. <ries, N. G. <. $eeuwen, P '. Achten 3%ds.5, 8omuter Aided Architectural Fesign; 4utures +22) 3. )A1I+225. /ringer.+22).
()1* /amia Azough,,Mostafa >ellaf"ih,%l 'oussine >ouya"hf,6 Adative %&learning using :enetic Algorithms6, International Nournal of 8omuter /cience and Ketwor" /ecurity3IN8/K/5, <E$.)2 Ko.1,+2)2.
()A* Marcus, K., >en&Kaim, F., >ain, M. ,6Instructional /uort 4or Teachers and :uided 4eedbac" 4or /tudents In An Adative e$earning %nvironment6, In; %ighth International 8onference on Information Technology; Kew :enerations 3ITK:5, +2)) Institute of %lectrical and %lectronics %ngineers 3I%%%5,+2)).
()B* A. M. Liad , 'amdy O. %l&Minir , 'aitham A. %l&:hareeb,6 Leview of e&$earning /ystems 8onvergence from Traditional /ystems to /ervices based Adative and Intelligent /ystems6, Nournal of 8onvergence Information Technology <ol -, Ko +,+22B.
(+2* %nrique :arcUa, 8ristVbal Lomero), /ebastiWn <entura, 8arlos de 8astro), Toon 8alders,6Mining learner rofile utilizing association rule for web&based learning diagnosis 6, www.elsevier.comXlocateXeswa, %!ert /ystems with Alications ,, ,.),,+221.
(+)* Fe $aat, M. 4., $ally, <., $ionen, $., P /imons, G. L. N. ,6 Gatterns of interaction in a networ"ed learning community; /quaring the circle6, International Nournal of 8omuter&/uorted 8ollaborative $earning. FEI )2.)221Xs))-)+&221&B220&-,+221.
(++* Ce"a, available at htt;XXwww.cs.wai"ato.ac.nzXmlXwe"aX 3accessed March )05, +22B
(+,* Moodle, available at htt;XXmoodle.orgX, 3accessed March )05,+22B.
(+-* %isenberg, M. ,6%ducational technology6, reimagined. Kew Firections for Youth Feveloment, +2)23)+A5, +.I,,. doi;)2.)22+Xyd.,1+,+2)2.
(+.* A. >or", ?The 4uture of $earning; An Interview with Alfred >or"@, Leview, %ducom, <ol. ,-.
(+0* A.M. Laid, 3+22.5, 6 e#niversity; An Intelligent /ystem for %&$earning6, %gytian Information Nournal, 4aculty of comuter and information, <ol.0, Ko +, 8airo #niv,)BBB
(+1* 'ill, N. L., 6$earning communities; Theoretical foundations for ma"ing connections6, In F. Nonassen P /. $and 3%ds.5, Theoretical 4oundations of $earning %nvironments 3+nd ed., . +0A&+A.5. Kew Yor", KY; Loutledge +2)+.