1. INFSCI 2955 Adaptive Web Systems Session 1-2: Adaptive E-Learning Systems Peter Brusilovsky School of Information Sciences University of Pittsburgh, USA http://www.sis.pitt.edu/~peterb/2955-092/

3. Overview The Context Technologies Adaptive E-Learning Systems vs. Learning Management Systems (LMS)

4. The Context Adaptive systems Why adaptive? Adaptive vs. intelligent

5. Adaptive systems Classic loop user modeling - adaptation in adaptive systems

6. Adaptive Software Systems Intelligent Tutoring Systems Adaptive Help Systems Adaptive Interfaces Adaptive Hypermedia Systems Adaptive Search Adaptive Filtering/Recommendation Adaptive …

7. Major Issues What is adaptive? –Adaptive sequencing of educational tasks –Adaptive content presentation –Adaptive ordering of search results What kinds of information about user? –User knowledge –User interests –User individual traits

8. Why Adaptive E-Learning? Adaptation was always an issue in education - what is special about the Web? greater diversity of users –“user centered” systems may not work new “unprepared” users –traditional systems are too complicated users are “alone” –limited help from a peer or a teacher

9. Intelligent vs. Adaptive 1. Intelligent but not adaptive (no student model!) 2. Adaptive but not really intelligent 3. Intelligent and adaptive Intelligent ES Adaptive ES 2 3 1

10. Technologies Origins of AEL technologies ITS Technologies AH Technologies Native Web Technologies

11. Origins of AEL Technologies Intelligent Tutoring Systems Adaptive Hypermedia Systems Adaptive Web-based Educational Systems

12. Origins of AEL Technologies (1) Adaptive Hypermedia Systems Intelligent Tutoring Systems Adaptive Hypermedia Intelligent Tutoring Adaptive Presentation Adaptive Navigation Support Curriculum Sequencing Intelligent Solution Analysis Problem Solving Support

13. Technology inheritance examples Intelligent Tutoring Systems (since 1970) –CALAT (CAIRNE, NTT) –PAT-ONLINE (PAT, Carnegie Mellon) Adaptive Hypermedia Systems (since 1990) –AHA (Adaptive Hypertext Course, Eindhoven) –KBS-HyperBook (KB Hypertext, Hannover) ITS and AHS –ELM-ART (ELM-PE, Trier, ISIS-Tutor, MSU)

14. Technology Fusion Adaptive Web Adaptive Educational Systems Adaptive E-Learning

15. Adaptive Web Adaptive information filtering Adaptive search Adaptive recommendation Social navigation Log mining Machine learning CSCW Brusilovsky, P., Kobsa, A., and Neidl, W. (eds.) (2007) The Adaptive Web: Methods and Strategies of Web Personalization. Lecture Notes in Computer Science, Vol. 4321, Berlin Heidelberg New York: Springer-Verlag.

16. Origins of AEL Technologies (2) Adaptive Hypermedia Systems Intelligent Tutoring Systems Information Retrieval Adaptive Hypermedia Adaptive Information Filtering Intelligent Monitoring Intelligent Collaborative Learning Intelligent Tutoring Machine Learning, Data Mining CSCL

17. Inherited Technologies Intelligent Tutoring Systems –course sequencing –intelligent analysis of problem solutions –interactive problem solving support –example-based problem solving Adaptive Hypermedia Systems –adaptive presentation –adaptive navigation support

18. Inherited Technologies What is modeled? –User knowledge of the subject –User individual traits What is adapted? –Order of educational activities –Presentation of hypertext links –Presented content –Problem solving feedback

19. How to Model User Knowledge Domain model –The whole body of domain knowledge is decomposed into set of smaller knowledge units –A set of concepts, topics, etc Student model –Overlay model –Student knowledge is measured independently for each knowledge unit

20. Vector vs. network models Vector - no relationships Precedence (prerequisite) relationship is-a, part-of, analogy: (Wescourt et al, 1977) Genetic relationships (Goldstein, 1979)

21. Vector model Concept 1 Concept 2 Concept 3 Concept 4 Concept 5 Concept N

22. Network model Concept 1 Concept 2 Concept 3 Concept 4 Concept 5 Concept N

23. Simple overlay model Concept 1 Concept 2 Concept 3 Concept 4 Concept 5 Concept N yes no yes

24. Simple overlay model Concept 1 Concept 2 Concept 3 Concept 4 Concept 5 Concept N yes no yes

25. Weighted overlay model Concept 1 Concept 2 Concept 3 Concept 4 Concept 5 Concept N 10 3 0 2 7 4

26. Course Sequencing Oldest ITS technology –SCHOLAR, BIP, GCAI... Goal: individualized “best” sequence of educational activities –information to read –examples to explore –problems to solve... Curriculum sequencing, instructional planning,...

27. Course Sequencing What is modeled? –User knowledge of the subject –User individual traits What is adapted? –Order of educational activities –Presentation of hypertext links –Presented content –Problem solving feedback

28. Active vs. passive sequencing Active sequencing –goal-driven expansion of knowledge/skills – achieve an educational goal predefined (whole course) flexible (set by a teacher or a student) Passive sequencing (remediation) –sequence of actions to repair misunderstanding or lack of knowledge

29. Levels of sequencing High level and low level sequencing

30. Sequencing options On each level sequencing decisions can be made differently –Which item to choose? –When to stop? Options –predefined –random –adaptive –student decides

31. Simple cases of sequencing No topics One task type –Problem sequencing and mastery learning –Question sequencing –Page sequencing

32. What do we need for sequencing? Domain model –Network of concepts User model –Overlay model of user knowledge Model of Educational Material –Content Indexing Goal model (for active sequencing)

33. Simple case: one concept per ULM Concept 1 Concept 2 Concept 3 Concept 4 Concept 5 Concept N Random selection if there are no links -Scholar Links can be used to restrict the order

34. Indexing ULMs with concepts Example 2Example M Example 1 Problem 1 Problem 2Problem K Concept 1 Concept 2 Concept 3 Concept 4 Concept 5 Concept N Examples Problems Concepts

35. Simple goal model Learning goal as a set of topics

36. More complicated models Sequence, stack, tree

37. Indexing teaching material Types of indexing –One concept per ULM –Indexing of ULMs with concepts How to get the ULMs indexed? –Manual indexing (closed corpus) –Computer indexing (open corpus)

38. Sequencing with models Given the state of UM and the current goal pick up the best topic or ULM within a subset of relevant ones (defined by links) Special cases with multi-topic indexing and several kinds of ULM Applying explicit pedagogical strategy to sequencing

39. Sequencing for AES Simplest technology to implement with CGI Important for WBE –“no perfect order” –lack of guidance No student modeling capability! –Requires external sources of knowledge about student –Problem/question sequencing is self-sufficient

40. ELM-ART: question sequencing

41. SIETTE: Adaptive Quizzes Combination of CAT and concept Based adaptation

42. Models in SIETTE

43. Beyond Sequencing: Generation

44. Adaptive Problem Solving Support The “main duty” of ITS From diagnosis to problem solving support Highly-interactive support –interactive problem solving support Low-interactive support –intelligent analysis of problem solutions

45. Adaptive Problem Solving Support What is modeled? –User knowledge of the subject –User individual traits What is adapted? –Order of educational activities –Presentation of hypertext links –Presented content –Problem solving feedback

46. Models for interactive problem- solving support and diagnosis Domain model –Concept model (same as for sequencing) –Bug model –Constraint model Student model –Generalized overlay model (works with bug model and constraint model too) Teaching material - feedback messages for bugs/constraints

47. Bug models Concept A Concept B Concept C Each concept/skill has a set of associated bugs/misconceptions and sub-optimal skills There are help/hint/remediation messages for bugs

48. Intelligent analysis of problem solutions Classic system: PROUST Support: Identifying misconceptions (bug model) and broken constraints (CM) Provides feedback adapted to the user model: remediation, positive help Low interactivity: Works after the (partial) solution is completed

49. Example: ELM-ART

50. Example: SQL-Tutor

51. Interactive Problem Solving Support Classic System: Lisp-Tutor The “ultimate goal” of many ITS developers Several kinds of adaptive feedback on every step of problem solving –Coach-style intervention –Highlight wrong step –What is wrong –What is the correct step –Several levels of help by request

52. Example: PAT-Online

53. Example: WADEIn http://adapt2.sis.pitt.edu/cbum/

54. Problem-solving support Important for WBE –problem solving is a key to understanding –lack of problem solving help Hardest technology to implement –research issue –implementation issue Excellent student modeling capability!

55. Do we need bug models? Lots of works on bug models in the between 1974-1985 Bugs has limited applicability –Problem solving feedback only. Sequencing does not take bugs into account: whatever misconceptions the student has - effectively we only can re-teach the same material –Short-term model: once corrected should disappear, so not necessary to keep

56. Models for example-based problem solving support Need to represent problem-solving cases Episodic learner model –Every solution is decomposed on smaller components, but not concepts! –Keeping track what components were used and when - not an overlay! ELM-PE and ELM-ART - only systems that use this model

57. Adaptive hypermedia Hypermedia systems = Pages + Links Adaptive presentation –content adaptation Adaptive navigation support –link adaptation Could be considered as “soft” sequencing –Helping the user to get to the right content

58. Adaptive problem solving support What is modeled? –User knowledge of the subject –User individual traits What is adapted? –Order of educational activities –Presentation of hypertext links –Presented content –Problem solving feedback

59. Adaptive Navigation Support Direct guidance Hiding, restricting, disabling Generation Sorting Annotation Map adaptation

60. Adaptive Annotation: Icons Annotations for topic states in Manuel Excell: not seen (white lens) ; partially seen (grey lens) ; and completed (black lens)

61. Adaptive Annotation: Icons 1. Concept role 2. Current concept state 3. Current section state 4. Linked sections state 4 3 2 1 √ InterBook system

62. Adaptive Annotation: Font Color Annotations for concept states in ISIS-Tutor: not ready (neutral); ready and new (red); seen (green); and learned (green+)

63. Adaptive Hiding Hiding links to concepts in ISIS-Tutor: not ready (neutral) links are removed. The rest of 64 links fits one screen.

64. What Size of “concept”? How much domain knowledge should a concept cover? Two practical approaches Topic-based student modeling –Large topics, one per ULM/page Concept-based student modeling –Small concepts, many per ULM/page

65. Topic-based Student Modeling Benefits –Easier for students and teachers to grasp –Easier for teachers to index content –Clear interface for presentation of progress Shortcomings –The user model is too coarse-grained –Precision of user modeling is low

66. Demo: QuizGuide

67. Concept-based Student Modeling Benefits –The user model is fine-grained –Precision of user modeling is good Shortcomings –Harder for students and teachers to grasp –Harder for teachers to index content –Presentation of progress is harder to integrate into the system interface

68. Demo: NavEx

69. Indexing Examples in NavEx Concepts derived from language constructs –C-code parser (based on UNIX lex & yacc) –51 concepts totally (include, void, main_func, decl_var, etc) Ask teacher to assign examples to lectures –Use a subsetting approach to divide extracted concepts into prerequisite and outcome concepts

70. ANS: Evaluation ISIS-Tutor: hypermedia-based ITS, adapting to user knowledge on the subject Fixed learning goal setting Learning time and number of visited nodes decreased No effect for navigation strategies and recall

71. Adaptive Presentation Conditional text filtering ITEM/IP, PT, AHA! Adaptive stretchtext MetaDoc, KN-AHS, PUSH, ADAPTS Frame-based adaptation Hypadapter, EPIAIM, ARIANNA, SETA Full natural language generation ILEX, PEBA-II, Ecran Total

72. Adaptive Presentation What is modeled? –User knowledge of the subject –User individual traits What is adapted? –Order of educational activities –Presentation of hypertext links –Presented content –Problem solving feedback

73. Example: PUSH (stretchtext)

74. Example: SASY http://www.cs.usyd.edu.au/~marek/sasy/ Scrutable adaptive presentation

75. Adaptive presentation: evaluation MetaDoc: On-line documentation system, adapting to user knowledge on the subject Reading comprehension time decreased Understanding increased for novices No effect for navigation time, number of nodes visited, number of operations

76. Models for Adaptive Hypermedia Domain model - same as for sequencing User model - same as for sequencing Goal model - same as for sequencing Model of the learning material –For ANS - same as for sequencing –For AP - could use fragment or frame indexing

77. Indexing of nodes Domain model Concept 1 Concept 2 Concept 3 Concept 4 Concept m Concept n Hyperspace

78. Indexing of page fragments Fragment 1 Fragment 2 Fragment K Concept 1 Concept 2 Concept 3 Concept 4 Concept 5 Concept N NodeConcepts

79. Adapting to User Knowledge: Other Ideas Adaptive interface –Presence of menus and widgets in an educational applet can be adapted to user knowledge Educational animation and simulation –Adaptive explanations –Adaptive visualization

80. Demo: WADEIn

81. Demo: Improve

82. Adapting to Individual Traits Source of knowledge –educational psychology research on individual differences Known as cognitive or learning styles –Field dependence, wholist/serialist (Pask) –Kolb, VARK, Felder-Silverman classifiers

83. Style-Adaptive Hypermedia What is modeled? –User knowledge of the subject –User individual traits What is adapted? –Order of educational activities –Presentation of hypertext links –Presented content –Problem solving feedback

84. Style-Adaptive Hypermedia Different content for different style –Recommended/ordered links –Generated on a page –Mixed evidences in favor Different navigation tools for different styles –Adding/removing maps, advanced organizers, etc. Good review: –Bajraktarevic, N., Hall, W., and Fullick, P. 2003. Incorporating Learning Styles in Hypermedia Environment: Empirical Evaluation, In Proceedings of Workshop on Adaptive Hypermedia and Adaptive Web-Based Systems, Nottingham, 41-52. http://wwwis.win.tue.nl/ah2003/proceedings/paper4.pdf

85. Example: AES-CS Interface for field-independent learners

86. Example: AES-CS Interface for field-dependent learners

87. Style-Adaptive Feedback What is modeled? –User knowledge of the subject –User individual traits What is adapted? –Order of educational activities –Presentation of hypertext links –Presented content –Problem solving feedback

88. Overview: Classic Technologies What?KnowledgeStyles Order of activities Sequencing? FeedbackAdaptive diagnosis Style-adaptive feedback ContentAdaptive presentation LinksAdaptive navigation support

89. Origins of AEL Technologies (2) Adaptive Hypermedia Systems Intelligent Tutoring Systems Information Retrieval Adaptive Hypermedia Adaptive Information Filtering Intelligent Monitoring Intelligent Collaborative Learning Intelligent Tutoring Machine Learning, Data Mining CSCL

90. Native Web Technologies Availability of logs - helping the teacher! –Log-mining –Intelligent class monitoring - class progress is available! One system, many users - group adaptation! –Adaptive collaboration support Web is a large information resource - helping to find relevant open corpus information –Adaptive content recommendation Possible combinations of the above –Collaborative recommendation –Social navigation

91. What You Can Get from Logs? Log processing and presentation –Presenting student progress on topic and concept level: making sense of class Course/site improvements Grouping users by learning styles Intelligent class monitoring –Comparing progress, identifying students way ahead and behind

92. Adaptive Collaboration Support Peer help Collaborative group formation Group collaboration support –Collaborative work support –Forum discussion support Mutual awareness support More information –Proceedings of the Workshop on Personalization in E-learning Environments at Individual and Group Level at the 11th International Conference on User Modeling, http://hermis.di.uoa.gr/PeLEIGL/PING07-proceedings.pdf

93. Personalized Access to Educational Resources A lot of resources are available on the Web and in educational DL/Repostitories A new direction of adaptation - provide personalized access to these resources –Content based recommender Adding advantage of community wisdom –Collaborative recommender systems –Social navigation systems

94. Modeling User Interests Concept-level modeling –Same domain models as in knowledge modeling, but the overlay models level of interests, not level of knowledge Keyword-level modeling –Uses a long list of keywords (terms) in place of domain model –User interests are modeled as weigthed vector or terms –Originated from adaptive filtering/search area

95. Keyword User Profiles

96. Use of Profiles in Adaptive Web

97. Use of Profiles in AES: ML Tutor

98. Social Computing Web 2.0 for education Collaborative filtering –AlteredVista Collaborative resource discovery systems –CoFIND –UMtella (Demo) Presence-based collaboration (Educo) Social navigation support for open corpus resources (Knowledge Sea II)

99. Demo: UMtella

100. Example: EDUCO

101. Demo: Knowledge Sea II

102. KSII: Map Interface Keywords related to documents inside the cell Indicator of density of document inside the cell Lecture Notes (landmarks) Indicator of user traffic Background color: indicator of group traffic Indicator of existence of annotation

103. Overview The Context Technologies Adaptive E-Learning Systems vs. Learning Management Systems (LMS)

104. Learning Management Systems From separate tools to Learning Management Systems (LMS) –University-level Cyberprof, Mallard, CM Online... –Consulting eCollege, Eduprise... –Commercial TopClass, WebCT, LearningSpace, Blackboard... –Open Source Moodle, SAKAI… Standardization: IMS, IEEE LTCS, SCORM...

105. What LMS Can Do For students –Course information and content delivery –Assessment and grades –Communication and collaboration For teachers –Authoring –Learning control –Student monitoring –Communication

106. What AES Can Do for Students Presentation –Adaptive presentation, adaptive navigation support, adaptive sequencing Assessment –Adaptive testing Communication and collaboration –Peer help and collaborative group formation –Collaboration coach Learning by doing –Problem solving support

107. What AES Can Do for Teachers Student monitoring –Identifying students in trouble Control –Sequencing –Adaptive navigation support Authoring –Concept-based authoring and courseware engineering

108. AES vs. LMS Adaptive E-Learning systems can provide a more advanced support for most functions –Course material presentation - InterBook, AHA –Assessment with quizzes - SIETTE –Threaded discussions - collaboration agents –Student management - intelligent monitoring Why LMS are not really adaptive? –Except simple control and learning design

109. Challenges How to make it working in practice? –AES systems use advanced techniques - hard to develop –AWBES content is based on knowledge - hard to create –AES require login and user modeling - hard to integrate Possible solutions - (watch, PhD students!) –Component-based architectures for AWBES –Authoring support –Open Corpus adaptive systems

110. Component-based Architectures Research systems can provide a better support of almost each function of E-Learning process Adaptive systems show how to implement nearly each component adaptively We need the ability to assemble from components –Course authors can choose best components and best content for their needs –Components providers and content providers have a chance to compete in developing better products

111. Current State Several component-based frameworks –ADAPT 2, ActiveMath, MEDEA,… Attempts to develop systems with internal components Reusable user/student model servers Some matching work in the standardization movement

112. Re-use/Standards Movement Learning Object Re-use supported by coming standards is another major research direction in E- Learning The re-use movement joins many existing streams of work driven by similar ideas –Create content once, use many times –Content independent from the “host” system –Content and interfaces with the host system are based on standards (metadata, CMI, etc) Let content providers be players in E-Learning The future is components and re-use

113. What is the Future? How to use good component/content if you have a Blackboard, Moodle or other major CMS? Is the future model a Blackboard-style giant system where all components are advanced and adaptive? –Wait for the CMS giants to integrate better tools? –Create our own “adaptive Blackboards” Is there any other choice?

114. ADAPT 2 Architecture Portal Activity Server Student Modeling Server Value-added Service

115. Knowledge Tree II Portal

116. Authoring Support Powerful tools for authors to create intelligent content ITS content editors –Algebra Tutor (Ritter) Adaptive Hypermedia authoring tools –MetaLinks (Murray) –AHA! (De Bra) –NetCoach (Weber)

117. MetaLinks (Murray)

118. AHA! (De Bra)

119. NetCoach (Weber)

120. Open Corpus Adaptive Systems Classic AES - Closed Corpus of pre-processed content Integrate Open Corpus content Bringing open corpus content in by indexing –KBS-HyperBook, SIGUE Processing open corpus content without manual indexing –NavEx (content-based guidance) –Knowledge Sea (social guidance)

121. Further Reading Brusilovsky, P. (2001) Adaptive hypermedia. User Modeling and User Adapted Interaction 11 (1/2), 87-110 Brusilovsky, P. and Peylo, C. (2003) Adaptive and intelligent Web- based educational systems. International Journal of Artificial Intelligence in Education 13 (2-4), 159-172. Brusilovsky, P. (2003) Developing Adaptive Educational Hypermedia Systems: From Design Models to Authoring Tools. In: T. Murray, S. Blessing and S. Ainsworth (eds.): Authoring Tools for Advanced Technology Learning Environments: Toward cost-effective adaptive, interactive, and intelligent educational software. Kluwer: Dordrecht, pp. 377-409. Brusilovsky, P. and Millan, E. (2007) User models for adaptive hypermedia and adaptive educational systems. In: P. Brusilovsky, A. Kobsa and W. Neidl (eds.): The Adaptive Web: Methods and Strategies of Web Personalization. Lecture Notes in Computer Science, Vol. 4321, Berlin Heidelberg New York: Springer-Verlag, pp. 3-53.