Download Educational data mining : applications and trends

Alejandro Pefia-Ayala
Editor
Educational Data
Applications
^
Springer
and Trends
Mining
Contents
Part I
1
Profile
Which Contribution Does EDM Provide to Computer-Based
Learning
Environments?
3
Nabila Bousbia and Idriss Belamri
1.1
Introduction
1.2
Educational Data
1.3
5
1.2.1
Definition
1.2.2
Areas in Relation to EDM
6
1.2.3
Objectives
6
1.2.4
The Used Methods
1.2.5
The
1.2.6
Process of
1.2.7
Some
Examples
5
of the EDM
Analyzed
8
Data
11
Applying the
Technological
of EDM
1.3.1
EDM
1.3.2
EDM
EDM
12
Tools Used in EDM
Applications
Learning Environments
1.3.3
1.4
4
Mining
in
Computer-Based
13
for
Applications
Predicting and Evaluating
Learning Performance
Applications for Analyzing
A
Behaviors
20
Discussion
23
Conclusions
Survey
15
Learners'
24
References
2
13
25
on
Pre-Processing Educational Data
29
Cristobal Romero, Jose Raul Romero and Sebastian Ventura
2.1
2.2
Introduction
Types
30
of Educational Environments
31
2.2.1
Learning Management Systems
32
2.2.2
Massive
Online Courses
Open
Intelligent Tutoring Systems
32
2.2.3
2.2.4
Adaptive
33
2.2.5
Test and
2.2.6
Other
33
Intelligent Hypermedia Systems
Quiz Systems
and
Types
of Educational
Systems
33
34
ix
Contents
x
2.3
2.4
2.5
Types
of Data
Relational Data
34
2.3.2
Transactional Data
35
2.3.3
Temporal, Sequence
2.3.4
Text Data
37
2.3.5
Multimedia Data
37
2.3.6
World Wide Web Data
38
Pre-Processing
and Time Series Data
Tasks
40
Data
40
2.4.2
Data
43
2.4.3
Gathering
Aggregation/Integration
Data Cleaning
45
2.4.4
User and Session Identification
47
2.4.5
Attribute/Variable Selection
48
2.4.6
Data
50
2.4.7
Data Transformation
Filtering
Pre-Processing
2.5.2
52
Tools
General
Purpose
Specific Purpose
56
Data
Data
Pre-Processing
Pre-Processing
Tools
56
Tools
57
Conclusions
58
References
3
36
2.4.1
2.5.1
2.6
34
2.3.1
59
How Educational Data
Mining Empowers State
Study
Policies
to Reform Education: The Mexican Case
Alejandro Pefia-Ayala
65
and Leonor Cardenas
3.1
Introduction
3.2
Domain
Study
3.2.1
A Glance at Data
3.2.2
Educational Data
66
68
Mining
Mining
68
in
a
Nutshell
69
3.3
Related Works
70
3.4
Context
71
3.5
3.4.1
The Mexican State
72
3.4.2
Educational
72
3.4.3
National Assessments
74
3.4.4
The Constitutional Reform in Education
75
3.4.5
Community
3.7
Reaction
76
Source Data
3.5.1
EXCALE Databases
3.5.2
Source Data Students'
3.5.3
Framework
3.5.4
3.6
Community
77
77
Opinions
78
78
Exploration Analysis
Educational Data Mining Approach
3.6.1
Essential Mining
3.6.2
Supplementary Mining
80
Discussion
90
82
82
87
Contents
xi
3.7.1
3.7.2
3.7.3
3.8
Interpretations of the Basic Findings
Interpretation of Supplementary Mining
A Diagnostic of Students Opinions
90
92
93
Conclusions
96
References
Part II
4
Student
Modeling
Modeling Student Performance in Higher Education Using
Data
Mining
Huseyin
105
Guruler and
Ayhan Istanbullu
4.1
Introduction
106
4.2
Background
109
4.2.1
The Decision Tree Classification Model
110
4.2.2
The Decision Tree Mechanism
Ill
4.3
System Overview, Software
4.4
Case
4.4.1
4.4.2
4.4.3
5
97
Interface and Architecture
Study: Modeling Student
Data Description
112
Performance
115
115
Data
116
Preparation
Analyzer Model
117
4.5
Discussion of Results
117
4.6
Conclusions
121
References
122
Using Data Mining Techniques to Detect the Personality
of Players in an Educational Game
Fazel Keshtkar, Candice Burkett, Haiying Li
125
and Arthur C. Graesser
5.1
Introduction
5.2
Literature Review
5.2.1
126
Personality
128
in
Computer-Based Learning
Environments
5.2.2
128
Emotion Detection
Using Leary's
Rose Frameboard
5.2.3
5.2.4
5.2.5
128
Automatic Detection of
Personality
128
and Student Behavior
Personality
Relationship
The
Between
Personality
129
Traits
and Information
5.2.6
Competency
Personality Traits and Learning Style
129
in Academic
Performance
130
5.2.7
A Neural Network Model for Human
5.2.8
Relationships
and
Personality
...
130
Between Academic Motivation
Personality Among
the Students
130
Contents
xjj
5.2.9
from Errors
Learning
Relation Between
131
Personality
and
5.2.10
Academic Achievement and
5.2.11
The
Big
Five Model
131
Big
Personality, Learning Styles,
Five
132
and Academic Achievement
5.2.12
and
Using Personality
Cognitive Ability
to Predict
132
Academic Achievement
5.3
Leary's Interpersonal
5.3.1
5.3.2
5.4
135
135
137
5.5.1
5.5.2
Feature Extraction
5.5.5
5.7.2
138
Linguistic Inquiry
Automated Approaches
and Word Count Features
...
Personality
to
140
Classification Method
141
143
and Results
143
Classification Results
145
Analysis
Trait
145
Tracking Analysis
Personality
ANOVA Analysis
146
148
Conclusion and Future Research
149
References
6
139
Classification
Discussion and
5.7.1
138
The
Experience
5.6.1
5.8
134
Human Annotation
Lexicon Resources
5.5.4
5.7
Participants
and Data Set Construction
Model
5.5.3
5.6
133
Land Science Game
Annotation Scheme
5.4.1
5.5
132
Frame Board
Students' Performance Prediction
Using
Multi-Channel
151
Decision Fusion
H. Moradi, S. Abbas Moradi and L. Kashani
6.1
Introduction
152
6.2
Student
153
6.3
Performance Prediction
6.4
Modeling
156
6.3.1
Performance Prediction in ITS
6.3.2
Data
Mining Approaches
156
for Prediction
157
Multi-Channel Decision Fusion Performance Prediction
6.4.1
Determining
the Performance Level in
....
158
Assignment
Categories
159
161
6.4.3
Determining Overall Performance Levels
Mapping from the Performance in Assignment
6.4.4
The Characteristics of
6.4.2
Categories
to
Overall Performance
6.5
Experimental
6.6
Conclusion and Future Work
References
Assignment Categories
Results and Discussion
162
163
164
172
173
Contents
7
xiii
Predicting Student
Performance from Combined
Data Sources
175
Annika Wolff, Zdenek Zdrahal, Drahomira Herrmannova
and Petr Knoth
7.1
Introduction
7.2
Defining the
7.3
176
Problem
177
7.2.1
Problem
7.2.2
Problem
Specification
Specification
1
178
2
7.2.3
179
Problem
Specification 3
7.2.4
179
Problem
Specification
4
179
Sources of Student Data
7.3.1
Student
179
Data from the Virtual
Activity
Learning
Environment
180
7.3.2
Demographic
7.3.3
Past
7.3.4
Assessment Data
Data
180
181
Study
181
7.4
Feature Selection and
181
7.5
Classifiers for
184
Data-Filtering
Predicting Student Outcome
7.5.1
Support
7.5.2
General
Vector Machines and Decision Trees
7.5.3
7.6
Evaluation Framework
7.7
Real-Time Prediction
7.8
Revisiting
7.9
Specification in Light
7.8.1
Problem
Specification
7.8.2
Problem
7.8.3
Problem
Specification 5
Specification 6
Developing
7.11
Beyond
OU:
Testing
of Results
....
4 (Revised)
Models
196
197
198
199
on
Open University
Data
199
Applying
Models
on
Alternative
Data Sources
200
Conclusions
201
References
8
189
196
the Problem
and
188
192
(A Case Study)
7.10
184
Unary Hypotheses Automaton
Bayesian Networks and Regression
201
Predicting Learner Answers Correctness Through Eye
Movements with Random Forest
Alper Bayazit,
Petek Askar and Erdal
203
Cosgun
8.1
Introduction
204
8.2
Background
205
8.3
8.2.1
Related Work
205
8.2.2
Cognitive
206
Processes
8.2.3
Eye
8.2.4
Random Forest
Method
Movement Data
207
209
210
Contents
xjv
8.3.1
The
8.3.5
Purpose of
Design
Pre-Application
Application
Study Group
8.3.6
Data Collection Instruments
8.3.2
8.3.3
8.3.4
210
Study
210
211
211
211
212
8.4
Analyses
8.5
Conclusion and Discussion
218
8.6
Future Work
220
of Results
214
Appendix: Supplementary
220
References
225
Part III
9
the
Assessment
Mining Domain Knowledge for Coherence Assessment
of Students Proposal Drafts
Samuel Gonzalez Lopez and Aurelio Lopez-Lopez
9.1
Introduction
230
9.2
Background
231
9.2.1
Global Coherence
231
9.2.2
Latent Semantic
232
9.2.3
Related Work
9.3
Analyzer
9.4
Data
9.5
9.6
9.7
9.8
Analysis
232
Model of Global Coherence
233
Description (Corpus)
Experiments
9.5.1
Experimental Design
236
9.5.2
Agreement
238
9.5.3
Across Section
Analysis
9.6.1
236
237
Evaluation
247
Exploration
and Discussion of Results
Across Section
248
Exploration
248
System Overview
249
9.7.1
Intelligent Tutoring System
249
9.7.2
Web Interface
251
Conclusions
252
References
10
229
Adaptive Testing
Educational Data
253
in
Programming Courses
Mining Techniques
Based
on
257
Vladimir Ivancevic, Marko Knezevic, Bojan Pusic
and Ivan Lukovic
10.1
Introduction
259
10.2
Related Work
261
10.3
Background
263
Contents
10.4
10.5
xv
10.3.1
Environment
263
10.3.2
Data Set
264
Modeling Programming Knowledge
10.4.1
Programming Knowledge Overview
10.4.2
Modeling Programming Competencies
10.4.3
Modeling Programming Concepts
of the C Language
Estimating Test Difficulty
10.5.1
Estimating Test Item Difficulty
10.5.2
10.6
Capacity
Estimating
Algorithm
Student
Test Generation
Application
10.8
Conclusion and Future Work
Plan
267
268
269
270
271
276
279
282
and Results
10.7
284
285
References
11
264
Recognition and Visualization
Learning Environments
Ofra Amir, Kobi Gal, David Yaron,
in
Exploratory
289
Michael Karabinos
and Robert Belford
290
11.1
Introduction
11.2
Related Work
291
11.2.1
Plan
292
11.2.2
Assessment of Students' Activities
Recognition
11.3
The Virtual Labs Domain
11.4
Plan
Recognition
11.6
11.7
11.8
296
Actions, Recipes, and Plans
296
11.4.2
The Plan
299
11.4.3
Empirical Methodology
Complete Algorithms
302
Visualizing Students'Activities
309
Recognition Algorithm
304
11.5.1
Visualization Methods
311
11.5.2
Empirical Methodology
314
11.5.3
Results
316
11.5.4
Discussion
319
Conclusion and Future Work
Experimental
The Recipe Library for
320
321
Problems
11.8.1
11.9
294
in Virtual Laboratories
11.4.1
11.4.4
11.5
293
the Dilution Problem
Dilution Problem
Recipes
323
323
11.8.2
324
User
324
References
Recipes Explanation
Study Questionnaire
325
Contents
xvj
12
Finding Dependency
of Test Items from Students'
329
Response Data
Xiaoxun Sun
330
12.1
Introduction
12.2
Related Work
330
Mutual Independency
12.3
12.3.1
Preliminaries
331
12.3.2
Mutual Information Measure
332
12.3.3
Finding the
12.3.4
An
12.3.5
Extensions
12.5
Best
Dependency
333
Tree
335
Example
Proof-of-Concept
12.4
331
Measure
337
337
Experiments
337
12.4.1
Data
12.4.2
Results on
Synthetic
12.4.3
Results
Real Data
340
Conclusions and Future Work
341
on
338
Data Sets
341
References
Part IV
13
Trends
Mining Texts,
Learner Productions and
Strategies
345
with ReaderBench
Mihai Dascalu, Philippe
Dessus,
Maryse Bianco,
Stefan Trausan-Matu and Aurelie
Nardy
346
13.1
Introduction
13.2
Applications
13.2.1
Comprehension
Predicting
Textual Complexity Assessment for Comprehension
Data and Text
Mining
for Educational
Learner
13.3
Prediction
13.3.1
Reading Strategies
Self-Explanations for Comprehension
Assessment
13.4
Cohesion-Based Discourse
the Cohesion
355
Extraction
13.6
Cohesion-Based Scoring Mechanism of the
Analysis
359
Elements
Reading Strategies
Assessing Textual
13.7
Identification Heuristics for
13.8
Multi-Dimensional Model for
364
Results
Comparison
13.11
Conclusions
References
360
363
Complexity
A
350
352
Graph
Topics
13.10
348
Analysis: Building
13.5
13.9
348
Extracted
The Impact of
from
347
of ReaderBench with Previous Work
370
372
373
xvii
Contents
14
Maximizing the Value of Student Ratings Through
Data Mining
Kathryn Gates, Dawn Wilkins, Sumali Conlon, Susan Mossing
379
and Maurice Eftink
14.1
Introduction
14.2
Description
14.2.1
380
of the Data Set
and
14.2.2
14.3
Collecting
Presenting Results
Evaluations
382
383
Details About the Data Set
14.2.3
Questions
14.2.4
Selected Results from the Statistical
The
383
and Variables of Interest
Analysis
A
385
View of the Process
High-Level
386
14.3.4
Analysis
Corpus
Category Selection
The Domain-Specific Lexicon
14.3.5
The Assessment Process
389
14.3.6
Refining
14.3.7
The
14.3.2
14.3.3
Word
387
388
the Lexicon
392
393
Algorithm
397
Assessment Results
14.4.1
14.4.2
Qualitative Validity
Scores by Teaching
Summary
Performance
14.4.3
Assessment of
Category
and Learning Specialists
Quantitative Assessment Through the Comparison
of
398
Ratings
Quantitative
Assessment
of
and
Category
Through
Summary
the
Scores for
Comparison
Teaching
Evaluation of Instruction at the
of
14.5.2
14.6
403
404
Applications of the Methodology
14.5.1
University
404
Mississippi
Other Educational
405
Applications
407
Future Work
409
References
15
Data
Mining
397
Scores with Overall Instructor
Award Winners with All Instructors
14.5
384
385
Methodology
14.3.1
14.4
382
The Process for
and Social Network
Analysis in the
Educational
Field: An Application for Non-Expert Users
411
Diego Garci'a-Saiz, Camilo Palazuelos and Marta Zorrilla
15.1
15.2
412
Introduction
Background
15.2.1
15.2.2
414
and Related Work
Social Network Analysis
Classification
Applied
Students' Performance and
15.2.3
Data
Mining
Tools for
415
to the Educational Context:
Dropout
Non-Expert Users
417
418
Contents
xviii
15.3
15.4
420
15.3.3
New Services Provided
422
15.3.4
Mode of
423
Case
Working
426
Study
426
15.4.1
Courses
15.4.2
Social Network
15.4.3
Prediction of Students' Performance
Analysis
in
E-Learning
Courses.
Collaborative
.
.
428
435
436
Learning
of Students in Online Discussion
Forums: A Social Network
Reihaneh
.
432
Dropouts
Conclusions
References
16
419
Architecture
and
15.5
419
E-Learning Web Miner
15.3.1
Description of E-Learning Web Miner
General View of the E-Learning Web Miner
15.3.2
Rabbany,
Analysis Perspective
441
Samira Elatia, Mansoureh Takaffoli
and Osmar R. Zai'ane
16.1
Introduction
16.2
Background
16.2.1
16.3
16.4
and Related Works
444
16.2.2
Learning and E-Learning:
An Educational Perspective
Social Networks: A Data Mining Perspective
16.2.3
Social Network
On Collaborative
Analysis of
444
445
Online Educational
Forums: Related Works
448
Network Analysis in E-Learning
450
16.3.1
Students Interaction Network
450
16.3.2
Term Co-Occurrence Network
453
Case Studies
16.4.1
16.4.2
16.4.3
16.4.4
16.5
442
Extracting
Interpreting
Interpreting
Objective
Conclusions
References
Author Index
456
Networks
458
Students Interaction Network
459
Term Co-Occurrence Network
462
Evaluation
463
464
464
467