第1页 / 共477页
第2页 / 共477页
第3页 / 共477页
第4页 / 共477页
第5页 / 共477页
第6页 / 共477页
第7页 / 共477页
第8页 / 共477页
Educational Data Mining: Applications and Trends.pdf
Preface
Contents
Part I Profile
1 Which Contribution Does EDM Provide to Computer-Based Learning Environments?
Abstract
1.1…Introduction
1.2…Educational Data Mining
1.2.1 Definition
1.2.2 Areas in Relation to EDM
1.2.3 Objectives of the EDM
1.2.4 The Used Methods
1.2.5 The Analyzed Data
1.2.6 Process of Applying the EDM
1.2.7 Some Technological Tools Used in EDM
1.3…Examples of EDM Applications in Computer-Based Learning Environments
1.3.1 EDM Applications for Predicting and Evaluating Learning Performance
1.3.2 EDM Applications for Analyzing Learners’ Behaviors
1.3.3 Discussion
1.4…Conclusions
References
2 A Survey on Pre-Processing Educational Data
Abstract
2.1…Introduction
2.2…Types of Educational Environments
2.2.1 Learning Management Systems
2.2.2 Massive Open Online Courses
2.2.3 Intelligent Tutoring Systems
2.2.4 Adaptive and Intelligent Hypermedia Systems
2.2.5 Test and Quiz Systems
2.2.6 Other Types of Educational Systems
2.3…Types of Data
2.3.1 Relational Data
2.3.2 Transactional Data
2.3.3 Temporal, Sequence and Time Series Data
2.3.4 Text Data
2.3.5 Multimedia Data
2.3.6 World Wide Web Data
2.4…Pre-Processing Tasks
2.4.1 Data Gathering
2.4.2 Data Aggregation/Integration
2.4.3 Data Cleaning
2.4.4 User and Session Identification
2.4.5 Attribute/Variable Selection
2.4.6 Data Filtering
2.4.7 Data Transformation
2.5…Pre-Processing Tools
2.5.1 General Purpose Data Pre-Processing Tools
2.5.2 Specific Purpose Data Pre-Processing Tools
2.6…Conclusions
Acknowledgments
References
3 How Educational Data Mining Empowers State Policies to Reform Education: The Mexican Case Study
Abstract
3.1…Introduction
3.2…Domain Study
3.2.1 A Glance at Data Mining
3.2.2 Educational Data Mining in a Nutshell
3.3…Related Works
3.4…Context
3.4.1 The Mexican State
3.4.2 Educational Community
3.4.3 National Assessments
3.4.4 The Constitutional Reform in Education
3.4.5 Community Reaction
3.5…Source Data
3.5.1 EXCALE Databases
3.5.2 Source Data Students’ Opinions
3.5.3 Framework
3.5.4 Exploration Analysis
3.6…Educational Data Mining Approach
3.6.1 Essential Mining
3.6.2 Supplementary Mining
3.7…Discussion
3.7.1 Interpretations of the Basic Findings
3.7.2 Interpretation of Supplementary Mining
3.7.3 A Diagnostic of Students Opinions
3.8…Conclusions
Acknowledgments
References
Part II Student Modeling
4 Modeling Student Performance in Higher Education Using Data Mining
Abstract
4.1…Introduction
4.2…Background
4.2.1 The Decision Tree Classification Model
4.2.2 The Decision Tree Mechanism
4.3…System Overview, Software Interface and Architecture
4.4…Case Study: Modeling Student Performance
4.4.1 Data Description
4.4.2 Data Preparation
4.4.3 Analyzer Model
4.5…Discussion of Results
4.6…Conclusions
References
5 Using Data Mining Techniques to Detect the Personality of Players in an Educational Game
Abstract
5.1…Introduction
5.2…Literature Review
5.2.1 Personality in Computer-Based Learning Environments
5.2.2 Emotion Detection Using Leary’s Rose Frameboard
5.2.3 Automatic Detection of Personality
5.2.4 Personality and Student Behavior
5.2.5 The Relationship Between Personality Traits and Information Competency
5.2.6 Personality Traits and Learning Style in Academic Performance
5.2.7 A Neural Network Model for Human Personality
5.2.8 Relationships Between Academic Motivation and Personality Among the Students
5.2.9 Relation Between Learning from Errors and Personality
5.2.10 Academic Achievement and Big Five Model
5.2.11 The Big Five Personality, Learning Styles, and Academic Achievement
5.2.12 Using Personality and Cognitive Ability to Predict Academic Achievement
5.3…Leary’s Interpersonal Frame Board
5.3.1 Land Science Game
5.3.2 Participants and Data Set Construction
5.4…Annotation Scheme
5.4.1 Human Annotation
5.5…Model
5.5.1 Lexicon Resources
5.5.2 Feature Extraction
5.5.3 The Linguistic Inquiry and Word Count Features
5.5.4 Automated Approaches to Personality Classification
5.5.5 Classification Method
5.6…Experience and Results
5.6.1 Classification Results
5.7…Discussion and Analysis
5.7.1 Personality Trait Tracking Analysis
5.7.2 ANOVA Analysis
5.8…Conclusion and Future Research
Acknowledgments
References
6 Students’ Performance Prediction Using Multi-Channel Decision Fusion
Abstract
6.1…Introduction
6.2…Student Modeling
6.3…Performance Prediction
6.3.1 Performance Prediction in ITS
6.3.2 Data Mining Approaches for Prediction
6.4…Multi-Channel Decision Fusion Performance Prediction
6.4.1 Determining the Performance Level in Assignment Categories
6.4.2 Determining Overall Performance Levels
6.4.3 Mapping from the Performance in Assignment Categories to Overall Performance
6.4.4 The Characteristics of Assignment Categories
6.5…Experimental Results and Discussion
6.6…Conclusion and Future Work
Acknowledgements
References
7 Predicting Student Performance from Combined Data Sources
Abstract
7.1…Introduction
7.2…Defining the Problem
7.2.1 Problem Specification 1
7.2.2 Problem Specification 2
7.2.3 Problem Specification 3
7.2.4 Problem Specification 4
7.3…Sources of Student Data
7.3.1 Student Activity Data from the Virtual Learning Environment
7.3.2 Demographic Data
7.3.3 Past Study
7.3.4 Assessment Data
7.4…Feature Selection and Data-Filtering
7.5…Classifiers for Predicting Student Outcome
7.5.1 Support Vector Machines and Decision Trees
7.5.2 General Unary Hypotheses Automaton
7.5.3 Bayesian Networks and Regression
7.6…Evaluation Framework
7.7…Real-Time Prediction
7.8…Revisiting the Problem Specification in Light of Results
7.8.1 Problem Specification 4 (Revised)
7.8.2 Problem Specification 5
7.8.3 Problem Specification 6
7.9…Developing and Testing Models on Open University Data (A Case Study)
7.10…Beyond OU: Applying Models on Alternative Data Sources
7.11…Conclusions
Acknowledgments
References
8 Predicting Learner Answers Correctness Through Eye Movements with Random Forest
Abstract
8.1…Introduction
8.2…Background
8.2.1 Related Work
8.2.2 Cognitive Processes
8.2.3 Eye Movement Data
8.2.4 Random Forest
8.3…Method
8.3.1 The Purpose of the Study
8.3.2 Design
8.3.3 Pre-Application
8.3.4 Application
8.3.5 Study Group
8.3.6 Data Collection Instruments
8.4…Analyses of Results
8.5…Conclusion and Discussion
8.6…Future Work
A.1. Appendix: Supplementary
A.1.1 Question-1
A.1.2 Question-2
A.1.3 Question-3
A.1.4 Question-4
A.1.5 Question-5
A.1.6 Question-6
A.1.7 Question-7
A.1.8 Question-8
A.1.9 Question-9
References
Part III Assessment
9 Mining Domain Knowledge for Coherence Assessment of Students Proposal Drafts
Abstract
9.1…Introduction
9.2…Background
9.2.1 Global Coherence
9.2.2 Latent Semantic Analysis
9.2.3 Related Work
9.3…Analyzer Model of Global Coherence
9.4…Data Description (Corpus)
9.5…Experiments
9.5.1 Experimental Design
9.5.2 Agreement Evaluation
9.5.3 Across Section Exploration
9.6…Analysis and Discussion of Results
9.6.1 Across Section Exploration
9.7…System Overview
9.7.1 Intelligent Tutoring System
9.7.2 Web Interface
9.8…Conclusions
Acknowledgments
References
10 Adaptive Testing in Programming Courses Based on Educational Data Mining Techniques
Abstract
10.1…Introduction
10.2…Related Work
10.3…Background
10.3.1 Environment
10.3.2 Data Set
10.4…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
10.5…Estimating Test Difficulty
10.5.1 Estimating Test Item Difficulty
10.5.2 Estimating Student Capacity
10.6…Test Generation Algorithm
10.7…Application and Results
10.8…Conclusion and Future Work
Acknowledgments
References
11 Plan Recognition and Visualization in Exploratory Learning Environments
Abstract
11.1…Introduction
11.2…Related Work
11.2.1 Plan Recognition
11.2.2 Assessment of Students’ Activities
11.3…The Virtual Labs Domain
11.4…Plan Recognition in Virtual Laboratories
11.4.1 Actions, Recipes, and Plans
11.4.2 The Plan Recognition Algorithm
11.4.3 Empirical Methodology
11.4.4 Complete Algorithms
11.5…Visualizing Students’ Activities
11.5.1 Visualization Methods
11.5.2 Empirical Methodology
11.5.3 Results
11.5.4 Discussion
11.6…Conclusion and Future Work
11.7…Experimental Problems
11.8…The Recipe Library for the Dilution Problem
11.8.1 Dilution Problem Recipes
11.8.2 Recipes Explanation
11.9…User Study Questionnaire
References
12 Finding Dependency of Test Items from Students’ Response Data
Abstract
12.1…Introduction
12.2…Related Work
12.3…Mutual Independency Measure
12.3.1 Preliminaries
12.3.2 Mutual Information Measure
12.3.3 Finding the Best Dependency Tree
12.3.4 An Example
12.3.5 Extensions
12.4…Proof-of-Concept Experiments
12.4.1 Data
12.4.2 Results on Synthetic Data Sets
12.4.3 Results on Real Data
12.5…Conclusions and Future Work
Acknowledgment
References
Part IV Trends
13 Mining Texts, Learner Productions and Strategies with ReaderBench
Abstract
13.1…Introduction
13.2…Data and Text Mining for Educational Applications
13.2.1 Predicting Learner Comprehension
13.3…Textual Complexity Assessment for Comprehension Prediction
13.3.1 The Impact of Reading Strategies Extracted from Self-Explanations for Comprehension Assessment
13.4…Cohesion-Based Discourse Analysis: Building the Cohesion Graph
13.5…Topics Extraction
13.6…Cohesion-Based Scoring Mechanism of the Analysis Elements
13.7…Identification Heuristics for Reading Strategies
13.8…Multi-Dimensional Model for Assessing Textual Complexity
13.9…Results
13.10…A Comparison of ReaderBench with Previous Work
13.11…Conclusions
Acknowledgments
References
14 Maximizing the Value of Student Ratings Through Data Mining
Abstract
14.1…Introduction
14.2…Description of the Data Set
14.2.1 The Process for Collecting Evaluations and Presenting Results
14.2.2 Details About the Data Set
14.2.3 Questions and Variables of Interest
14.2.4 Selected Results from the Statistical Analysis
14.3…The Methodology
14.3.1 A High-Level View of the Process
14.3.2 Corpus Word Analysis
14.3.3 Category Selection
14.3.4 The Domain-Specific Lexicon
14.3.5 The Assessment Process
14.3.6 Refining the Lexicon
14.3.7 The Algorithm
14.4…Assessment Results
14.4.1 Qualitative Validity Assessment of Category Scores by Teaching and Learning Specialists
14.4.2 Quantitative Assessment Through the Comparison of Summary Scores with Overall Instructor Performance Ratings
14.4.3 Quantitative Assessment Through the Comparison of Category and Summary Scores for Teaching Award Winners with All Instructors
14.5…Applications of the Methodology
14.5.1 Evaluation of Instruction at the University of Mississippi
14.5.2 Other Educational Applications
14.6…Future Work
Acknowledgments
References
15 Data Mining and Social Network Analysis in the Educational Field: An Application for Non-Expert Users
Abstract
15.1…Introduction
15.2…Background and Related Work
15.2.1 Social Network Analysis
15.2.2 Classification Applied to the Educational Context: Students’ Performance and Dropout
15.2.3 Data Mining Tools for Non-expert Users
15.3…E-Learning Web Miner
15.3.1 Description of E-Learning Web Miner
15.3.2 General View of the E-Learning Web Miner Architecture
15.3.3 New Services Provided
15.3.4 Mode of Working
15.4…Case Study
15.4.1 Courses
15.4.2 Social Network Analysis in E-Learning Courses
15.4.3 Prediction of Students’ Performance and Dropouts
15.5…Conclusions
Acknowledgements
References
16 Collaborative Learning of Students in Online Discussion Forums: A Social Network Analysis Perspective
Abstract
16.1…Introduction
16.2…Background and Related Works
16.2.1 On Collaborative Learning and E-Learning: An Educational Perspective
16.2.2 Social Networks: A Data Mining Perspective
16.2.3 Social Network Analysis of Online Educational Forums: Related Works
16.3…Network Analysis in E-Learning
16.3.1 Students Interaction Network
16.3.2 Term Co-Occurrence Network
16.4…Case Studies
16.4.1 Extracting Networks
16.4.2 Interpreting Students Interaction Network
16.4.3 Interpreting Term Co-Occurrence Network
16.4.4 Objective Evaluation
16.5…Conclusions
References
Author Index
Studies in Computational Intelligence 524
Alejandro Peña-Ayala Editor
Educational Data
Mining
Applications and Trends
Studies in Computational Intelligence
Volume 524
Series Editor
Janusz Kacprzyk, Polish Academy of Sciences, Warsaw, Poland
e-mail: kacprzyk@ibspan.waw.pl
For further volumes:
http://www.springer.com/series/7092
About this Series
‘‘Studies
Intelligence’’
in Computational
The series
(SCI) publishes new
developments and advances in the various areas of computational intelligence—
quickly and with a high quality. The intent is to cover the theory, applications, and
design methods of computational
intelligence, as embedded in the fields of
engineering, computer science, physics and life sciences, as well as the
methodologies behind them. The series contains monographs, lecture notes and
edited volumes in computational
intelligence spanning the areas of neural
networks, connectionist systems, genetic algorithms, evolutionary computation,
artificial intelligence, cellular automata, self-organizing systems, soft computing,
fuzzy systems, and hybrid intelligent systems. Of particular value to both the
contributors and the readership are the short publication timeframe and the world-
wide distribution, which enable both wide and rapid dissemination of research
output.
Alejandro Peña-Ayala
Editor
Educational Data Mining
Applications and Trends
123
Editor
Alejandro Peña-Ayala
World Outreach Light to the
Nations Ministries
Escuela Superior de Ingeniería
Mecánica y Eléctrica, Zacatenco
Instituto Politécnico Nacional
Mexico City
Mexico
ISSN 1860-949X
ISBN 978-3-319-02737-1
DOI 10.1007/978-3-319-02738-8
Springer Cham Heidelberg New York Dordrecht London
ISSN 1860-9503 (electronic)
ISBN 978-3-319-02738-8
(eBook)
Library of Congress Control Number: 2013953247
Ó Springer International Publishing Switzerland 2014
This work is subject to copyright. All rights are reserved by the Publisher, whether the whole or part of
the material is concerned, specifically the rights of translation, reprinting, reuse of illustrations,
recitation, broadcasting, reproduction on microfilms or in any other physical way, and transmission or
information storage and retrieval, electronic adaptation, computer software, or by similar or dissimilar
methodology now known or hereafter developed. Exempted from this legal reservation are brief
excerpts in connection with reviews or scholarly analysis or material supplied specifically for the
purpose of being entered and executed on a computer system, for exclusive use by the purchaser of the
work. Duplication of this publication or parts thereof is permitted only under the provisions of
the Copyright Law of the Publisher’s location, in its current version, and permission for use must
always be obtained from Springer. Permissions for use may be obtained through RightsLink at the
Copyright Clearance Center. Violations are liable to prosecution under the respective Copyright Law.
The use of general descriptive names, registered names, trademarks, service marks, etc. in this
publication does not imply, even in the absence of a specific statement, that such names are exempt
from the relevant protective laws and regulations and therefore free for general use.
While the advice and information in this book are believed to be true and accurate at the date of
publication, neither the authors nor the editors nor the publisher can accept any legal responsibility for
any errors or omissions that may be made. The publisher makes no warranty, express or implied, with
respect to the material contained herein.
Printed on acid-free paper
Springer is part of Springer Science+Business Media (www.springer.com)
Preface
Educational Data Mining (EDM) is a new discipline based on the Data Mining
(DM) grounds (i.e., the baseline is composed of models, tasks, methods, and
algorithms) to explore data from educational settings to find out descriptive pat-
terns and predictions that characterize learners behaviors and achievements,
domain knowledge
and
applications.
functionalities,
assessments,
educational
content,
This book introduces concepts, models, frameworks,
tasks, methods, and
algorithms, as well as tools and case studies of the EDM field. The chapters make
up a sample of the work currently achieved in countries from the five continents,
which illustrates the world labor of the EDM arena. According to the nature of the
contributions accepted for this volume, four kinds of topics are identified as
follows:
• Profile shapes a conceptual view of the EDM. It provides an introduction of the
nature, purpose, components, processes, and applications. Through this section,
readers are encouraged to: make an incursion in the EDM field, facilitate the
extraction of source data to be mined, and acquire consciousness of the use-
fulness of this sort of approaches to support education policies.
• Student Modeling is an essential functionality of Computer-Based Educational
Systems (CBES) to adapt their performance according to users needs. Most of
the EDM approaches are oriented to characterize diverse student traits, such as:
behavior, acquired domain knowledge, personality, and academic achievements
by means of machine learning methods.
• Assessment evaluates learners’ domain knowledge acquisition, skills develop-
ment, and achieved outcomes, as well as reflection, inquiring, and sentiments
are essential subjects to be taken into account by CBES. The purpose is to
differentiate student proficiency at the finer grained level through static and
dynamic testing, as well as online and offline assessment.
• Trends focus on some of the new demands for applying EDM, such as text
mining and social networks analysis. Both targets represent challenges to cope
with huge, dynamical, and heterogeneous information that new generations of
students produce in their every day life. These paradigms represent new edu-
cational settings such as: ubiquitous-learning and educational networking.
v
vi
Preface
This volume is the result of one year of effort, where more than 30 chapters
were rigorous peer reviewed by a team of 60 reviewers. After several cycles of
chapter submission, revision, and tuning based on the Springer quality principles,
16 works were approved, edited as chapters, and organized according to the prior
four topics. So the Part I corresponds to Profile that includes Chaps. 1–3; the Part
II represents Student Modeling, which embraces Chaps. 4–8; the Part III concerns
Assessment and has Chaps. 9–12;
the Part IV is related to Trends through
Chaps. 13–16. A profile of the chapters is given next:
1. Chapter 1 provides a bibliographic review of studies made in the field of
Educational Data Mining (EDM) to identify diverse aspects related to tech-
niques and contributions in the field of computer-based learning. Authors
pursue to facilitate the use and understanding of Data Mining (DM) techniques
to help the educational specialists to develop EDM approaches.
2. Chapter 2 overcomes the lack of data preprocessing literature through the
detailed exposition of the tasks involved to extract, clean, transform, and
provide suitable data worthy to be mined. The work depicts educational
environments and data they offer; as well as gives examples of Moodle data
and tools.
3. Chapter 3 illustrates how EDM is able to support government policies devoted
to enhance education. The work shapes the context of basic education and how
the government aims at reforming the current practices of evaluation to aca-
demics and students. Several findings extracted from surveys are shown to
highlight the opinion of the community and provide an initial diagnostic.
4. Chapter 4 presents the Student Knowledge Discovery Software, a tool to
explore the factors having an impact on the student success based on student
profiling. Authors deeply outline how to implement the software to help
educational organizations
to better understand knowledge discovery
processes.
5. Chapter 5 explains how to automate the detection of student’s personality and
behavior in an educational game called Land Science. The work includes a
model to learn vector space representations for various extracted features.
Learner personality is detected by combining the features spaces from psy-
cholinguistics and computational linguistics.
6. Chapter 6 attempts to predict student performance to better adjust educational
materials and strategies throughout the learning process. Thus, authors design
a multichannel decision fusion approach to estimate the overall student per-
formance. Such an approach is based on the performance achieved in
assignment categories.
7. Chapter 7 explores predictive modeling methods for identifying students who
will most benefit from tutor interventions. Authors assert how the predictive
capacity of diverse sources of data changes as the course progresses, as well as
how a student’s pattern of behavior changes during the course.
8. Chapter 8 predicts learner achievements by recording learner eye movements
and mouse click counts. The findings claim: the most important eye metrics
Preface
vii
that predict answers in reasoning questions include total fixation duration,
number of mouse clicks, fixation count, and visit duration.
9. Chapter 9 focuses on coherence expressed in research protocols and thesis.
Authors develop a coherence analyzer that employs Latent Semantic Analysis
to mine domain knowledge. The analysis outcomes are used to grade students
and provide online support with the aim at improving their writings.
10. Chapter 10 tailors an approach to automatically generate tests. It recognizes
competence areas and matches the overall competence level of target students.
The approach makes use of a concept map of programming competencies and
a method for estimating the test item difficulty. The contribution is evaluated
in a setting where its results are compared against a solution that randomly
searches within the item space to find an adequate test.
11. Chapter 11 outlines methods oriented to support teachers’ understanding of
students’ activities on Exploratory Learning Environments (ELE). The work
includes an algorithm that
intelligently recognizes student activities and
visualization facilities for presenting these activities to teachers. The approach
is evaluated using real data obtained from students using an ELE to solve six
representative problems from introductory chemistry courses.
12. Chapter 12 adopts the concept of entropy from information theory to find the
most dependent test items in student responses. The work defines a distance
metric to estimate the amount of mutual independency between two items that
is used to quantify how independent two items are in a test. The trials show:
the approach for finding the best dependency tree is fast and scalable.
13. Chapter 13 proposes ReaderBench, an environment for assessing learner
productions and supporting teachers. It applies text mining to perform:
assessment of the reading materials, assignment of texts to learners, detection
of reading strategies, and comprehension evaluation to fostering learner’s self-
regulation process. All of these tasks were subject of empirical validations.
14. Chapter 14 analyzes a data set consisting of student narrative comments that
were collected using an online process. The approach uses category vectors to
depict instructor traits and a domain-specific lexicon. Sentiment analysis is
also used to detect and gauge attitudes embedded in comments about each
category. The approach is useful to instructors and administrators, and is a
vehicle to analyze student perceptions of teaching to feedback the educational
process.
15. Chapter 15 introduces E-learning Web Miner, a tool that assists academics to
discover student behavior profiles, models of how they collaborate, and their
performance with the purpose of enhancing the teaching-learning process. The
tool applies Social Network Analysis (SNA) and classification techniques.
16. Chapter 16 depicts an approach to assess the students’ participation by the
analysis of their interactions in social networks. It includes metrics for ranking
and determining roles to analyze the student communications, the forming of
groups, the role changes, and the interpretation of exchanged messages.
相关推荐
2023年江西萍乡中考道德与法治真题及答案.doc
2012年重庆南川中考生物真题及答案.doc
2013年江西师范大学地理学综合及文艺理论基础考研真题.doc
2020年四川甘孜小升初语文真题及答案I卷.doc
2020年注册岩土工程师专业基础考试真题及答案.doc
2023-2024学年福建省厦门市九年级上学期数学月考试题及答案.doc
2021-2022学年辽宁省沈阳市大东区九年级上学期语文期末试题及答案.doc
2022-2023学年北京东城区初三第一学期物理期末试卷及答案.doc
2018上半年江西教师资格初中地理学科知识与教学能力真题及答案.doc
2012年河北国家公务员申论考试真题及答案-省级.doc
2020-2021学年江苏省扬州市江都区邵樊片九年级上学期数学第一次质量检测试题及答案.doc
2022下半年黑龙江教师资格证中学综合素质真题及答案.doc
