Intelligent Support for Computer Science Education: Pedagogy Enhanced by Artificial Intelligence

Cover
Half Title
Title Page
Copyright Page
Dedication
Contents
About the Authors
Contributors
Preface
Acknowledgments
SECTION I: Four Scientific Pillars
	CHAPTER 1: Introduction
		1.1. AN INTERDISCIPLINARY PERSPECTIVE
		1.2. THE STRUCTURE OF THE BOOK
	CHAPTER 2: Related Work
		2.1. COGNITION AND MULTIPLE MODES OF LEARNING
			2.1.1. Background
			2.1.2. Nine Modes of Learning
				2.1.2.1. Discussion
		2.2. PRAGMATICS AND DIALOGUE PROCESSING
		2.3. INTRODUCTORY COMPUTER SCIENCE EDUCATION
			2.3.1. Elementary and Secondary Education
			2.3.2. From high school to college
			2.3.3. Post-Secondary Education for CS Majors
		2.4. INTELLIGENT TUTORING SYSTEMS (ITSS)
			2.4.1. Natural Language Processing (NLP) for ITSs
			2.4.2. Modes of Learning and ITSs
				2.4.2.1. Positive and Negative feedback
				2.4.2.2. Worked-Out Examples
				2.4.2.3. Analogy
		2.5. ITSS FOR COMPUTER SCIENCE AND NLP
			2.5.1. ITSs for CS
				2.5.1.1. NLP in ITSs for CS
SECTION II: From Human Tutoring to ChiQat-Tutor
	CHAPTER 3: Human Tutoring Dialogues and their Analysis
		3.1. DATA COLLECTION
			3.1.1. Learning Outcomes in Human Tutoring
			3.1.2. Measuring Learning Gains
			3.1.3. Learning E ects
		3.2. TRANSCRIPTION AND ANNOTATION
			3.2.1. Annotation
				3.2.1.1. Validating the Corpus Annotation
		3.3. DISTRIBUTIONAL ANALYSIS
			3.3.1. Elementary Dialogue Acts
			3.3.2. Student Initiative
			3.3.3. Episodic Strategies
		3.4. INSIGHTS FROM THE CORPUS: PEDAGOGICAL MOVES AND LEARNING
			3.4.1. Individual Dialog Acts (Type 1 Models)
			3.4.2. Sequences of Dialogue Acts (Type 2 Models)
				3.4.2.1. Bigram Models
				3.4.2.2. Trigram Models
			3.4.3. Episodic Strategies (Type 3 Models)
				3.4.3.1. Worked-Out Examples
				3.4.3.2. Analogies
		3.5. SUMMARY: INSIGHTS FROM HUMAN TUTORING ANALYSIS
	CHAPTER 4: ChiQat-Tutor and its Architecture
		4.1. THE DOMAIN MODEL
			4.1.1. Problem Definitions
			4.1.2. Solution Definitions
			4.1.3. Worked-Out Examples
			4.1.4. The Procedural Knowledge Model
		4.2. USER INTERFACE
		4.3. A BIRD'S EYE VIEW OF ChiQat-Tutor IN ACTION
			4.3.1. Solution Evaluator
		4.4. TUTOR MODULE
			4.4.1. Code Feedback: Syntax and Executability
			4.4.2. Reactive & Proactive Feedback
				4.4.2.1. Reactive Procedural Feedback
				4.4.2.2. Proactive Procedural Feedback
		4.5. TRAINING THE PKM GRAPHS
	CHAPTER 5: Evaluation in the Classroom
		5.1. EVALUATION METRICS
		5.2. LEARNING WITH PROACTIVE AND REACTIVE FEEDBACK
			5.2.1. Insights on Learning from Student Behavior and Perceptions of ChiQat-Tutor-v1
				5.2.1.1. Student Behavior
				5.2.1.2. Student Satisfaction
			5.2.2. Chiqat-Tutor, Version 1: Summary of Findings
		5.3. LEARNING WITH WORKED-OUT EXAMPLES AND ANALOGY
			5.3.1. WOE and Analogy Conditions
				5.3.1.1. Standard WOEs
				5.3.1.2. Length and Usage of WOEs
				5.3.1.3. Analogical Content in WOEs
			5.3.2. Learning Linked Lists among Non-Majors
			5.3.3. Learning Linked Lists Among Majors
			5.3.4. Learning and Initial Student Knowledge
				5.3.4.1. Mining the Logs: Predicting Initial Knowledge
			5.3.5. Chiqat-Tutor, Version 2: Summary of Findings
SECTION III: Extending ChiQat-Tutor
	CHAPTER 6: Beyond Linked Lists: Binary Search Trees and Re-cursion
		6.1. BINARY SEARCH TREES
			6.1.1. Pilot Evaluation
		6.2. RECURSION
			6.2.1. Models for Teaching Recursion
				6.2.1.1. Conceptual Models
				6.2.1.2. Program Visualization
			6.2.2. A Hybrid Model for Teaching Recursion in ChiQat-Tutor
			6.2.3. Evaluation of the Recursion Module
				6.2.3.1. Experimental Protocol
				6.2.3.2. Experiments at CMU Qatar
				6.2.3.3. Experiments at UIC
			6.2.4. Analysis of Students' Interactions with the System
		6.3. SUMMARY
	CHAPTER 7: A Practical Guide to Extending ChiQat-Tutor
		7.1. AN IMPLEMENTATION ARCHITECTURE
		7.2. CASE STUDY: THE STACK TUTOR PLUGIN
			7.2.1. Stack Plugin Design
			7.2.2. Class Structure
			7.2.3. Setting up the Stage
			7.2.4. Graphical Interface
			7.2.5. Stack Problem Logic and Feedback
	CHAPTER 8: Conclusions
		8.1. WHERE WE ARE, AND LESSONS LEARNED
		8.2. FUTURE WORK
			8.2.1. Extending the Curriculum
			8.2.2. Enhancing Communication with the Student
			8.2.3. Mining the User Logs, and Deep Learning
APPENDIX A: A Primer on Data Structures
	A.1. LINKED LISTS (LISTS)
	A.2. STACKS
	A.3. BINARY SEARCH TREES (BSTS)
APPENDIX B: Pre-/Post-Tests
	B.1. PRE-/POST-TEST FOR HUMAN TUTORING
	B.2. PRE-/POST-TEST FOR CHIQAT (LINKED LIST PROBLEMS)
APPENDIX C: Annotation Manuals
	C.1. DIALOGUE ACT MANUAL
		C.1.1. Direct Procedural Instruction: DPI
		C.1.2. Direct Declarative Instruction: DDI
			C.1.2.1. NOT DDI (NO!)
			C.1.2.2. DDI (YES!)
		C.1.3. Prompt
			C.1.3.1. Types of Prompts
		C.1.4. Feedback
			C.1.4.1. Positive Feedback
			C.1.4.2. Negative Feedback
			C.1.4.3. General Guidelines and Special Cases
	C.2. STUDENT INITIATIVE (SI)
	C.3. WORKED-OUT EXAMPLES
		C.3.1. Coding Categories
		C.3.2. Marking Worked-Out Examples
			C.3.2.1. Outline
			C.3.2.2. Examples
	C.4. ANALOGY CODING MANUAL
		C.4.1. Definition
		C.4.2. Analogous Terms
		C.4.3. Coding Category
		C.4.4. Marking Analogies
			C.4.4.1. Examples
APPENDIX D: Linked List Problem Set
	D.1. PROBLEM 1
	D.2. PROBLEM 2
	D.3. PROBLEM 3
	D.4. PROBLEM 4
	D.5. PROBLEM 5
	D.6. PROBLEM 6
	D.7. PROBLEM 7
APPENDIX E: Stack Plugin Full Code
	E.1. PLUGININSTANCE.JAVA
	E.2. STACKVIEW.JAVA
	E.3. STACKPROBLEM.JAVA
	E.4. STACKPROBLEMSTEP.JAVA
	E.5. STACKPROBLEMFEEDBACK.JAVA
	E.6. STACKSTYLE.CSS
Bibliography