Advances in Social Networking-based Learning: Machine Learning-based User Modelling and Sentiment Analysis (Intelligent Systems Reference Library, 181)

✍ Scribed by Christos Troussas, Maria Virvou

• Publisher: Springer
• Year: 2020
• Tongue: English
• Leaves: 185
• Edition: 1st ed. 2020
• Category: Library

✦ Synopsis

This book discusses three important, hot research issues: social networking-based learning, machine learning-based user modeling and sentiment analysis. Although these three technologies have been widely used by researchers around the globe by academic disciplines and by R&D departments in the IT industry, they have not yet been used extensively for the purposes of education. The authors present a novel approach that uses adaptive hypermedia in e-learning models to personalize educational content and learning resources based on the needs and preferences of individual learners. According to reports, in 2018 the vast majority of internet users worldwide are active on social networks, and the global average social network penetration rate as of 2018 is close to half the population. Employing social networking technologies in the field of education allows the latest technological advances to be used to create interactive educational environments where students can learn, collaborate with peers and communicate with tutors while benefiting from a social and pedagogical structure similar to a real class.

The book first discusses in detail the current trend of social networking-based learning. It then provides a novel framework that moves further away from digital learning technologies while incorporating a wide range of recent advances to provide solutions to future challenges. This approach incorporates machine learning to the student-modeling component, which also uses conceptual frameworks and pedagogical theories in order to further promote individualization and adaptivity in e-learning environments. Moreover, it examines error diagnosis, misconceptions, tailored testing and collaboration between students are examined and proposes new approaches for these modules. Sentiment analysis is also incorporated into the general framework, supporting personalized learning by considering the user’s emotional state, and creating a user-friendly learning environment tailored to students’ needs. Support for students, in the form of motivation, completes the framework. This book helps researchers in the field of knowledge-based software engineering to build more sophisticated personalized educational software, while retaining a high level of adaptivity and user-friendliness within human–computer interactions. Furthermore, it is a valuable resource for educators and software developers designing and implementing intelligent tutoring systems and adaptive educational hypermedia systems.

✦ Table of Contents

Foreword
Preface
Contents
1 Introduction
1.1 Current Topics
1.2 Social Networks as Learning Tools
1.2.1 Facebook
1.2.2 Google+
1.2.3 Twitter
1.2.4 Elgg
1.2.5 Edmodo
1.3 Comparative Analysis
1.3.1 Criteria
1.3.2 Results for Using SNs in Educational Contexts
1.4 Related Fields and Open Research Questions
References
2 Related Work
2.1 Social Media Language Learning
2.2 Related Literature for Social e-Learning
2.2.1 Methodology and Model Used
2.2.2 Selected Systems in the Review
2.2.3 Multicriteria Framework for Social e-Learning Systems
2.3 Comparative Discussion
References
3 Intelligent, Adaptive and Social e-Learning in POLYGLOT
3.1 Intelligent Tutoring Systems (ITSs)
3.1.1 Architecture of ITS
3.1.2 Function of ITSs
3.2 Intelligent Computer-Assisted Language Learning (ICALL)
3.3 User Modeling and Adaptivity
3.3.1 Student Models Characteristics
3.3.2 Using a Student Modeling in an ITS
3.4 Platforms for Social e-Learning
3.4.1 Research Method
3.4.2 Comparative Analysis and Discussion
References
4 Computer-Supported Collaborative Learning: A Novel Framework
4.1 Computer-Supported Collaborative Learning (CSCL): An Introduction
4.2 Precursor Theories
4.3 Collaboration Theory and Group Cognition
4.4 Strategies
4.5 Instructor Roles in CSCL
4.6 Effects
4.7 Applications of CSCL
4.8 CSCL for Foreign Language Acquisition
4.8.1 Effectiveness and Perception
4.9 Win-Win Collaboration Module
References
5 Affective Computing and Motivation in Educational Contexts: Data Pre-processing and Ensemble Learning
5.1 Affective Computing
5.1.1 Affective States
5.2 Frustration as an Affective State
5.2.1 Rosenweig’s Frustration Theory
5.2.2 Frustration Aggression Hypothesis
5.2.3 Frustration and Goal-Blockage
5.2.4 Frustration and Cause in Computer Users
5.3 Motivation Theory
5.3.1 Hull’s Drive Theory
5.3.2 Lewin’s Field Theory
5.3.3 Atkinson’s Theory of Achievement Motivation
5.3.4 Rotter’s Social Learning Theory
5.3.5 Attribution Theory
5.3.6 Discussion on Motivational Theories
5.4 Responding to Frustration
5.5 Pre-processing Techniques and Ensemble Classifiers for Sentiment Analysis Through Social Networks
5.5.1 Methodology
5.5.2 Twitter Datasets
5.5.3 Evaluation of Data Preprocessing Techniques
5.5.4 Evaluation of Stand-Alone Classifiers
5.5.5 Evaluation of Ensemble Classifiers
References
6 Blending Machine Learning with Krashen’s Theory and Felder-Silverman Model for Student Modeling
6.1 Employing the Stephen Krashen’s Theory of Second Language Acquisition in POLYGLOT
6.2 POLYGLOT Learning Content
6.3 POLYGLOT Student Model
6.3.1 Approximate String Matching for Error Diagnosis
6.3.2 String Meaning Similarity for Error Diagnosis
6.4 Automatic Detection of Learning Styles Based on Felder-Silverman Model Using the K-NN Algorithm
6.5 Tailored Assessments
6.5.1 Criteria for Tailored Assessment
6.5.2 Overview of the Building of the Adaptive Test Algorithm
References
7 Regression-Based Affect Recognition and Handling Using the Attribution Theory
7.1 Declaration and Handling of Affective States
7.2 Automatic Detection of Frustration
7.3 Rectilinear Regression Model to Detect Frustration
7.4 Incorporation of the Rectilinear Regression Model in POLYGLOT
7.5 Respond to Frustration
7.6 Delivery of Motivational Messages Based on the Attribution Theory
References
8 Overview of POLYGLOT Architecture and Implementation
8.1 POLYGLOT Architecture
8.2 POLYGLOT Implementation
References
9 Evaluation Results for POLYGLOT and Discussion
9.1 Evaluation Process and Framework Used
9.1.1 Criteria
9.1.2 Method
9.1.3 Population
9.2 Results
9.2.1 Satisfaction
9.2.2 Performance
9.2.3 Individual State of Learners
9.2.4 Students’ Progress
9.2.5 Validity of the Detection of the Students’ Learning Style
9.2.6 Validity of Win-Win Collaboration
9.3 Questionnaires
References
10 Conclusions
10.1 Conclusions and Discussion
10.2 Contribution to Science
10.2.1 Contribution to Intelligent Tutoring Systems
10.2.2 Contribution to Computer-Supported Collaborative Learning
10.2.3 Contribution to Student Modeling
10.2.4 Contribution to Computer-Assisted Language Learning
10.2.5 Contribution to Affective Computing
10.3 Future Work
References