This book introduces Meaningful Purposive Interaction Analysis (MPIA) theory, which combines social network analysis (SNA) with latent semantic analysis (LSA) to help create and analyse a meaningful learning landscape from the digital traces left by a learning community in the co-construction of knowledge.     The hybrid algorithm is implemented in the statistical programming language and environment R, introducing packages which capture – through matrix algebra – elements of learners’ work with more knowledgeable others and resourceful content artefacts. The book provides comprehensive package-by-package application examples, and code samples that guide the reader through the MPIA model to show how the MPIA landscape can be constructed and the learner’s journey mapped and analysed. This building block application will allow the reader to progress to using and building analytics to guide students and support decision-making in learning.

lgli/K:\!genesis\!repository8\springer\10.1007%2F978-3-319-28791-1.pdf

lgrsnf/K:\!genesis\!repository8\springer\10.1007%2F978-3-319-28791-1.pdf

nexusstc/Learning Analytics in R with SNA, LSA, and MPIA/947eb128c0650e360a3fae36f18d3d18.pdf

scihub/10.1007/978-3-319-28791-1.pdf

zlib/Computers/Databases/Fridolin Wild (auth.)/Learning Analytics in R with SNA, LSA, and MPIA_2741699.pdf

Adobe InDesign CS6 (Windows)

Wild, Fridolin

Springer Nature Switzerland AG

Springer London, Limited

Springer Nature, Switzerland, 2016

Switzerland, Switzerland

Apr 05, 2016

2, 20160404

sm52632424

producers:
Adobe PDF Library 10.0.1

{"edition":"1","isbns":["3319287893","3319287915","9783319287898","9783319287911"],"publisher":"Springer"}

Acknowledgement 8
Contents 12
Application examples included 16
Chapter 1: Introduction 17
1.1 The Learning Software Market 18
1.2 Unresolved Challenges: Problem Statement 19
1.3 The Research Area of Technology-Enhanced Learning 21
1.4 Epistemic Roots 23
1.5 Algorithmic Roots 25
1.6 Application Area: Learning Analytics 26
1.7 Research Objectives and Organisation of This Book 30
References 33
Chapter 2: Learning Theory and Algorithmic Quality Characteristics 38
2.1 Learning from Purposive, Meaningful Interaction 40
2.2 Introducing Information 42
2.3 Foundations of Culturalist Information Theory 43
2.4 Introducing Competence 45
2.5 On Competence and Performance Demonstrations 47
2.6 Competence Development as Information Purpose of Learning 47
2.7 Performance Collections as Purposive Sets 49
2.8 Filtering by Purpose 50
2.9 Expertise Clusters 50
2.10 Assessment by Equivalence 51
2.11 About Disambiguation 52
2.12 A Note on Texts, Sets, and Vector Spaces 53
2.13 About Proximity as a Supplement for Equivalence 54
2.14 Feature Analysis as Introspection 54
2.15 Algorithmic Quality Characteristics 55
2.16 A Note on Educational Practice 56
2.17 Limitations 56
2.18 Summary 57
References 58
Chapter 3: Representing and Analysing Purposiveness with SNA 60
3.1 A Brief History and Standard Use Cases 60
3.2 A Foundational Example 63
3.3 Extended Social Network Analysis Example 71
3.4 Limitations 83
References 84
Chapter 4: Representing and Analysing Meaning with LSA 86
4.1 Mathematical Foundations 88
4.2 Analysis Workflow with the R Package `lsa ́ 92
4.3 Foundational Example 95
4.4 State of the Art in the Application of LSA for TEL 106
4.5 Extended Application Example: Automated Essay Scoring 110
4.6 Limitations of Latent Semantic Analysis 116
References 116
Chapter 5: Meaningful, Purposive Interaction Analysis 122
5.1 Fundamental Matrix Theorem on Orthogonality 123
5.2 Solving the Eigenvalue Problem 127
5.3 Example with Incidence Matrices 129
5.4 Singular Value Decomposition 132
5.5 Stretch-Dependent Truncation 137
5.6 Updating Using Ex Post Projection 140
5.7 Proximity and Identity 141
5.8 A Note on Compositionality: The Difference of Point, Centroid, and Pathway 143
5.9 Performance Collections and Expertise Clusters 144
5.10 Summary 145
References 146
Chapter 6: Visual Analytics Using Vector Maps as Projection Surfaces 147
6.1 Proximity-Driven Link Erosion 149
6.2 Planar Projection With Monotonic Convergence 150
6.3 Kernel Smoothing 155
6.4 Spline Tile Colouring With Hypsometric Tints 156
6.5 Location, Position, and Pathway Revisited 159
6.6 Summary 160
References 161
Chapter 7: Calibrating for Specific Domains 163
7.1 Sampling Model 164
7.2 Investigation 166
7.3 Results 169
7.4 Discussion 172
7.5 Summary 176
References 177
Chapter 8: Implementation: The MPIA Package 178
8.1 Use Cases for the Analyst 178
8.2 Analysis Workflow 180
8.3 Implementation: Classes of the mpia Package 183
8.3.1 The DomainManager 187
8.3.2 The Domain 188
8.3.3 The Visualiser 188
8.3.4 The HumanResourceManager 189
8.3.5 The Person 189
8.3.6 The Performance 190
8.3.7 The Generic Functions 190
8.4 Summary 194
References 194
Chapter 9: MPIA in Action: Example Learning Analytics 195
9.1 Brief Review of the State of the Art in Learning Analytics 196
9.2 The Foundational SNA and LSA Examples Revisited 197
9.3 Revisiting Automated Essay Scoring: Positioning 208
9.4 Learner Trajectories in an Essay Space 224
9.5 Summary 233
References 234
Chapter 10: Evaluation 235
10.1 Insights from Earlier Prototypes 237
10.2 Verification 242
10.3 Validation 243
10.3.1 Scoring Accuracy 244
10.3.2 Structural Integrity of Spaces 246
10.3.3 Annotation Accuracy 250
10.3.4 Visual (in-)Accuracy 252
10.3.5 Performance Gains 253
10.4 Summary and Limitations 254
References 255
Chapter 11: Conclusion and Outlook 258
11.1 Achievement of Research Objectives 259
11.1.1 Objective 1: Represent Learning 260
11.1.2 Objective 2: Provide Instruments for Analysis 261
11.1.3 Objective 3: Re-represent to the User 263
11.1.4 Summary 267
11.2 Open Points for Future Research 269
11.3 Connections to Other Areas in TEL Research 271
11.4 Concluding Remarks 274
References 274
Erratum to: Learning Analytics in R with SNA, LSA, and MPIA 276
Annex A: Classes and Methods of the mpia Package 277
A.1 Class `DomainManager ́ 277
A.2 Class `Domain ́ 278
A.3 Class `Visualiser ́ 280
A.4 Class `HumanResourceManager ́ 282
A.5 Class `Person ́ 283
A.6 Class `Performance ́ 286

Front Matter....Pages i-xv
Introduction....Pages 1-21
Learning Theory and Algorithmic Quality Characteristics....Pages 23-44
Representing and Analysing Purposiveness with SNA....Pages 45-70
Representing and Analysing Meaning with LSA....Pages 71-106
Meaningful, Purposive Interaction Analysis....Pages 107-131
Visual Analytics Using Vector Maps as Projection Surfaces....Pages 133-148
Calibrating for Specific Domains....Pages 149-163
Implementation: The MPIA Package....Pages 165-181
MPIA in Action: Example Learning Analytics....Pages 183-222
Evaluation....Pages 223-245
Conclusion and Outlook....Pages 247-264
Erratum to: Learning Analytics in R with SNA, LSA, and MPIA....Pages E1-E1
Back Matter....Pages 265-275

This text introduces Meaningful Purposive Interaction Analysis (MPIA) theory, which combines social network analysis (SNA) with latent semantic analysis (LSA) to help create and analyse a meaningful learning landscape from the digital traces left by a learning community in the co-construction of knowledge. The hybrid algorithm is implemented in the statistical programming language and environment R, introducing packages which capture - through matrix algebra - elements of learners' work with more knowledgeable others and resourceful content artefacts

2016-07-20