Advances in Analytics and Applications (eBook)

Arnab Kumar Laha (Herausgeber)

eBook Download: PDF

2018 | 1st ed. 2019
XI, 297 Seiten
Springer Singapore (Verlag)
978-981-13-1208-3 (ISBN)

Infos

This book includes selected papers submitted to the ICADABAI-2017 conference, offering an overview of the new methodologies and presenting innovative applications that are of interest to both academicians and practitioners working in the area of analytics. It discusses predictive analytics applications, machine learning applications, human resource analytics, operations analytics, analytics in finance, methodology and econometric applications. The papers in the predictive analytics applications section discuss web analytics, email marketing, customer churn prediction, retail analytics and sports analytics. The section on machine learning applications then examines healthcare analytics, insurance analytics and machine analytics using different innovative machine learning techniques. Human resource analytics addresses important issues relating to talent acquisition and employability using analytics, while a paper in the section on operations analytics describe an innovative application in oil and gas industry. The papers in the analytics in finance part discuss the use of analytical tools in banking and commodity markets, and lastly the econometric applications part presents interesting banking and insurance applications.

Prof. Arnab K Laha takes keen interest in understanding how analytics, machine learning and artificial intelligence can be leveraged to solve complex problems of business and society.  His areas of research and teaching interest include Advanced Data Analytics, Quality Management and Risk Modeling.  He has published papers in national and international journals of repute in these areas and has served on the editorial board of several journals including Statistical Analysis and Data Mining: The ASA Data Science Journal. He was featured among India's best business school professors by Business Today in 2006 and Business India in 2012 and was named as one of the '10 Most Prominent Analytics Academicians in India' by Analytics India Magazine in 2014 and 2017. He is the convener of the biennial IIMA series of conferences on Advanced Data Analysis, Business Analytics and Intelligence. He is the author of the popular book on analytics entitled 'How to Make the Right Decision' published by Penguin-Random House.  He has conducted large number of training programmes and undertaken consultancy work in the fields of business analytics, quality management and risk management.

Inhalt

Preface 5
Contents 7
About the Editor 10
Brief Overviews 11
Machine Learning: An Introduction 12
1 Introduction 12
2 Supervised Learning 13
2.1 K-Nearest Neighbor (KNN) 14
2.2 Artificial Neural Networks (ANN) 14
2.3 Support Vector Machines (SVM) 15
2.4 Random Forest 16
3 Unsupervised Learning 17
3.1 Clustering 17
4 Deep Learning 18
5 Conclusion 19
References 19
Linear Regression for Predictive Analytics 21
1 Introduction 21
2 Linear Regression Model 22
3 Prediction Using Linear Regression Model 23
4 Hidden Extrapolation 23
5 Prediction Accuracy 24
6 Use of Validation and Test Data 24
7 Tracking Model Performance 26
References 27
Directional Data Analysis 28
1 Introduction 28
2 Descriptive Statistics on the Circle 29
3 Probability Models on the Circle 30
4 Inference on the Circle 32
5 Robustness with Circular Data 34
6 Conclusion 35
References 35
Branching Processes 38
1 Introduction and History 38
2 Simple Branching Process 39
3 Variants of Simple Branching Process 42
3.1 Bisexual Branching Processes 42
3.2 Varying Environments 43
3.3 Multi-type Branching Processes 44
4 Applications 47
References 48
Predictive Analytics Applications 49
Click-Through Rate Estimation Using CHAID Classification Tree Model 50
1 Introduction 50
1.1 Need for CTR 51
1.2 Factors Impacting CTR 52
1.3 Issues with CTR 54
1.4 Research Purpose 55
2 Literature Review 55
3 Methodology 58
4 Results 59
5 Conclusion 60
References 61
Predicting Success Probability in Professional Tennis Tournaments Using a Logistic Regression Model 64
1 Introduction 64
2 Literature Review 65
3 Model, Data and Results 66
4 Conclusion 70
References 70
Hausdorff Path Clustering and Hidden Markov Model Applied to Person Movement Prediction in Retail Spaces 71
1 Introduction 71
2 Data Description 71
3 Preliminaries 72
4 Part 1: Movement Prediction 72
4.1 Room Assignment 72
4.2 Movement Modeling Using Hidden Markov Models 74
5 Part 2: Path Clustering 75
6 Experimental Results 75
7 Part 2: Path Clustering (Experimental Results) 77
8 Experimental Results of Path Clustering Using Hausdorff Distance 78
9 Conclusions and Further Work 79
References 80
Improving Email Marketing Campaign Success Rate Using Personalization 81
1 Introduction 81
2 Language Modelling 83
3 Two-Step Personalization Process 85
3.1 Step 1 85
3.2 Step 2 85
4 Results and Future Scope 86
References 87
Predicting Customer Churn for DTH: Building Churn Score Card for DTH 88
1 Introduction and Motivation 89
2 Literature Survey 89
3 Dataset 90
4 Methodology 91
4.1 Sampling Strategy and Base Creation 91
4.2 Database Creation 92
4.3 Character Variable Generation 93
4.4 Segmentation Analysis 93
4.5 Modelling Methodology 94
5 Results and Findings 95
5.1 Comparison of Result at Population Level 95
5.2 Segment Level Model Result 97
6 Validation Result 100
6.1 Circlewise 100
6.2 Monthwise 100
7 Conclusion 101
Appendix 102
References 107
Applying Predictive Analytics in a Continuous Process Industry 108
1 Introduction and Literature Review 109
2 Data 111
3 Data Analysis and Results 111
4 Findings and Interpretation of Results 115
5 Conclusion 118
References 118
Machine Learning Applications 119
Automatic Detection of Tuberculosis Using Deep Learning Methods 120
1 Introduction 121
2 Related Work 122
3 Methodology 123
3.1 Data, Software and Hardware 123
3.2 TB Diagnosis 124
4 Results and Discussion 127
4.1 Results 127
4.2 Discussion 128
5 Conclusion 129
References 129
Connected Cars and Driving Pattern: An Analytical Approach to Risk-Based Insurance 131
1 Introduction 131
2 Our Approach 132
2.1 Data 132
2.2 Clustering 133
2.3 Cluster Labelling 133
2.4 Sampling 134
2.5 Feature Selection 134
2.6 Building Classifier 134
2.7 Analytical Results 135
2.8 Performance Monitoring 136
3 Conclusion 136
References 136
Human Resource Analytics 138
Analytics-Led Talent Acquisition for Improving Efficiency and Effectiveness 139
1 Introduction 140
2 Analytics for Talent Acquisition 141
3 Data Mining and Text Mining-Based Solution Components 144
3.1 Resume Information EXtractor (RINX) 144
3.2 RINX Search Engine (RINX SE) 149
3.3 Skill Similarity Computation 150
3.4 JD Extraction 151
3.5 JD Completion 154
4 Conclusion and Future Work 157
References 158
Assessing Student Employability to Help Recruiters Find the Right Candidates 159
1 Introduction 159
2 Research Objectives 160
3 Data Analysis 160
3.1 Approach 160
3.2 Data Collection 161
3.3 Assumptions, Constraints, and Mitigation 161
3.4 Exploratory Data Analysis 162
3.5 Data Modeling and Methodology 163
3.6 Results 167
4 Discussion 172
References 172
Operations Analytics 173
Estimation of Fluid Flow Rate and Mixture Composition 174
1 Introduction 174
2 Materials and Methods 175
2.1 Signal Description 175
2.2 Signal Filtering and Down-Sampling 175
2.3 Auto-regression Model: AR(1) Process 176
2.4 Hidden Markov Model 176
2.5 Mel-Frequency Cepstral Coefficients (MFCC) 177
2.6 Relating HMM and MFCC to Flow Rate Classification/Estimation 178
2.7 Choice of Hidden Markov Model Parameters and Training the HMM 178
2.8 Log-Likelihood Estimation for the AR(1) Process 179
3 Multiclass AUC Computation (In-House Developed) 180
4 Results 180
5 Conclusion 181
References 182
Analytics in Finance 183
Loan Loss Provisioning Practices in Indian Banks 184
1 Introduction 185
2 Literature 189
3 Research Methodology 190
3.1 Sample 190
3.2 Statistical Tool for Analysis of Data 191
3.3 Descriptive Study of the Sample 191
4 Analysis 192
4.1 OLS Method 192
4.2 Dynamic GMM (Generalized Method of Moments) 193
4.3 Impact of GDP and Earnings on LLP 194
5 Conclusion 194
References 195
Modeling Commodity Market Returns: The Challenge of Leptokurtic Distributions 197
1 Introduction 197
1.1 Generalized Secant Hyperbolic Distribution 198
1.2 Mixture of Normal Distributions 199
1.3 Sampling Importance Resampling (SIR) Algorithm 200
1.4 The Variance Gamma Model 200
2 Modeling Daily Gold Returns 201
2.1 Mixture of Normal Model 202
2.2 Variance Gamma Distribution Model 204
2.3 Generalized Secant Hyperbolic Distribution Model 205
3 Modeling Daily Silver Returns 206
3.1 Mixture of Normal Distributions Model 207
3.2 Variance Gamma Distribution Model 209
3.3 Generalized Secant Hyperbolic Distribution Model 210
4 Modeling Daily Crude Oil Returns 211
4.1 Mixture of Normal Model 211
4.2 Variance Gamma Distribution Model 215
4.3 Generalized Secant Hyperbolic Distribution Model 216
5 Summary and Conclusions 216
References 217
Methodology 219
OLS: Is That So Useless for Regression with Categorical Data? 220
1 Introduction 221
2 Modeling Categorical Data 223
2.1 Logistic Regression 223
2.2 Proposed Ordinary Least Square (OLS) Based Methodology 225
3 Simulations 227
4 Relative Entropy Based Assessment 232
5 Concluding Remarks 234
References 235
Estimation of Parameters of Misclassified Size Biased Borel Tanner Distribution 236
1 Introduction 236
2 Size Biased Borel–Tanner Distribution (SBBTD) 238
3 Misclassified Size Biased Borel–Tanner Distribution (MSBBTD) 240
4 Methods of Estimation of the Parameters of MSBBTD 242
5 Simulation Study 250
References 252
A Stochastic Feedback Queuing Model with Encouraged Arrivals and Retention of Impatient Customers 254
1 Introduction 255
2 Formulation of Stochastic Model 256
3 Steady-State Equations 256
4 Steady-State Solution 257
5 Measures of Performance 257
5.1 Expected System Size (Ls) 257
5.2 Expected Queue Length (Lq) 258
5.3 Average Rate of Reneging Is Given by (Rr) 258
5.4 Average Rate of Retention Is Given by (RR) 258
6 Numerical Illustration 258
7 Economic Analysis of the System 261
8 Conclusion and Future Scope 264
References 265
Econometric Applications 266
Banking Competition and Banking Stability in SEM Countries: The Causal Nexus 267
1 Introduction 267
2 Review of Literature and Rationale of Analysis 269
2.1 Hypotheses Tested 271
3 Data, Variables, and Econometric Model 271
4 Empirical Results 275
4.1 Short-Run Causality Results Between Banking Competition and Banking Stability 275
4.2 Long-Run Causality Results Between the Variables 282
4.3 Results from Innovation Accounting 282
5 Concluding Comments and Policy Implications 283
Appendix 1: Description of Variables 284
Appendix 2: Devising of Composite Index of Financial Stability by Using PCA 285
References 286

Produkt-Details

• Erscheint lt. Verlag: 7.9.2018
• Reihe/Serie: Springer Proceedings in Business and Economics
• Reihe/Serie: Springer Proceedings in Business and Economics
• Zusatzinfo: XI, 297 p. 64 illus.
• Verlagsort: Singapore
• Sprache: englisch
• Themenwelt: Mathematik / Informatik ► Informatik ► Datenbanken
• Themenwelt: Mathematik / Informatik ► Mathematik ► Finanz- / Wirtschaftsmathematik
• Themenwelt: Recht / Steuern ► Wirtschaftsrecht
• Themenwelt: Wirtschaft ► Betriebswirtschaft / Management ► Finanzierung
• Themenwelt: Wirtschaft ► Betriebswirtschaft / Management ► Unternehmensführung / Management
• Schlagworte: Big Data • Business Analytics • Business Intelligence • machine learning • Stream Data
• ISBN-10: 981-13-1208-7 / 9811312087
• ISBN-13: 978-981-13-1208-3 / 9789811312083