Applied Advanced Analytics: 6th IIMA International Conference on Advanced Data Analysis, Business Analytics and Intelligence (Springer Proceedings in Business and Economics)

Description

This book covers several new areas in the growing field of analytics with some innovative applications in different business contexts, and consists of selected presentations at the 6th IIMA International Conference on Advanced Data Analysis, Business Analytics and Intelligence. The book is conceptually divided in seven parts. The first part gives expository briefs on some topics of current academic and practitioner interests, such as data streams,  binary prediction and reliability shock models. In the second part, the contributions look at artificial intelligence applications with chapters related to explainable AI,  personalized search and recommendation, and customer retention management. The third part deals with credit risk analytics, with chapters on optimization of credit limits and mitigation of agricultural lending risks. In its fourth part, the book explores analytics and data mining in the retail context. In the fifth part, the book presents some applications of analytics to operations management. This part has chapters related to improvement of furnace operations, forecasting food indices and analytics for improving student learning outcomes. The sixth part has contributions related to adaptive designs in clinical trials, stochastic comparisons of systems with heterogeneous components and stacking of models. The seventh and final part contains chapters related to finance and economics topics, such as role of infrastructure and taxation on economic growth of countries and connectedness of markets with heterogenous agents, The different themes ensure that the book would be of great value to practitioners, post-graduate students, research scholars and faculty teaching advanced business analytics courses. Contents Editor and Contributors Machine Learning for Streaming Data: Overview, Applications and Challenges 1 Introduction to Machine Learning and Streaming Data 2 Classification and Regression in Streaming Data 2.1 Classification Algorithms 2.2 Regression Algorithms 3 Drift Detection Algorithms 3.1 Supervised Drift Detection 3.2 Unsupervised Drift Detection 4 Applications 5 Conclusion References Binary Prediction 1 Introduction 2 Logistic Regression 3 Application to Binary Prediction 4 Maximum Score Predictor 5 Examples 5.1 Example 1 5.2 Example 2 6 Conclusion References Reliability Shock Models: A Brief Excursion 1 Introduction 2 The Discrete Definitions 3 The Shock Models 3.1 Homogeneous Poisson Shock Model 3.2 Nonhomogeneous Poisson Shock Model 3.3 Pure Birth Shock Model 4 The Cumulative Damage Shock Model References Explainable Artificial Intelligence Model: Analysis of Neural Network Parameters 1 Introduction 2 Transparent Neural Network Model (TRANN) 3 TraNN Parameter Estimation 4 TRANN Model Parameter Test for Significance 5 Simulation Study 6 Conclusion References Style Scanner—Personalized Visual Search and Recommendations 1 Introduction 2 Related Work 3 Our Approach 4 Training Dataset Generation 5 Implementation Details 6 Dataset and Evaluation 7 Production Pipeline 8 Future Developments 9 Conclusion References Artificial Intelligence-Based Cost Reduction for Customer Retention Management in the Indian Life Insurance Industry 1 Introduction 1.1 Company Information 1.2 Background 1.3 Purpose of Research 1.4 Overview of Customer Retention Operations 2 Source of Data 2.1 Description of Datasets 3 Methodology 3.1 Defining Research Objective 3.2 Data Extraction 3.3 Data Exploration 3.4 Data Processing 3.5 Model Building 3.6 Deep Learning Model (Neural Networks) 4 Results 5 Conclusion 6 Implications References Optimization of Initial Credit Limit Using Comprehensive Customer Features 1 Introduction and Motivation 2 Literature Review 3 Methodology 4 Benefits to the Business References Mitigating Agricultural Lending Risk: An Advanced Analytical Approach 1 Introduction 2 Literature Review 3 Agriculture Credit in India: Trends and Current Scenario 4 Research Methodology 5 Results 6 Policy Implications 7 Limitations and Conclusion Appendix 1 Appendix 2: Key Summary Statistic Kolmogorov–Smirnov Test Gini Coefficient References Application of Association Rule Mining in a Clothing Retail Store 1 Introduction 1.1 Literature Review 1.2 Pricing Intelligence 2 Methodology 3 Data 3.1 Data Analysis, Results and Findings 4 Conclusion Appendix References Improving Blast Furnace Operations Through Advanced Analytics 1 Introduction 1.1 Literature Review 1.2 Purpose of Research 2 Data Availability 2.1 Understanding the Present Control System 3 Data Analysis and Modeling 3.1 Data Preparation 3.2 Modeling 3.3 Model Validation 4 Results and Conclusion References Food Index Forecasting 1 Business Problem 2 Data Gathering 3 Model Development 3.1 Pre-modeling 3.2 Modeling Process 3.3 Evaluation Metrics 4 Results 5 Conclusion References Implementing Learning Analytic Tools in Predicting Students’ Performance in a Business School 1 Introduction 2 Purpose of Research 3 Research Objectives 4 Methodology 4.1 Data Collection 4.2 Data Preparation 4.3 Partition the Data 4.4 Build Models 4.5 Evaluate the Models 5 Results and Discussion 5.1 Descriptive Analytics 5.2 Scatter Plot 5.3 Model to Predict the Academic Status in a Course 5.4 Model Building to Predict the Grade of a Student in the Capstone Course 6 Implications and Conclusion References An Optimal Response-Adaptive Design for Multi-treatment Clinical Trials with Circular Responses 1 Introduction 2 The Proposed Allocation Design 3 Implementation of the Allocation Design in Practice 4 Performance Evaluation 4.1 Performance Measures 4.2 Simulation Studies 5 Redesigning a Real Clinical Trial: SICS Trial 6 Concluding Remarks References Stochastic Comparisons of Systems with Heterogeneous Log-Logistic Components 1 Introduction 2 Notations, Definitions, and Preliminaries 3 Order Relations for Parallel Systems 4 Order Relations for Series System References Stacking with Dynamic Weights on Base Models 1 Introduction 2 Literature Review 3 Stacking by Conventional Way 3.1 Steps of Stacking by Conventional Way 4 Proposed Method I: Stacking Using Neighbourhood-Based Dynamic Weights 4.1 Steps of Stacking Using Neighbourhood-Based Dynamic Weights 5 Proposed Method II: Stacking Using Distance-Based Dynamic Weights 5.1 Steps of Stacking Using Distance-Based Dynamic Weights 6 Findings 6.1 Wholesale Customer Data 6.2 Pima Indians Diabetes Data 6.3 Bank Note Authentication Data 6.4 Iris Data 7 Conclusion References The Effect of Infrastructure and Taxation on Economic Growth: Insights from Middle-Income Countries 1 Introduction 2 Methods of Study and Data 3 Empirical Results 4 Conclusion and Policy Implications Appendix 1: List of Middle-Income Countries (MICs) References Response Prediction and Ranking Models for Large-Scale Ecommerce Search 1 Problem Statement 2 Literature Survey 3 Algorithm 4 Feature Selection 5 Business Insights 6 ML Architecture 7 A/B Framework 8 Future Work References Connectedness of Markets with Heterogeneous Agents and the Information Cascades 1 Introduction 2 Related Work 3 Research Objectives 4 Methodology 4.1 Data Source and Details 4.2 Estimation Framework 5 Preliminary Results 6 Discussion 7 Concluding Remarks 8 Appendix References