Fuzzy Computing in Data Science
Description
This book comprehensively explains how to use various fuzzy-based models to solve real-time industrial challenges.
The book provides information about fundamental aspects of the field and explores the myriad applications of fuzzy logic techniques and methods. It presents basic conceptual considerations and case studies of applications of fuzzy computation. It covers the fundamental concepts and techniques for system modeling, information processing, intelligent system design, decision analysis, statistical analysis, pattern recognition, automated learning, system control, and identification. The book also discusses the combination of fuzzy computation techniques with other computational intelligence approaches such as neural and evolutionary computation.
Audience
Researchers and students in computer science, artificial intelligence, machine learning, big data analytics, and information and communication technology.
Sachi Nandan Mohanty, PhD, received his doctorate from IIT Kharagpur in 2015 and his PostDoc from IIT Kanpur in 2019. He has recently joined as an associate professor at VIT-AP University, Andhra Pradesh. He has edited 24 books and published more than 100 research papers in international journals and has been elected as Fellow of the Institute of Engineers and Senior member of IEEE Computer Society Hyderabad chapter. His research areas include data mining, big data analysis, cognitive science, fuzzy decision-making, brain-computer interface, and computational intelligence.
Prasenjit Chatterjee, PhD, is an associate professor in the Mechanical Engineering Department at MCKV Institute of Engineering, India. He has more than 80 research papers in various international SCI journals. Dr. Chatterjee is one of the developers of a new multiple-criteria decision-making method called Measurement of Alternatives and Ranking according to Compromise Solution (MARCOS).
Bui Thanh Hung, PhD, is the Director of Artificial Intelligence Laboratory, Faculty of Information Technology, Ton Duc Thang University, Vietnam, and received his doctorate from Japan Advanced Institute of Science and Technology (JAIST) in 2013. He has published numerous research articles in international journals and conferences as well as 14 book chapters. His main research interests are natural language processing, machine learning, machine translation, text processing, data analytics, computer vision, and artificial intelligence.
More details
Other editions
Additional editions
Persons
Sachi Nandan Mohanty, PhD, received his doctorate from IIT Kharagpur in 2015 and his PostDoc from IIT Kanpur in 2019. He has recently joined as an associate professor at VIT-AP University, Andhra Pradesh. He has edited 24 books and published more than 100 research papers in international journals and has been elected as Fellow of the Institute of Engineers and Senior member of IEEE Computer Society Hyderabad chapter. His research areas include data mining, big data analysis, cognitive science, fuzzy decision-making, brain-computer interface, and computational intelligence.
Prasenjit Chatterjee, PhD, is an associate professor in the Mechanical Engineering Department at MCKV Institute of Engineering, India. He has more than 80 research papers in various international SCI journals. Dr. Chatterjee is one of the developers of a new multiple-criteria decision-making method called Measurement of Alternatives and Ranking according to Compromise Solution (MARCOS).
Bui Thanh Hung, PhD, is the Director of Artificial Intelligence Laboratory, Faculty of Information Technology, Ton Duc Thang University, Vietnam, and received his doctorate from Japan Advanced Institute of Science and Technology (JAIST) in 2013. He has published numerous research articles in international journals and conferences as well as 14 book chapters. His main research interests are natural language processing, machine learning, machine translation, text processing, data analytics, computer vision, and artificial intelligence.
Content
Preface xvii
Acknowledgement xxi
1 Band Reduction of HSI Segmentation Using FCM 1
V. Saravana Kumar, S. Anantha Sivaprakasam, E.R. Naganathan, Sunil Bhutada and M. Kavitha
1.1 Introduction 2
1.2 Existing Method 3
1.2.1 K-Means Clustering Method 3
1.2.2 Fuzzy C-Means 3
1.2.3 Davies Bouldin Index 4
1.2.4 Data Set Description of HSI 4
1.3 Proposed Method 5
1.3.1 Hyperspectral Image Segmentation Using Enhanced Estimation of Centroid 5
1.3.2 Band Reduction Using K-Means Algorithm 6
1.3.3 Band Reduction Using Fuzzy C-Means 7
1.4 Experimental Results 8
1.4.1 DB Index Graph 8
1.4.2 K-Means-Based PSC (EEOC) 9
1.4.3 Fuzzy C-Means-Based PSC (EEOC) 10
1.5 Analysis of Results 12
1.6 Conclusions 16
References 17
2 A Fuzzy Approach to Face Mask Detection 21
Vatsal Mishra, Tavish Awasthi, Subham Kashyap, Minerva Brahma, Monideepa Roy and Sujoy Datta
2.1 Introduction 22
2.2 Existing Work 23
2.3 The Proposed Framework 26
2.4 Set-Up and Libraries Used 26
2.5 Implementation 27
2.6 Results and Analysis 29
2.7 Conclusion and Future Work 33
References 34
3 Application of Fuzzy Logic to the Healthcare Industry 37
Biswajeet Sahu, Lokanath Sarangi, Abhinadita Ghosh and Hemanta Kumar Palo
3.1 Introduction 38
3.2 Background 41
3.3 Fuzzy Logic 42
3.4 Fuzzy Logic in Healthcare 45
3.5 Conclusions 49
References 50
4 A Bibliometric Approach and Systematic Exploration of Global Research Activity on Fuzzy Logic in Scopus Database 55
Sugyanta Priyadarshini and Nisrutha Dulla
4.1 Introduction 56
4.2 Data Extraction and Interpretation 58
4.3 Results and Discussion 59
4.3.1 Per Year Publication and Citation Count 59
4.3.2 Prominent Affiliations Contributing Toward Fuzzy Logic 60
4.3.3 Top Journals Emerging in Fuzzy Logic in Major Subject Areas 61
4.3.4 Major Contributing Countries Toward Fuzzy Research Articles 63
4.3.5 Prominent Authors Contribution Toward the Fuzzy Logic Analysis 66
4.3.6 Coauthorship of Authors 67
4.3.7 Cocitation Analysis of Cited Authors 68
4.3.8 Cooccurrence of Author Keywords 68
4.4 Bibliographic Coupling of Documents, Sources, Authors, and Countries 70
4.4.1 Bibliographic Coupling of Documents 70
4.4.2 Bibliographic Coupling of Sources 71
4.4.3 Bibliographic Coupling of Authors 72
4.4.4 Bibliographic Coupling of Countries 73
4.5 Conclusion 74
References 76
5 Fuzzy Decision Making in Predictive Analytics and Resource Scheduling 79
Rekha A. Kulkarni, Suhas H. Patil and Bithika Bishesh
5.1 Introduction 80
5.2 History of Fuzzy Logic and Its Applications 81
5.3 Approximate Reasoning 82
5.4 Fuzzy Sets vs Classical Sets 83
5.5 Fuzzy Inference System 84
5.5.1 Characteristics of FIS 85
5.5.2 Working of FIS 85
5.5.3 Methods of FIS 86
5.6 Fuzzy Decision Trees 86
5.6.1 Characteristics of Decision Trees 87
5.6.2 Construction of Fuzzy Decision Trees 87
5.7 Fuzzy Logic as Applied to Resource Scheduling in a Cloud Environment 88
5.8 Conclusion 90
References 91
6 Application of Fuzzy Logic and Machine Learning Concept in Sales Data Forecasting Decision Analytics Using ARIMA Model 93
S. Mala and V. Umadevi
6.1 Introduction 94
6.1.1 Aim and Scope 94
6.1.2 R-Tool 94
6.1.3 Application of Fuzzy Logic 94
6.1.4 Dataset 95
6.2 Model Study 96
6.2.1 Introduction to Machine Learning Method 96
6.2.2 Time Series Analysis 96
6.2.3 Components of a Time Series 97
6.2.4 Concepts of Stationary 99
6.2.5 Model Parsimony 100
6.3 Methodology 100
6.3.1 Exploratory Data Analysis 100
6.3.1.1 Seed Types-Analysis 101
6.3.1.2 Comparison of Location and Seeds 101
6.3.1.3 Comparison of Season (Month) and Seeds 103
6.3.2 Forecasting 103
6.3.2.1 Auto Regressive Integrated Moving Average (ARIMA) 103
6.3.2.2 Data Visualization 106
6.3.2.3 Implementation Model 108
6.4 Result Analysis 108
6.5 Conclusion 110
References 110
7 Modified m-Polar Fuzzy Set ELECTRE-I Approach 113
Madan Jagtap, Prasad Karande and Pravin Patil
7.1 Introduction 114
7.1.1 Objectives 114
7.2 Implementation of m-Polar Fuzzy ELECTRE-I Integrated Shannon's Entropy Weight Calculations 115
7.2.1 The m-Polar Fuzzy ELECTRE-I Integrated Shannon's Entropy Weight Calculation Method 115
7.3 Application to Industrial Problems 118
7.3.1 Cutting Fluid Selection Problem 118
7.3.2 Results Obtained From m-Polar Fuzzy ELECTRE-I for Cutting Fluid Selection Problem 122
7.3.3 FMS Selection Problem 125
7.3.4 Results Obtained From m-Polar Fuzzy ELECTRE-I for FMS Selection 130
7.4 Conclusions 143
References 143
8 Fuzzy Decision Making: Concept and Models 147
Bithika Bishesh
8.1 Introduction 148
8.2 Classical Set 149
8.3 Fuzzy Set 150
8.4 Properties of Fuzzy Set 151
8.5 Types of Decision Making 153
8.5.1 Individual Decision Making 153
8.5.2 Multiperson Decision Making 157
8.5.3 Multistage Decision Making 158
8.5.4 Multicriteria Decision Making 160
8.6 Methods of Multiattribute Decision Making (MADM) 162
8.6.1 Weighted Sum Method (WSM) 162
8.6.2 Weighted Product Method (WPM) 162
8.6.3 Weighted Aggregates Sum Product Assessment (WASPAS) 163
8.6.4 Technique for Order Preference by Similarity to Ideal Solutions (TOPSIS) 166
8.7 Applications of Fuzzy Logic 167
8.8 Conclusion 169
References 169
9 Use of Fuzzy Logic for Psychological Support to Migrant Workers of Southern Odisha (India) 173
Sanjaya Kumar Sahoo and Sukanta Chandra Swain
9.1 Introduction 174
9.2 Objectives and Methodology 175
9.2.1 Objectives 175
9.2.2 Methodology 176
9.3 Effect of COVID-19 on the Psychology and Emotion of Repatriated Migrants 176
9.3.1 Psychological Variables Identified 176
9.3.2 Fuzzy Logic for Solace to Migrants 176
9.4 Findings 178
9.5 Way Out for Strengthening the Psychological Strength of the Migrant Workers through Technological Aid 178
9.6 Conclusion 179
References 180
10 Fuzzy-Based Edge AI Approach: Smart Transformation of Healthcare for a Better Tomorrow 181
B. RaviKrishna, Sirisha Potluri, J. Rethna Virgil Jeny, Guna Sekhar Sajja and Katta Subba Rao
10.1 Significance of Machine Learning in Healthcare 182
10.2 Cloud-Based Artificial Intelligent Secure Models 183
10.3 Applications and Usage of Machine Learning in Healthcare 183
10.3.1 Detecting Diseases and Diagnosis 183
10.3.2 Drug Detection and Manufacturing 183
10.3.3 Medical Imaging Analysis and Diagnosis 184
10.3.4 Personalized/Adapted Medicine 185
10.3.5 Behavioral Modification 185
10.3.6 Maintenance of Smart Health Data 185
10.3.7 Clinical Trial and Study 185
10.3.8 Crowdsourced Information Discovery 185
10.3.9 Enhanced Radiotherapy 186
10.3.10 Outbreak/Epidemic Prediction 186
10.4 Edge AI: For Smart Transformation of Healthcare 186
10.4.1 Role of Edge in Reshaping Healthcare 186
10.4.2 How AI Powers the Edge 187
10.5 Edge AI-Modernizing Human Machine Interface 188
10.5.1 Rural Medicine 188
10.5.2 Autonomous Monitoring of Hospital Rooms-A Case Study 188
10.6 Significance of Fuzzy in Healthcare 189
10.6.1 Fuzzy Logic-Outline 189
10.6.2 Fuzzy Logic-Based Smart Healthcare 190
10.6.3 Medical Diagnosis Using Fuzzy Logic for Decision Support Systems 191
10.6.4 Applications of Fuzzy Logic in Healthcare 193
10.7 Conclusion and Discussions 193
References 194
11 Video Conferencing (VC) Software Selection Using Fuzzy TOPSIS 197
Rekha Gupta
11.1 Introduction 197
11.2 Video Conferencing Software and Its Major Features 199
11.2.1 Video Conferencing/Meeting Software (VC/MS) for Higher Education Institutes 199
11.3 Fuzzy TOPSIS 203
11.3.1 Extension of TOPSIS Algorithm: Fuzzy TOPSIS 203
11.4 Sample Numerical Illustration 207
11.5 Conclusions 213
References 213
12 Estimation of Nonperforming Assets of Indian Commercial Banks Using Fuzzy AHP and Goal Programming 215
Kandarp Vidyasagar and Rajiv Kr. Dwivedi
12.1 Introduction 216
12.1.1 Basic Concepts of Fuzzy AHP and Goal Programming 217
12.2 Research Model 221
12.2.1 Average Growth Rate Calculation 227
12.3 Result and Discussion 233
12.4 Conclusion 234
References 234
13 Evaluation of Ergonomic Design for the Visual Display Terminal Operator at Static Work Under FMCDM Environment 237
Bipradas Bairagi
13.1 Introduction 238
13.2 Proposed Algorithm 240
13.3 An Illustrative Example on Ergonomic Design Evaluation 245
13.4 Conclusions 249
References 249
14 Optimization of Energy Generated from Ocean Wave Energy Using Fuzzy Logic 253
S. B. Goyal, Pradeep Bedi, Jugnesh Kumar and Prasenjit Chatterjee
14.1 Introduction 254
14.2 Control Approach in Wave Energy Systems 255
14.3 Related Work 257
14.4 Mathematical Modeling for Energy Conversion from Ocean Waves 259
14.5 Proposed Methodology 260
14.5.1 Wave Parameters 261
14.5.2 Fuzzy-Optimizer 262
14.6 Conclusion 264
References 264
15 The m-Polar Fuzzy TOPSIS Method for NTM Selection 267
Madan Jagtap and Prasad Karande
15.1 Introduction 268
15.2 Literature Review 268
15.3 Methodology 270
15.3.1 Steps of the mFS TOPSIS 270
15.4 Case Study 272
15.4.1 Effect of Analytical Hierarchy Process (AHP) Weight Calculation on the mFS TOPSIS Method 273
15.4.2 Effect of Shannon's Entropy Weight Calculation on the m-Polar Fuzzy Set TOPSIS Method 277
15.5 Results and Discussions 281
15.5.1 Result Validation 281
15.6 Conclusions and Future Scope 283
References 284
16 Comparative Analysis on Material Handling Device Selection Using Hybrid FMCDM Methodology 287
Bipradas Bairagi
16.1 Introduction 288
16.2 MCDM Techniques 289
16.2.1 Fahp 289
16.2.2 Entropy Method as Weights (Influence) Evaluation Technique 290
16.3 The Proposed Hybrid and Super Hybrid FMCDM Approaches 291
16.3.1 Topsis 291
16.3.2 FMOORA Method 292
16.3.3 FVIKOR 292
16.3.4 Fuzzy Grey Theory (FGT) 293
16.3.5 COPRAS -G 293
16.3.6 Super Hybrid Algorithm 294
16.4 Illustrative Example 295
16.5 Results and Discussions 298
16.5.1 FTOPSIS 298
16.5.2 FMOORA 298
16.5.3 FVIKRA 298
16.5.4 Fuzzy Grey Theory (FGT) 299
16.5.5 COPRAS-G 299
16.5.6 Super Hybrid Approach (SHA) 299
16.6 Conclusions 302
References 302
17 Fuzzy MCDM on CCPM for Decision Making: A Case Study 305
Bimal K. Jena, Biswajit Das, Amarendra Baral and Sushanta Tripathy
17.1 Introduction 306
17.2 Literature Review 307
17.3 Objective of Research 308
17.4 Cluster Analysis 308
17.4.1 Hierarchical Clustering 309
17.4.2 Partitional Clustering 309
17.5 Clustering 310
17.6 Methodology 314
17.7 TOPSIS Method 316
17.8 Fuzzy TOPSIS Method 318
17.9 Conclusion 325
17.10 Scope of Future Study 326
References 326
Index 329
1
Band Reduction of HSI Segmentation Using FCM
V. Saravana Kumar1*, S. Anantha Sivaprakasam2, E.R. Naganathan3, Sunil Bhutada1 and M. Kavitha4
1 Department of IT, SreeNidhi Institute of Science and Technology, Hyderabad, India
2 Department of CSE, Rajalakshmi Engineering College, Chennai, India
3 Department of CSE, Koneru Lakshmaiah Education Foundation, Vaddeswaram, AP, India
4 Research Scholar, Manonmaniam Sundaranar University, Tirunelveli, India
Abstract
Hyperspectral has carried hundreds of nonoverlapping spectral channels of a specified scene, clustering is one of the approaches for diminishing the size of these large data sets. Segmentation is intricate for the raw data; however, it is likely for the reduced band of HSI. To lessen the band size, the classical clustering methods for example K-means, Fuzzy C-means are accomplished. An integrated image segmentation procedure built on interband clustering and intraband clustering is proposed. The interband clustering is performed by K-means clustering and Fuzzy C-means clustering algorithms, despite the fact the intraband clustering is executed using particle swarm segmentation (PSO) clustering algorithm. The performance of the K-means algorithm is subject to initial cluster centers. Besides, the final partition should be contingent on the initial configuration. The clustering consequences have profoundly been subject to the number of clusters stated. It is essential to provide refined direction for defining the number of clusters with the purpose of attaining appropriate clustering consequences. Davies Bouldin (DB) index is one of the reliable methods to outline the number of clusters for these clustering methods. The hyperspectral bands are clustered, and a band which has extreme variance from each cluster is preferred. This tactic forms the diminished set of bands. PSC (EEOC) accomplishes the segmentation process on the reduced bands. In conclusion, there is a comparison of the result produced for K-means worked with EEOC and FCM worked with EEOC in various HSI scenes.
Keywords: K-means, Fuzzy C-means, band reduction, PSO, cluster, centroid
1.1 Introduction
Hyperspectral [14] has carried hundreds of nonoverlapping spectral channels [11] of a given scene, clustering [13, 21] is one of the methods for reducing the size of these large data sets. Despite displaying the size of hyperspectral scene [17], the device does not support to display the scene directly. Segmentation is complicated for the raw data, whereas it is possible for the reduced band scene. Even though hyperspectral data [3] can provide finely resolved details about the spectral properties [4] to be identified, it also has some limitation. When dealing with such high-dimensional data [24, 25], one is faced with the "curse of dimensionality" problem One popular way to tackle the curse of dimensionality [7] is to employ a feature extraction technique [22]. To diminish the band size, the classical clustering methods [30], such as K-means, Fuzzy C-means [27] are handled. The former can be sensitive to the initial centers, while the results from the latter depend on the initial weights. Here, an integrated image segmentation [2] process based on interband clustering and intraband clustering is proposed. The interband clustering is performed by K-means clustering or Fuzzy C-means clustering algorithms, whereas the intraband clustering is executed using particle swarm segmentation (PSO) clustering algorithm. The performance of the K-means algorithm depends on initial cluster centers. Besides, the final partition depends on the initial configuration. The clustering results have heavily depended on the number of clusters specified. It is necessary to provide educated guidance for determining the number of clusters in order to achieve appropriate clustering results. Davies Bouldin (DB) index is one of the reliable methods to determine the number of clusters for these clustering methods. The hyperspectral bands [18] are clustered and a band [34], which has maximum variance, from each cluster is chosen. This forms the reduced set of bands. PSC (EEOC) [29] performs the segmentation process on the reduced bands. Finally, the result produced from K-means worked with EEOC and FCM worked with EEOC in various HSI [19] scenes was compared.
1.2 Existing Method
1.2.1 K-Means Clustering Method
Theoretically, K-means [1] is a typical algorithm. Now that it is elementary and expeditious, it is attractive in practice. To begin with, it segregates the input dataset into K-lusters. Each cluster is described by an adaptively changing centroid, starting from some initial values named seed points. K-means enumerates the squared distances between the inputs and centroids, and assigns inputs to the nearest centroid. The procedure continues until there is no significant change in the location of class mean vectors between successive iteration of the algorithms. Apparently, the performance of the K-means algorithm depends on initial cluster centers, whereas the final partition depends on the initial configuration.
1.2.2 Fuzzy C-Means
In Fuzzy C-means clustering [8], data elements can belong to more than one cluster and associated with each element is a set of membership levels. FCM clustering [9] is a process of assigning these membership levels, and then using them to assign data elements to one or more clusters. The aim of FCM [10] is to minimize an objective function. Contrary to traditional clustering analysis methods, which distribute each object to a unique group, fuzzy clustering algorithms gain membership values between 0 and 1 that indicate the degree of membership for each object to each group. Obviously, the sum of the membership values for each object to all the groups is definitely equal to 1. Different membership values show the probability of each object to different groups.
The main limitation of the FCM algorithm [16] is its sensitivity to noise. The FCM algorithm implements the clustering task for a data set by minimizing an objective-function subject to the probabilistic constraint that the summation of all the membership degrees of every data point to all clusters must be one. This constraint results in the problem of this membership assignment, that noise is treated the same as points close to the cluster centers. However, in reality, these points should be assigned very low or even zero membership in either cluster.
Like K-means, the clustering results have heavily depended on the number specified. It is also necessary to provide an educated guidance for determining the number cluster in order to achieve appropriate clustering results. Davies Bouldin (DB) index is one of the reliable methods to determine the number of clusters for these clustering methods.
1.2.3 Davies Bouldin Index
Davies Bouldin index was introduced in 1979 by David L. Davies and Donald W. Bouldin. It is one of the methods for evaluating clustering algorithms. This is an internal evaluation scheme, where the validation of how well the clustering has done using quantities and features inherent to the dataset.
Many other distance metrics can be used, in the case of manifolds and higher dimensional data, where the Euclidean distance may not be the best measure for determining the clusters. It is important to note that this distance metric has to match with the metric used in the clustering scheme itself for meaningful results.
Davies-Bouldin (DB) index is dependent both on the data, as well as the algorithm. The Davies-Bouldin index measures the average of similarity between each cluster and its most similar one. As the clusters have to be compact and separated the lower Davies-Bouldin index produces a better cluster configuration.
In this work, the interband clustering and intraband clustering approach has proposed. In interband clustering part; existing method such as K-means and Fuzzy C-means method are applied to reduce the band size. In intraband clustering part; a lightweight algorithm, namely Enhanced Estimation of Centroid (EEOC) [33] is proposed. This method is examined with the abovementioned clustering method by applying in various hyperspectral scenes.
1.2.4 Data Set Description of HSI
The dataset contains a variety of hyperspectral remote sensing [15], which are acquired from airborne and satellite. In this work, certain data, such as Salinas A, Salinas Valley, Indian Pines, Pavia University, and Pavia Centre, are handled.
The original scene and its corresponding ground truth image are downloaded from the link. http://www.ehu.eus/ccwintco/index.p...Hyperspectral_Remote_Sensing_Scenes
These scenes are a widely used benchmark for testing the accuracy of hyperspectral data [20] classification [23, 26] and segmentation [31].
1.3 Proposed Method
1.3.1 Hyperspectral Image Segmentation Using Enhanced Estimation of Centroid
These topics explain the integration of intraband and interband cluster approach for segmentation of hyperspectral image in a synergistic fashion. It demonstrates the advantage of the advanced properties of both analysis techniques in combined fashion of clustering method. The ultimate goal is to improve the analysis and interpretation of hyperspectral image. Owing that, classifications [5, 12], as well as segmentation [6], face problems related to the...
