Recent Advances in Hybrid Metaheuristics for Data Clustering

An authoritative guide to an in-depth analysis of various state-of-the-art data clustering approaches using a range of computational intelligence techniques
 
Recent Advances in Hybrid Metaheuristics for Data Clustering offers a guide to the fundamentals of various metaheuristics and their application to data clustering. Metaheuristics are designed to tackle complex clustering problems where classical clustering algorithms have failed to be either effective or efficient. The authors-noted experts on the topic-provide a text that can aid in the design and development of hybrid metaheuristics to be applied to data clustering.
 
The book includes performance analysis of the hybrid metaheuristics in relationship to their conventional counterparts. In addition to providing a review of data clustering, the authors include in-depth analysis of different optimization algorithms. The text offers a step-by-step guide in the build-up of hybrid metaheuristics and to enhance comprehension. In addition, the book contains a range of real-life case studies and their applications. This important text:
* Includes performance analysis of the hybrid metaheuristics as related to their conventional counterparts
* Offers an in-depth analysis of a range of optimization algorithms
* Highlights a review of data clustering
* Contains a detailed overview of different standard metaheuristics in current use
* Presents a step-by-step guide to the build-up of hybrid metaheuristics
* Offers real-life case studies and applications
 
Written for researchers, students and academics in computer science, mathematics, and engineering, Recent Advances in Hybrid Metaheuristics for Data Clustering provides a text that explores the current data clustering approaches using a range of computational intelligence techniques.

List of contents

List of Contributors xiii
 
Series Preface xv
 
Preface xvii
 
1 Metaheuristic Algorithms in Fuzzy Clustering 1
Sourav De, Sandip Dey, and Siddhartha Bhattacharyya
 
1.1 Introduction 1
 
1.2 Fuzzy Clustering 1
 
1.2.1 Fuzzy c-means (FCM) clustering 2
 
1.3 Algorithm 2
 
1.3.1 Selection of Cluster Centers 3
 
1.4 Genetic Algorithm 3
 
1.5 Particle Swarm Optimization 5
 
1.6 Ant Colony Optimization 6
 
1.7 Artificial Bee Colony Algorithm 7
 
1.8 Local Search-Based Metaheuristic Clustering Algorithms 7
 
1.9 Population-Based Metaheuristic Clustering Algorithms 8
 
1.9.1 GA-Based Fuzzy Clustering 8
 
1.9.2 PSO-Based Fuzzy Clustering 9
 
1.9.3 Ant Colony Optimization-Based Fuzzy Clustering 10
 
1.9.4 Artificial Bee Colony Optimization-Based Fuzzy Clustering 10
 
1.9.5 Differential Evolution-Based Fuzzy Clustering 11
 
1.9.6 Firefly Algorithm-Based Fuzzy Clustering 12
 
1.10 Conclusion 13
 
References 13
 
2 Hybrid Harmony Search Algorithm to Solve the Feature Selection for Data Mining Applications 19
Laith Mohammad Abualigah, Mofleh Al-diabat, Mohammad Al Shinwan, Khaldoon Dhou, Bisan Alsalibi, Essam Said Hanandeh, and Mohammad Shehab
 
2.1 Introduction 19
 
2.2 Research Framework 21
 
2.3 Text Preprocessing 22
 
2.3.1 Tokenization 22
 
2.3.2 StopWords Removal 22
 
2.3.3 Stemming 23
 
2.3.4 Text Document Representation 23
 
2.3.5 TermWeight (TF-IDF) 23
 
2.4 Text Feature Selection 24
 
2.4.1 Mathematical Model of the Feature Selection Problem 24
 
2.4.2 Solution Representation 24
 
2.4.3 Fitness Function 24
 
2.5 Harmony Search Algorithm 25
 
2.5.1 Parameters Initialization 25
 
2.5.2 Harmony Memory Initialization 26
 
2.5.3 Generating a New Solution 26
 
2.5.4 Update Harmony Memory 27
 
2.5.5 Check the Stopping Criterion 27
 
2.6 Text Clustering 27
 
2.6.1 Mathematical Model of the Text Clustering 27
 
2.6.2 Find Clusters Centroid 27
 
2.6.3 Similarity Measure 28
 
2.7 k-means text clustering algorithm 28
 
2.8 Experimental Results 29
 
2.8.1 Evaluation Measures 29
 
2.8.1.1 F-measure Based on Clustering Evaluation 30
 
2.8.1.2 Accuracy Based on Clustering Evaluation 31
 
2.8.2 Results and Discussions 31
 
2.9 Conclusion 34
 
References 34
 
3 Adaptive Position-Based Crossover in the Genetic Algorithm for Data Clustering 39
Arnab Gain and Prasenjit Dey
 
3.1 Introduction 39
 
3.2 Preliminaries 40
 
3.2.1 Clustering 40
 
3.2.1.1 k-means Clustering 40
 
3.2.2 Genetic Algorithm 41
 
3.3 RelatedWorks 42
 
3.3.1 GA-Based Data Clustering by Binary Encoding 42
 
3.3.2 GA-Based Data Clustering by Real Encoding 43
 
3.3.3 GA-Based Data Clustering for Imbalanced Datasets 44
 
3.4 Proposed Model 44
 
3.5 Experimentation 46
 
3.5.1 Experimental Settings 46
 
3.5.2 DB Index 47
 
3.5.3 Experimental Results 49
 
3.6 Conclusion 51
 
References 57
 
4 Application of Machine Learning in the Social Network 61
Belfin R. V., E. Grace Mary Kanaga, and Suman Kundu
 
4.1 Introduction 61
 
4.1.1 Social Media 61
 
4.1.2 Big Data 62
 
4.1.3 Machine Learning 62
 
4.1.4 Natural Language Processing (NLP) 63
 
4.1.5 Social Network Analysis 64
 
4.2 Application of Classification Models in Social Networks 64

About the author

Sourav De, PhD, is an Associate Professor of Computer Science and Engineering at Cooch Behar Government Engineering College, West Bengal, India. Sandip Dey, PhD, is an Assistant Professor of Computer Science at Sukanta Mahavidyalaya, Dhupguri, Jalpaiguri, India. Siddhartha Bhattacharyya, PhD, is a Professor of Computer Science and Engineering at CHRIST (Deemed to be University), Bangalore, India.

Summary

An authoritative guide to an in-depth analysis of various state-of-the-art data clustering approaches using a range of computational intelligence techniques

Recent Advances in Hybrid Metaheuristics for Data Clustering offers a guide to the fundamentals of various metaheuristics and their application to data clustering. Metaheuristics are designed to tackle complex clustering problems where classical clustering algorithms have failed to be either effective or efficient. The authors-noted experts on the topic-provide a text that can aid in the design and development of hybrid metaheuristics to be applied to data clustering.

The book includes performance analysis of the hybrid metaheuristics in relationship to their conventional counterparts. In addition to providing a review of data clustering, the authors include in-depth analysis of different optimization algorithms. The text offers a step-by-step guide in the build-up of hybrid metaheuristics and to enhance comprehension. In addition, the book contains a range of real-life case studies and their applications. This important text:
* Includes performance analysis of the hybrid metaheuristics as related to their conventional counterparts
* Offers an in-depth analysis of a range of optimization algorithms
* Highlights a review of data clustering
* Contains a detailed overview of different standard metaheuristics in current use
* Presents a step-by-step guide to the build-up of hybrid metaheuristics
* Offers real-life case studies and applications

Written for researchers, students and academics in computer science, mathematics, and engineering, Recent Advances in Hybrid Metaheuristics for Data Clustering provides a text that explores the current data clustering approaches using a range of computational intelligence techniques.

Product details

• Authors: Siddhartha Bhattacharyya, S De, Sourav De, Sourav Dey De, Sandip Dey
• Assisted by: Siddhartha Bhattacharyya (Editor), Bhattacharyya Siddhartha (Editor), Sourav De (Editor), De Sourav (Editor), Sandi Dey (Editor), Sandip Dey (Editor)
• Publisher: Wiley, John and Sons Ltd
• Languages: English
• Product format: Hardback
• Released: 31.07.2020
• EAN: 9781119551591
• ISBN: 978-1-119-55159-1
• No. of pages: 200
• Series: The Wiley Intelligent Signal a; The Wiley Series in Intelligent Signal and Data Processing
• Subjects: Natural sciences, medicine, IT, technology > IT, data processing > IT; Statistik, Datenverarbeitung, Informatik, Datenanalyse, Statistics, computer science, data analysis, Data Processing, Algorithms & data structures, Bauingenieur- u. Bauwesen, Civil Engineering & Construction, Datenverarbeitung im Bauwesen, Algorithmen u. Datenstrukturen