Advances in Applications of Data-Driven Computing (Advances in Intelligent Systems and Computing)

✍ Scribed by Jagdish Chand Bansal (editor), Lance C. C. Fung (editor), Milan Simic (editor), Ankush Ghosh (editor)

• Publisher: Springer
• Year: 2021
• Tongue: English
• Leaves: 187
• Edition: 1st ed. 2021
• Category: Library

✦ Synopsis

This book aims to foster machine and deep learning approaches to data-driven applications, in which data governs the behaviour of applications. Applications of Artificial intelligence (AI)-based systems play a significant role in today’s software industry. The sensors data from hardware-based systems making a mammoth database, increasing day by day. Recent advances in big data generation and management have created an avenue for decision-makers to utilize these huge volumes of data for different purposes and analyses. AI-based application developers have long utilized conventional machine learning techniques to design better user interfaces and vulnerability predictions. However, with the advancement of deep learning-based and neural-based networks and algorithms, researchers are able to explore and learn more about data and their exposed relationships or hidden features. This new trend of developing data-driven application systems seeks the adaptation of computational neural network algorithms and techniques in many application domains, including software systems, cyber security, human activity recognition, and behavioural modelling. As such, computational neural networks algorithms can be refined to address problems in data-driven applications. Original research and review works with model and build data-driven applications using computational algorithm are included as chapters in this book.

✦ Table of Contents

Preface
Contents
Editors and Contributors
Genetic Algorithm-Based Two-Tiered Load Balancing Scheme for Cloud Data Centers
1 Introduction
2 Motivation and Contributions
3 Related Work
4 Proposed Two-Tiered Load Balancing Algorithm
4.1 Architectural Prerequisites
4.2 General System Architecture
4.3 Load Balancing Scheme
5 Simulation and Experimental Analysis
5.1 Results and Discussion
6 Conclusion and Future Work
References
KNN-DK: A Modified K-NN Classifier with Dynamic k Nearest Neighbors
1 Introduction
1.1 Problem Definition
1.2 Motivation
1.3 Contributions
1.4 Paper Organization
2 Related Work
3 Proposed Modified k-nn Classifier
3.1 Computation of k for Each Class
3.2 Proximity Computation
3.3 Class Relevance Score Computation
3.4 Prediction of Class Label for a Unknown Object
3.5 Proposed Framework
3.6 The Proposed KNN-DK Method
3.7 Complexity Analysis
4 Experimental Results
4.1 Datasets Used
4.2 Result Analysis
4.3 Discussion
5 Conclusion and Future Work
References
Identification of Emotions from Sentences Using Natural Language Processing for Small Dataset
1 Introduction
2 Background
2.1 Previous Related Works
2.2 Gaps Identified in Previous Studies
3 Proposed Methods
3.1 Traditional Machine Learning Models
3.2 Deep Learning Models
4 Methodology
4.1 Workflow
4.2 Description
5 Experiments
5.1 Naive Bayes Classifier
5.2 Support Vector Classifier
5.3 Decision Tree Classifier
5.4 Random Forest Classifier
5.5 Convolutional Neural Network (Normal and Hybrid)
5.6 Long Short-Term Memory
5.7 Gated Recurrent Unit
6 Results
7 Conclusion
8 Limitations and Future Scope
References
Comparison and Analysis of RNN-LSTMs and CNNs for Social Reviews Classification
1 Introduction
2 Concepts
2.1 Artificial Neural Networks
2.2 Recurrent Neural Networks
2.3 Long Short-Term Memory Networks (LSTMs)
2.4 Convolutional Neural Networks (CNNs)
2.5 Word2Vec
3 Model Implementation
3.1 Methodology Using RNN with LSTM
3.2 Methodology Using One-Dimensional Convolutional Neural Networks
4 Results and Discussions
4.1 Comparison Between the Performance of CNN and RNN-LTSM
5 Conclusion
References
Blockchain-Based Model for Expanding IoT Device Data Security
1 Introduction
2 Blockchain Basics
2.1 Working of Blockchain
2.2 Define the Blockchain Type
2.3 Requirement of the Blockchain
3 Related Work
4 Proposed Model For Medical Blockchain
5 Conclusions and Future Work
References
Linear Dynamical Model as Market Indicator of the National Stock Exchange of India
1 Introduction
2 Definitions and Results
3 Linear Dynamical Model
4 Application of LDM to Indian Stock Data
5 Conclusion
References
E-Focused Crawler and Hierarchical Agglomerative Clustering Approach for Automated Categorization of Feature-Level Healthcare Sentiments on Social Media
1 Introduction
2 Motivation
3 Related Work
4 Proposed Methodology
4.1 Contraceptive Assesment Gathering
4.2 Novel Focused Crawler
4.3 Preprocessing
4.4 Identification Feature (Attribute) and Word Opinion Analysis
4.5 Identification of Frequent Feature
4.6 Frequent Identification Feature (Attribute)
4.7 Sentence Level Assessment of Opinion Polarity
4.8 Clustering
5 Evaluation and Analysis
6 Conclusion
References
Error Detection Algorithm for Cloud Outsourced Big Data
1 Introduction
1.1 Motivation
2 Literature Review
3 Research Gap
4 Proposed System
4.1 Matrix Computation
4.2 Algorithm Selection for Encryption
4.3 Data Downloading, Decryption, Attaching Files Together, and Error Detection
4.4 Algorithm
5 Results and Discussion
6 Conclusion and Future Scope
7 Glossary
References
Framing Fire Detection System of Higher Efficacy Using Supervised Machine Learning Techniques
1 Introduction
2 Related Work
3 Proposed approach
3.1 Integrated Control System
3.2 Sensor Dataset
3.3 Experimental Tools
4 Results and Discussion
4.1 Description of SVM Classifier
4.2 Naive Bayes Classifier
4.3 Graph Representation
4.4 Quantitative Measure Using SVM and Naïve Bayes
5 Conclusion
References
Twitter Data Sentiment Analysis Using Naive Bayes Classifier and Generation of Heat Map for Analyzing Intensity Geographically
1 Introduction
2 Literature Survey
3 System Model
3.1 Database Creation Using Twitter Data
3.2 Sentiments
3.3 Summarized Report on the Information
3.4 Pie Chart
3.5 Heat Map
4 System Architecture
4.1 Algorithms for Analyzing Sentiments
4.2 Naive Bayes Classifier
4.3 Google Maps Heat Maps Layer
5 Case Study
6 Naive Bayes Accuracy Test Using Weka
6.1 Classifier Output
6.2 Results
7 Conclusion
8 Future Scope
References
Computing Mortality for ICU Patients Using Cloud Based Data
1 Introduction
1.1 Motivation and Goal
1.2 ICU Data Sources
1.3 Variables and Attributes
1.4 Telemedicine
1.5 Machine Learning
1.6 Cloud Computing
1.7 Domain Hosting
2 Literature Survey
2.1 ICU with Cloud
3 Computing ICU Mortality
3.1 Model Development
3.2 Chapter Perspective and Scope
3.3 Product Design and Implementation
3.4 Data Flow Diagram (DFD)
3.5 Flowchart
4 Requirement of Data Driven Computing
4.1 Non-Fundamental Requirements
4.2 Hardware and Software Design
5 Results and Discussion
5.1 Result Analysis
5.2 System Design and Implications
5.3 Implementation on Cloud
6 Conclusion
References
Early Detection of Poisonous Gas Leakage in Pipelines in an Industrial Environment Using Gas Sensor, Automated with IoT
1 Introduction
1.1 Proposed Methodology
1.2 Material Properties and Design Specifications
1.3 Work Process
2 Creation of Database and Web site
2.1 Hosting PHP Application and Creation of MySQL Database
2.2 Creation of Application Programming Interfaces (API) Key
3 Mode of Communication
3.1 Output and Readings
4 Limitation
5 Future-Scope
6 Conclusion
References