Machine Learning for Advanced Functional Materials

This book presents recent advancements of machine learning methods and their applications in material science and nanotechnologies. It provides an introduction to the field and for those who wish to explore machine learning in modeling as well as conduct data analyses of material characteristics. The book discusses ways to enhance the material’s electrical and mechanical properties based on available regression methods for supervised learning and optimization of material attributes. In summary, the growing interest among academics and professionals in the field of machine learning methods in functional nanomaterials such as sensors, solar cells, and photocatalysis is the driving force for behind this book. This is a comprehensive scientific reference book on machine learning for advanced functional materials and provides an in-depth examination of recent achievements in material science by focusing on topical issues using machine learning methods. Preface Contents Solar Cells and Relevant Machine Learning 1 Introduction 1.1 Generations of Solar Cells 1.2 Machine Learning 2 Workflow of Machine Learning 2.1 Data Collection and Preparation 2.2 Model Building and Evaluation 3 Machine Learning for Solar Cells 3.1 Naïve Bayes (NB) 3.2 Artificial Neural Network (ANN) 3.3 Decision Trees (DT) 3.4 Other Machine Learning Techniques 4 Typical Applications of ML Tools for Solar Cells 4.1 Effect of Material Properties on PCE of Solar Cells 4.2 Prediction of Optimal Device Structure 5 Conclusion and Future Recommendations References Machine Learning-Driven Gas Identification in Gas Sensors 1 Introduction 2 Gas Sensor and Electronic Nose 2.1 Gas Sensors Classification 2.2 Characteristics of Chemiresistive Type Gas Sensors 2.3 Gas Sensor with Identification Capability: Electronic Nose 3 Gas Sensing Response Features 3.1 Steady-State Features 3.2 Transient-State Features 4 Gas Sensing Signal Modulation Methods 5 Machine Learning-Enabled Smart Gas Sensor for Industrial Gas Identification 6 Summary and Outlook References A Machine Learning Approach in Wearable Technologies 1 Introduction 2 Machine Learning Algorithms Commonly Used in Wearable Technologies 2.1 Supervised Machine Learning 2.2 Non-supervised Machine Learning 2.3 Deep Learning 2.4 Evaluation Metrics 3 Application of Machine Learning in Wearable Technologies 3.1 Healthcare Applications 3.2 Sports Analytics 3.3 Smart Farming and Precision Agriculture 4 Conclusion and Outlooks References Potential of Machine Learning Algorithms in Material Science: Predictions in Design, Properties, and Applications of Novel Functional Materials 1 Introduction 2 Fundamentals of Machine Learning Algorithms: In Context of Material Science 3 Adoption of Machine Learning in Material Science 3.1 Principle 3.2 Automatic Information Acquisition 3.3 Physical Insights from Materials Learning 4 Model Generalizability and Performance in the Real World 4.1 Case Study: Prediction of TATB Peak Stress 4.2 Model Generalizability Takeaways 5 Conclusions References The Application of Novel Functional Materials to Machine Learning 1 Introduction 2 Design of Experiments and Parameter Space Optimization 2.1 Device Fabrication 2.2 Synthesis of Materials 3 Identifying Next-Generation Materials 3.1 Plan for Achieving Carbon Neutrality 3.2 Technological Advancements 4 Algorithms for Machine Learning 5 Machine Learning Applications 5.1 Batteries 5.2 Photovoltaics and Light-Emitting Materials 6 Future Perspective 6.1 Materials for CO2 Capture 6.2 Materials for Catalysis 6.3 Model Visualization and an Automated Closed-Loop Optimization Roadmap 6.4 Machine-Learned Interatomic Potentials 6.5 Data Production and Accessibility 7 Conclusion References Recent Advances in Machine Learning for Electrochemical, Optical, and Gas Sensors 1 Machine Learning 1.1 Induction of Hypotheses 1.2 Types and Tasks 2 Machine Learning in Electrochemistry 3 Machine Learning in Colorimetric Biosensors 4 Colorimetric Biosensors 5 AI Feature Extraction 6 ML Algorithms in Colorimetric Biosensors 7 Machine Learning Selection of Color Spaces 8 Machine Learning-Assisted Colorimetric Testing 9 Deep Learning Colorimetric Detection 10 Gas Sensors with Machine Learning References Perovskite-Based Materials for Photovoltaic Applications: A Machine Learning Approach 1 Introduction 2 Implementing Machine Learning in Pb-Free Perovskites for Photovoltaic Applications 2.1 Targeted Properties 2.2 Constructing Datasets 2.3 Selecting Descriptors 2.4 Feature Engineering 2.5 Machine Learning Models 3 Review of Machine Learning in Lead-Free Stable Perovskites for Photovoltaic Application 4 Conclusion and Outlook References A Review of the High-Performance Gas Sensors Using Machine Learning 1 Introduction 2 Common Process to Collect Gas Sensing Features and Conduct Machine-Learning Algorithms 3 Enhanced Gas Sensing Behavior Using Machine-Learning Techniques 3.1 Enhancement in Gas Selectivity and Concentration Prediction 3.2 Enhanced Long-Term Drift Compensation 3.3 Accurate Classification of Food 3.4 Monitoring of the Freshness of Meat 3.5 Assisted Early Cancer Diagnosis 3.6 Low-Cost Air Quality Monitoring 4 Conclusions and Possible Challenges/Prospects References Machine Learning for Next‐Generation Functional Materials 1 Introduction 1.1 Need for Functional Materials 2 Advanced Functional Materials 2.1 Materials for Combating Corona Virus Pandemic 3 Machine Learning Platform for Biological Materials 3.1 Functional Polymeric Materials 3.2 Combinatorial and Automated Polymer Chemistry 3.3 The Evolution of Molecular Modeling for Polymers 3.4 Machine Learning for Polymer Modeling and Evaluation 3.5 Computing of Electronic Polymer Materials 4 Energy Material Used for Machine Learning Platform 4.1 Machine Learning in Renewable Energy Material 5 Composite Materials for Machine Learning Applications 6 Machine Learning for Biomedical Applications 7 Conclusions and Future Directions References Contemplation of Photocatalysis Through Machine Learning 1 Introduction 2 Machine Learning in Photocatalyst 2.1 Photocatalyst Domain Knowledge Acquirement 3 Photocatalyst ML Framework 4 Integration of Domain Knowledge with Machine Learning 5 Conclusion References Discovery of Novel Photocatalysts Using Machine Learning Approach 1 Literature Survey 1.1 Photocatalyst 1.2 Machine Learning 1.3 Machine Learning as a Major Technique to Discover Novel Photocatalysts 1.4 Scope of This Proposed Work 2 Materials Used for Photocatalytic Applications, ML Descriptors, and Key Parameters 2.1 Descriptors for Photocatalysis 2.2 Matminer 2.3 Python Materials Genomics (Pymatgen) 3 Tools and Approaches Used to Predict Novel Photocatalysis Using Machine Learning 4 ML Algorithms for Predicting Novel Photocatalysts 4.1 Random Forest 4.2 Gradient Boosting Regression 4.3 Artificial Neural Networks (ANN) 4.4 Case Study 5 Conclusion and Future Perspective 6 Future Perspective References Machine Learning in Impedance-Based Sensors 1 Introduction 2 Techniques for Chemical and Biomolecules Detection 2.1 Optical-Based Methods 2.2 Field Effect Transistors (FET) 3 Electrochemical Impedance-Based Sensors 3.1 Nyquist and Bode Plots 3.2 Applications of EIS Based Sensors 4 Machine Learning in Impedance Analysis 5 Conclusions References Machine Learning in Wearable Healthcare Devices 1 Introduction 2 Wearable Devices 3 Wearable Healthcare Devices 3.1 Prevention of Diseases and Maintenance of Health 3.2 Mental Status Monitoring 3.3 Patient Management 3.4 Disease Management 4 Key Challenges with the Wearable Healthcare Devices 5 Machine Learning 6 Applications of Machine Learning in the Wearable Healthcare Devices 7 Conclusion References