Big Data Analysis Using Machine Learning for Social Scientists and Criminologists
by Juyoung Song (Author)

About the Author
Dr Juyoung Song is an Assistant Professor of Criminal Justice and Criminology at Pennsylvania State University-Schuylkill, USA. She earned her Bachelors and Masters degrees from the College of Law at Hanyang University, South Korea, and then obtained her doctorate degree in Criminal Justice from Michigan State University, USA. She has published several articles on cyberbullying, underage prostitution, and juvenile delinquency in the International Journal of Offender Therapy, Comparative Criminology, and Journal of Criminal Justice, and has published five books about big data analysis in Korean. Dr Tae Min Song is a Professor in the Department of Health Management at Sahmyook University, South Korea, and is a Visiting Research Fellow at the Big Data Research Center of the Korean Institute for Health and Social Affairs, where he worked as a data scientist and policy maker for 36 years. He holds a PhD in Computer Science, and has authored numerous journal articles in Computers in Human Behavior, Journal of Adolescence Heath, and various other journals. He has also published five books related to big data analysis in Korean.

About this Book
This book provides a detailed description of the entire study process concerning gathering and analysing big data and making observations to develop a crime-prediction model that utilises its findings. It offers an in-depth discussion of several processes, including text mining, which extracts useful information from online documents; opinion mining, which analyses the emotions contained in documents; machine learning for crime prediction; and visualization analysis. To accurately predict crimes using machine learning, it is necessary to procure high-quality training data. Machine learning combined with high-quality data can be used to develop excellent crime-prediction artificial intelligences. As such, the book will serve to be a practical guide to anyone wishing to predict rapidly-changing social phenomena and draw creative conclusions using big-data analysis.

Brief Contents
Installation and Use of R 1
    Installation of R 1
    Use of R 7
Scientific Research Design  35
    Research Concepts 36
    Variable Measurement 37
    Unit of Analysis 39
    Sampling and Hypothesis Testing  39
    Statistical Analysis 44
Overview of Machine Learning 118
    Introduction 118
    Machine Learning Training Data 122
Development of a Cyber bullying Prediction Model Based on Machine Learning 124
    Naïve Bayes Classification Model 124
    Logistic Regression Model  130
    Random Forest Model 134
    Decision Tree Model 141
    Neural Network Model  149
    Support Vector Machine Model 162
    Association Analysis  170
    Cluster Analysis and Segmentation 179
Machine Learning Model Evaluation  186
    Machine Learning Model Evaluation Using Misclassification Tables 189
    Machine Learning Model Evaluation Using ROC Curves 208
Artificial Intelligence 215
    Calculate the Effect of Input Variables on Output Variables (Prediction Probability)  215
    Using Training Data with Input Variables to Create Dependent Variables 221
    Creating Data with the Same Training-Data and Predicted-Data Classifications 225
    Evaluating Existing Training Data and High Quality Training Data 228
    Creating an Artificial Intelligence with Machine Learning 230
Visualization  236
    Visualization of Text Data 236
    Visualization of Time Series Data 239
    Visualization of Geographical Data 250
Developing Machine Learning–Based Predictive Models of Adverse Drug Responses  258
    Introduction  258
    Research Subjects and Analysis Method 263
    Result 269
    Discussion and Conclusion  302
Index  307

Pages : 311
ISBN-10 : 1527533883
ISBN-13 : 978-1527533882
Publisher : Cambridge Scholars Publishing; 1st Edition (July 1, 2019)
Language : English