Applied Data Science Lessons Learned for the Data-Driven Business

This book has two main goals: to define data science through the work of data scientists and their results, namely data products, while simultaneously providing the reader with relevant lessons learned from applied data science projects at the intersection of academia and industry. As such, it is not a replacement for a classical textbook (i.e., it does not elaborate on fundamentals of methods and principles described elsewhere), but systematically highlights the connection between theory, on the one hand, and its application in specific use cases, on the other.  

With these goals in mind, the book is divided into three parts: Part I pays tribute to the interdisciplinary nature of data science and provides a common understanding of data science terminology for readers with different backgrounds. These six chapters are geared towards drawing a consistent picture of data science and were predominantly written by the editors themselves. Part II then broadens the spectrum by presenting views and insights from diverse authors – some from academia and some from industry, ranging from financial to health and from manufacturing to e-commerce. Each of these chapters describes a fundamental principle, method or tool in data science by analyzing specific use cases and drawing concrete conclusions from them. The case studies presented, and the methods and tools applied, represent the nuts and bolts of data science. Finally, Part III was again written from the perspective of the editors and summarizes the lessons learned that have been distilled from the case studies in Part II. The section can be viewed as a meta-study on data science across a broad range of domains, viewpoints and fields. Moreover, it provides answers to the question of what the mission-critical factors for success in different data science undertakings are.  

The book targets professionals as well as students of data science:first, practicing data scientists in industry and academia who want to broaden their scope and expand their knowledge by drawing on the authors’ combined experience. Second, decision makers in businesses who face the challenge of creating or implementing a data-driven strategy and who want to learn from success stories spanning a range of industries. Third, students of data science who want  to understand both the theoretical and practical aspects of data science, vetted by real-world case studies at the intersection of academia and industry.

 

Preface.- 1 Introduction.- 2 Data Science.- 3 Data Scientists.- 4 Data products.- 5 Legal Aspects of Applied Data Science.- 6 Risks and Side Effects of Data Science and Data Technology.- 7 Organization.- 8 What is Data Science?.- 9 On Developing Data Science.- 10 The ethics of Big Data applications in the consumer sector.- 11 Statistical Modelling.- 12 Beyond ImageNet - Deep Learning in Industrial Practice.- 13 THE BEAUTY OF SMALL DATA - AN INFORMATION RETRIEVAL PERSPECTIVE.- 14 Narrative Visualization of Open Data.- 15 Security of Data Science and Data Science for Security.- 16 Online Anomaly Detection over Big Data Streams.- 17 Unsupervised Learning and Simulation for Complexity Management in Business Operations.- 18 Data Warehousing and Exploratory Analysis for Market Monitoring.- 19 Mining Person-Centric Datasets for Insight, Prediction, and Public Health Planning.- 20 Economic Measures of Forecast Accuracy for Demand Planning - A Case-Based Discussion.- 21 Large-Scale Data-DrivenFinancial Risk Assessment.- 22 Governance and IT Architecture.- 23 Image Analysis at Scale for Finding the Links between Structure and Biology.- 24 Lessons Learned from Challenging Data Science Case Studies. 

Prof. Dr. Kurt Stockinger is a senior lecturer in computer science at ZHAW and Director of Studies in Data Science. His research focuses on data warehousing (DWH), business intelligence (BI) and Big Data. He is also on the Advisory Board of Callista Group AG. Before joining ZHAW, Kurt was a DWH and BI Architect at Credit Suisse, Zurich where he worked on designing and implementing algorithmsfor a terabyte-scale enterprise data warehouse, data security, and DWH/BI applications. Prior, Kurt worked for four years at Lawrence Berkeley National Laboratory performing research on multi-dimensional indexing and query methods for large-scale scientific data as well as high-performance visual analytics. From 2000 to 2003 Kurt was heading the Replica Optimization Team of the EU Data Grid Project at CERN. In 2001 Kurt was a visiting researcher at California Institute of Technology.