Contents
Editors and Contributors
Part I Introduction
1 Data Science: From Concept to Application
Abstract
1.1 What is Data Science?
1.2 What is and What Does a Data Scientist?
1.3 Introduction to Data Science
1.4 Systems, Tools and Methods
1.5 Applications
References
Part II Introduction to Data Science
2 Visualization and Deep Learning in Data Science
Abstract
2.1 Introduction
2.2 Methods for the Visual Preparation of Data
2.2.1 Representing Simple Data and Text
2.2.2 Simplifying and Representing Complex Data
2.2.2.1 Matrixplot
2.2.2.2 Principal Component Analysis and Multidimensional Scaling
2.2.2.3 t-SNE
2.3 Extract Image Information
2.3.1 Recognizing Visual Structures with Deep Learning
2.3.2 Architectures for Practice
2.4 Bringing Together Image and Data
2.4.1 Generation of Enriching Detail Information
2.4.2 Transformation of Visual Representations
2.4.3 Applications
2.5 Summary
References
3 Digital Ethics in Data-Driven Organizations and AI Ethics as Application Example
Abstract
3.1 Introduction
3.2 Data-driven Organizations
3.2.1 The concept of Data-driven Organization
3.2.2 Technology Use of Data-driven Organizations
3.2.3 Data-driven Corporate Culture
3.3 Digital Ethics
3.3.1 Terminology and Moral Theories
3.3.2 Overview of Digital Ethical Principles
3.4 Digital Ethics and Data-driven Organizations
3.4.1 Digital Ethical Principles and Data Value Creation
3.4.2 Consequences for the Design of Data-driven Organizations
3.5 Case Study Deutsche Telekom AG: Operationalization of AI Ethics
3.5.1 The Companyโs Motivation for Developing Digital Ethics
3.5.2 AI Ethics at the DTAG
3.6 Summary and Outlook
References
4 Multiple Perspectives for the Implementation of Innovative Technological Solutions in the Context of Data-Driven Decision-Making
Abstract
4.1 Why the Implementation of Innovative Technologies Requires a Comprehensive Approach
4.2 Models from the Literature and their Weaknesses
4.3 The Technological and Organizational Coherence Implementation Model (TOCI Model)
4.4 Benefits and Features of the TOCI Model
4.5 Possible Useful Extensions to the TOCI Model
4.6 Outlook
References
5 Donโt Be Afraid of FailureโInsights from a Survey on the Failure of Data Science Projects
Abstract
5.1 Introduction
5.2 Characteristics of and Hypotheses about Data Science Projects
5.3 Design and Conduct of the Survey
5.4 Evaluation of the Survey
5.5 Conclusion and Outlook
References
Part III Systems, Tools and Methods
6 Recommendation Systems and the Use of Machine Learning Methods
Abstract
6.1 Introduction
6.2 Collaborative Recommendation Systems
6.2.1 Approaches
6.2.2 Methods
6.3 Content-based Recommendation Systems
6.3.1 Approach
6.3.2 Methods
6.4 Other Concepts
6.4.1 Demographic Recommendation Systems
6.4.2 Knowledge-based Recommendation Systems
6.4.3 Hybrid Recommendation Systems
6.5 Current Developments
6.6 Summary
References
7 Comparison of Machine Learning Functionalities of Business Intelligence and Analytics Tools
Abstract
7.1 Introduction
7.2 Evaluation Framework for Business Intelligence Tools
7.2.1 Selection of BI Tools
7.2.2 Personas
7.2.2.1 Persona 1: Expert
7.2.2.2 Persona 2: Layperson
7.2.3 Comparison Criteria
7.2.4 Test Datasets
7.3 Comparison of ML Methods
7.3.1 SAP Analytics Cloud
7.3.2 Tableau Online/Tableau Desktop
7.3.3 Qlik Sense Business/Qlik Sense Desktop
7.3.4 TIBCO Cloud Spotfire
7.3.5 RapidMiner
7.4 Recommendations
References
8 Using the Data Science Process Model Version 1.1 (DASC-PM v1.1) for Executing Data Science Projects: Procedures, Competencies, and Roles
Abstract
8.1 Introduction
8.2 The Project Flow when Using DASC-PM v1.1
8.2.1 DASC-PM v1.1 at a Glance
8.2.2 Project Order
8.2.3 Data Provision
8.2.4 Analysis
8.2.5 Deployment
8.2.6 Application
8.3 Overarching Key Areas
8.4 Competence-driven Team Management Using Roles
8.5 Concluding Remarks
References
Part IV Applications
9 Integration of Renewable EnergiesโAI-Based Prediction Methods for Electricity Generation from Photovoltaic Systems
Abstract
9.1 Introduction and Motivation: Integration of Renewable Energies
9.2 Data Preparation
9.2.1 Data Collection
9.2.2 Data Exploration
9.2.3 Data Cleansing
9.2.4 Data Transformation
9.3 AI-based Prediction Methods
9.3.1 Approaches Based on Artificial Neural Networks
9.3.2 Approaches Based on Ensemble Machine Learning
9.4 Fusion of Results
9.5 Application Examples and Outlook
References
10 Machine Learning for Energy Management Optimization
Abstract
10.1 Digital Twin for an Air Conditioning System with Passive and Active Heat Recovery
10.2 Conception and Architecture
10.3 Analysis and Evaluation of the Data Processing Steps
10.3.1 Step 1: Data Collection
10.3.2 Step 2: Data Cleansing
10.3.3 Step 3: Classify Data
10.3.4 Step 4: Filter Data
10.3.5 Step 5: Calculate Prediction
10.4 Proof-of-Concept
10.4.1 Methods and Technologies Stack
10.4.2 Visualization of results
10.5 Conclusion
10.6 Outlook
10.6.1 Further Analysis Approaches
10.6.2 Applications
References
11 Text Mining in Scientific Literature Evaluation: Extraction of Keywords for Describing Content
Abstract
11.1 Introduction
11.2 Explainable Artificial Intelligence
11.3 Extraction of Keywords
11.4 Extraction of keywords for a literature review on โExplainable AIโ
11.5 Conclusion
References
12 Identification of Relevant Relationships in Data Using Machine Learning
Abstract
12.1 Introduction
12.2 Expertise Problem
12.3 Approaches to Reducing the Number of Rules
12.3.1 Association Rule Discovery
12.3.2 Subgroup Discovery
12.4 Determining the Quality of Reduced Rule Sets
12.5 Combination Schema
12.6 Results
12.7 Summary
References
13 Framework for the Management and Analysis of Vehicle Data for Model-Based Driver Assistance System Development in Teaching and Research
Abstract
13.1 Motivation
13.2 Wildauer Maschinen Werke at TH Wildau
13.3 Presentation of the Vehicle Fleet
13.3.1 Trikes
13.3.2 Trucks
13.4 Introduction to Infrastructure
13.4.1 ROS
13.4.2 Node-RED
13.4.3 MQTT-Bridge
13.4.4 ROS Car2X
13.4.5 Traffic Light Systems
13.4.6 VDI
13.5 Development Framework
13.5.1 Implementation of Vehicle Communication
13.5.2 Model-based Development and Code Generation for Vehicles
13.5.3 Agile Project Management, Knowledge Management and Source Code Management
13.6 Scenario-based Teaching and Research
13.6.1 ROS Car2X as Data Aggregation and Function Behavior Across Vehicles
13.6.2 NodeRED for Data Analysis
13.6.3 Interdisciplinary Scenario using the Example of Material Management
13.7 Summary and Outlook
References
Index