Index – Artificial Intelligence for Risk Management

AI. See Artificial intelligence

Artificial intelligence (AI)

capabilities, 9–10

characteristics of, 8

data to, 20–21

definition of, 8–9

distraction of, 17–20

horizontal, 11

mind map of, 3–5

purpose of, 11

strong, 9

superintelligence, 9

target audience, 1–2

types of, 9

vertical, 11

weak, 9

Artificial intelligence (AI) adoption

mass unemployment, 18

total human control, impossibility of, 18–20

Artificial intelligence (AI) projects

big data ecosystem, 14–15

business value vs. time, 13–14

definition of, 12

negative risk of, 15–16

positive risk of, 16–17

value proposition, 12–13

Artificial intelligence (AI) solutions

background, 68

data collection, 80–83

project plan, 68–73

purpose, 68

risk process life cycle, 73–80

Big data ecosystem, 14–15

Business risk, 48

Business value, 13–14

Classification, definition of, 34–35

Classification modeling, 35

Closed risks, 81

CNN. See Convolution neural network

Contract risk, 48

Convolution neural network (CNN), 26–27

Copeland, Jake, 8

Corporate risks, 48–49

Country risk, 48

Country-specific risks, 46

Data

to artificial intelligence, 20–21

definition of, 21

Data collection

risk matrix and risk trend chart, 81–82

risk register, 81

samples, 82–83

Data science, definition of, 21

Decision tree algorithm, 35

Deep learning, 25–27

Deep Q network, 33–34

Departmental risks, 47

Eager learners, 35

Fully connected neural network layers, 26

Functional area risks, 47

General risks, 57–58

Horizontal artificial
intelligence, 11

Human senses, 10–11

Industry-specific risks, 46–47

Information technology
risk, 48

Instance-based learning, 23

Intelligence, 7. See also Artificial intelligence (AI)

Knowledge acquisition, 65

Knowledge identification, 65

Knowledge sharing, 65–66

Lazy learners, 35

Liquidity risk, 48

Machine learning (ML)

categories of, 22–23

definition of, 22

reinforcement, 28

semisupervised, 25

supervised, 23–24

teaching machines, 22

transfer, 28

types of, 22–23

unsupervised, 24–25

Markov decision process (MDP), 28–30

McCarthy, John, 8

MDP. See Markov decision process

ML. See Machine learning

Model-based learning, 23

Monte Carlo (MC) algorithm

crude method, 31–32

definition of, 30

simulation, 30–31

uses of, 32–33

Natural risks, 46

Open risks, 81

Project plan

background, 68

budget, 69

constraints, assumptions, risks, and dependencies, 69–70

current state of risk process, 70

goals and scope, 69

key stakeholders, 69

milestones, 69

proposed to goal state, 70–71

risk processes, 71–73

Project risk, 48

Q-learning, 33

Q-value value network, 34

Reinforced learning, 28

Risk(s)

categorization, 58–61

concerns of, 38–39

definition of, 38

general, 57–58

identification of similarities, 60–61

response, 58

root causes of project, 59–60

types of, 46–50

Risk business case studies

alternatives and analysis, 44–45

Blockbuster and Netflix, 39–40

Enron case, 42–43

goals/scope, 44

high-level business impact, 44

risks/issues, 44

Taxi and Uber, 41–42

Risk categorization/classification model

accuracy vs. epoch graph, 103–104

conclusion, 107

confusion matrix, 102

data preparation, 100–101

defining goal, 97

design algorithm, 99

evaluating business process, 98

evaluating given dataset, 98

evaluating model, 102–107

evaluating project-related documents, 98

evaluation measure, 97

evaluation steps, 97–98

model conclusion, 107

multiclass classification model, 99–100

precision–recall curves, 103

publishing/production of model, 107

risk category dataset, 99

testing model, 102

training model, 101

tuning model, 101

weighted majority rule ensemble classifier, 104–107

Risk identification/Threat model

analytical step of evaluation, 87

binary classification models, 90

conclusion, 95

data preparation, 91

design algorithm, 89

evaluating business process, 86–87

evaluating given dataset, 87

evaluating model, 93–94

evaluating project-related documents, 87

evaluation measure, 86

model conclusion, 94

POS transaction dataset, 89

publishing/production of model, 94–95

receiver operating characteristic curve, 93

risk data collection, 88

testing model, 92

training model, 91

tuning model, 91–92

Risk impact score model

conclusion, 122

data collection, 111

data preparation, 113–116

deep neural network architecture, 116–117

defining goal, 109

design algorithm, 112

evaluating business process, 109–110

evaluating given dataset, 110

evaluating model, 119–122

evaluating project-related documents, 110–111

evaluation measure, 109

evaluation steps, 109–111

model conclusion, 122

Pearson correlation scatter plot, 114

publishing/production of model, 122

regression models, 112–113

testing model, 119

training model, 116–118

Risk knowledge base objectives, 64–65

Risk knowledge development, 64

Risk management process

Blockbuster and Netflix-case study, 54–55

description of, 50–52

development of, 52–53

icons of, 53

input and output process, 54

negative risk (threats) response strategies, 56–57

positive risk (opportunities) response strategies, 55–56

project diagram, 53

Risk management standard, 61–62

Risk matrix, 81–82

Risk mitigation strategy, 81

Risk priority model

conclusion, 130

data collection, 129

defining goal, 129

design algorithm, 129

evaluating model, 130

publishing/production of model, 130

testing model, 130

training model, 129

Risk probability occurrence model

conclusion, 128

data preparation, 126–127

data preprocessing, 127

defining goal, 123

design algorithm, 125

evaluating business process, 123–124

evaluating given dataset, 124

evaluating model, 127–128

evaluating project-related documents, 124

evaluation measure, 123

evaluation steps, 123–124

identifying features, 125–126

model conclusion, 128

publishing/production of model, 128

regression models, 126

testing model, 127

training model, 127

Risk process life cycle

identifying, 74–75

implementing, 78–79

monitoring and control, 79–80

planning, 73–74

qualifying, 75–76

quantifying, 76

responding, 77–78

Risk register, 81

Risk trend chart, 81–82

Samuel, Arthur, 22

Security risks, 49–50

Semisupervised machine learning, 25

Start-up risks, 49

Strong artificial intelligence, 9

Subject matter risks, 47–48

Superintelligence artificial intelligence, 9

Supervised machine learning

definition of, 23

linear vs. nonlinear, 23–24

Talent risk, 49

Technology risk, 48

Transfer learning, 28

Unsupervised machine learning, 24–25

Vertical artificial intelligence, 11

Weak artificial intelligence, 9