Deck 13: Advanced Intelligent Systems

Question 1

A personal daemon must be able to make decisions on its own and alter its course of behavior when it meets an obstacle and find ways around the impediment.

Answer 1

False

Question 2

Inductive learning allows the computer to use historical cases to solve new problems.

Answer 2

False

Question 3

Development of a fuzzy inference system is a four-step process: fuzzification,inferencing,summation,and defuzzification.

Answer 3

False

Question 4

Machine-learning techniques enable computers to acquire knowledge (i.e.,learn)from data that reflects the historical happenings.

Question 5

Case-based reasoning (CBR)is analogical reasoning that adapts solutions that had been used to solve old problems to solve new problems.

Question 6

The thinking behind fuzzy logic is that decision making is not always a matter of crisp logic,e.g.,black or white or true or false.It often involves tones of gray and varying degrees of truth.

Question 7

Support vector machines (SVM)have a solid mathematical foundation in statistical learning theory,but are too complex for use in real-world applications.

Question 8

Fuzzy logic systems can be used in conducting property appraisals since it is necessary to use judgment,experience,and intuition to generate estimates.

Question 9

Genetic algorithms (GA)are a part of global search techniques that are used to find or produce the best solution to a specific problem.

Question 10

An advantage of SVM models is that they tend to take less time to develop compared to other predictive models.

Question 11

The foundation of case-based reasoning (CBR)is a repository or library of cases called a case base that contains a number of previous cases for decision making.

Question 12

A hyperplane is a geometric concept commonly used to describe the separation surface between different classes of things within a multidimensional space.

Question 13

Interactivity is the most important characteristic of intelligent agents,making them capable of acting on their own or of being empowered.

Question 14

STRATASSIST is a fuzzy ES that helps small- to medium-sized firms plan strategically by asking questions in five strategic,competitive areas.

Question 15

Unsupervised learning is a process of inducing knowledge from a set of observations.

Question 16

Genetic algorithms can be used to solve business problems such as assembly-line balancing,facility layout,machine and job shop scheduling,creditworthiness determination,and aircraft design.

Question 17

Machines learning overcomes deficiencies of manual knowledge acquisition techniques by automating the learning process.

Question 18

In SVM,kernel trick allows us to construct a decision surface that is 3D in the feature space while it is inherently nonlinear in the original input space.

Question 19

The knapsack problem is a conceptually simple optimization problem that can be solved directly using analytical methods,but it is ideal for illustrating a genetic algorithm approach.

Question 20

In the SVM model development process,the main advantage of normalizing or scaling the data is to avoid attributes in greater numeric ranges dominating those in smaller numeric ranges.

Question 21

The foundation of case-based reasoning is a library of cases called ________.

A) case base
B) data base
C) legacy system
D) reasoning base

Question 22

Fuzzy logic in consumer products is sometimes called ________.

A) inferencing
B) defuzzification
C) continuous logic
D) crisp logic

Question 23

In unsupervised learning:

A) only input stimuli are fed into the network.
B) both the inputs and the desired outputs are known.
C) the desired output and the algorithms are known.
D) the network is controlled by the user.

Question 24

Which of the following are special cases that contain rich contents and special features of deep implications?

A) Descriptive cases
B) Foundation cases
C) Special cases
D) Stories cases

Question 25

Human learning is a combination of many complicated cognitive processes.Which of the following is not included in human's cognitive processes?

A) Induction
B) Simulation
C) Deduction
D) Analogy

Question 26

Machine learning is considered a(n)________ method.

A) trial-and-error
B) optimization
C) simulation
D) artificial intelligence

Question 27

Which of the following is not a method of machine learning?

A) Genetic algorithms
B) Generic algorithms
C) Neural computing
D) Case-based reasoning

Question 28

Machine learning is a family of artificial intelligence technologies that is primarily concerned with:

A) the use of use neural networks to support simple and repetitive decision making.
B) the design and development of algorithms that allow computers to learn based on historical data.
C) identifying outliers and patterns in data sets.
D) simulating creativity to generate courses of action.

Question 29

Crossover means choosing a random position in the string and exchanging the segments ________ of this point with another string.

A) to the right
B) to the left
C) to both left and right
D) to either left or right

Question 30

A chromosome in a genetic algorithm refers to ________.

A) crossover
B) summation
C) mutation
D) candidate solution

Question 31

Neural networks identify ________ in the data from which they generate recommended courses of action.

A) patterns
B) outliers
C) movements
D) associations

Question 32

Which of the following is not an advantage of fuzzy logic?

A) Greater flexibility
B) Faster calculations
C) More options
D) All of the above are advantages

Question 33

Survival of the fittest (natural selection)is used in ________.

A) genetic algorithms
B) fuzzy logic
C) case-based reasoning
D) fuzzy neural networks

Question 34

Which of the following is not a relevant observation on how learning relates to intelligent systems?

A) Automated learning systems are anchored in a formal bedrock; thus, their implications are well understood.
B) Automated learning systems demonstrate intriguing learning behaviors.
C) Although AI sometimes matches human-level learning capabilities, it is not able to learn as well as humans or in the same way that humans do.
D) A common thread running through most AI approaches to learning is the manipulation of symbols rather than numeric information.

Question 35

Autonomy implies that an intelligent agent takes initiative and exercises control over its own actions by having each of the following characteristics except:

A) An agent should accept high-level requests indicating what a human wants, and it should be responsible for deciding how and where to satisfy the requests.
B) An agent should be able to dynamically choose which actions to invoke, and in what sequence, in response to the state of its external environment.
C) The agent should obey commands and not be able to modify requests.
D) An agent should sense changes in its environment and decide when to act.

Question 36

Which of the following is a limitation of GA?

A) Difficult to determine good variables for a particular problem
B) The results of GA may vary
C) It might reach local optimum
D) All of the above

Question 37

In supervised learning:

A) only input stimuli are fed into the network.
B) both the inputs and the desired outputs are known.
C) the desired output and the algorithms are known.
D) the network is controlled by the user.

Question 38

________ allows the computer to examine historical cases and generate rules that can be used to solve new problems or that can be deployed for automating a decision support process that repeatedly deals with a specific class of problems (e.g.,evaluating loan applications).

A) Exploration learning
B) Case-based reasoning
C) Explanation-based learning
D) Inductive learning

Question 39

According to the Simplified Taxonomy of Machine Learning,all of the following are types of learning except:

A) Supervised learning
B) Reinforcement learning
C) Unreiforced learning
D) Unsupervised learning

Question 40

Which of the following is not a step in case-based reasoning process?

A) Retrieve
B) Reuse
C) Retain
D) Reserve

Question 41

________ allows computers to examine historical cases and generate rules that can be chained to solve new problems.

Question 42

________ segmentation is one of the unsupervised machine learning methods wherein data is put into several groups based on their similarity.

Question 43

________ deals with the kind of uncertainty and partial information that is inherently human in nature.

Question 44

A(n)________ is a geometric concept commonly used to describe the separation surface between different classes of things within a multidimensional space.

Question 45

________ is used to discover knowledge from a set of data whose outcomes are unknown.

Question 46

________ is one of the unintended by-products of the information age characterized by huge volumes of information that cross the desks of mangers and decision makers.

Question 47

A genetic algorithm represents its candidate solutions as strings of genes called ________ and measures their viability with a fitness function.

Question 48

________ are one of the most popular machine-learning techniques.They belong to the family of generalized linear models and are closely associated with feedforward-type artificial neural networks.

Question 49

An ________ is an autonomous computer program that observes and acts upon an environment and directs its activity toward achieving specific goals.

Question 50

Genetic algorithms are sets of computational procedures that conceptually follow steps inspired by the biological processes of ________.

Question 51

Fuzzy sets use a spectrum of possible values called ________.

Question 52

Genetic algorithms demonstrate self-organization and ________ in much the same way that the fittest biological organisms survive and reproduce.

Question 53

The process in which the input and output are processed by fuzzy logic before entering neural network in fuzzy neural networks is called ________.

Question 54

The basic principle of machine learning is that there are data that preserve previous decision experience,and these experience-based records are called ________.

Question 55

Human learning is a combination of many complicated ________ processes.

Question 56

________ is the degree of autonomy and authority vested in an agent and can be measured by the nature of the interaction between the agent and other entities in the system.

Question 57

________ is the degree to which agents themselves travel through a network.Some agents may be static,either residing on the client machine or initiated at the server.

Question 58

Genetic algorithms ________ by producing offspring that are better and better measured by a fitness-to-survive function.

Question 59

________ is a process of inducing knowledge from a set of observations whose outcomes are known.

Question 60

In machine learning,the ________ is a method for using a linear classifier algorithm to solve a nonlinear problem

Question 61

Define these three categories of cases--ossified cases,paradigmatic cases,and stories--based on their different characteristics and the different ways of handling them.

Question 62

What are the advantages of using case-based reasoning?

Question 63

Describe case-based reasoning.What type of leaning is it?

Question 64

Describe the genetic operators of Genetic Algorithm in reproduction.

Question 65

What are genetic algorithms?

Question 66

What is fuzzy logic?