M1 Flashcards
Intelligence is defined usign these characteristics (3)
Reasoning
Learning
Adaptivity
___ adapts itself to deal with changes in problems (automatic learning)
intelligent system
___ follows problem solving processes similar to humans
Machine intelligence
___ display machine intelligence, not necessarily self-adapting
Intelligent systems
Intelligent systems in ___ utilize one or more intelligence tools to aid decision making
business
Customer behaviour patterns, Market trend, Efficiency bottlenecks are examples of ___
business intelligence – information
Customer service, Scheduling, Data Mining, Financial market prediction, Quality control are examples of __ in business
successful intelligent systems applications
What intelligent systems is used in the example?
HNC software’s credit card fraud detector – 30-70% improvement
ANN
What intelligent systems is used in the example?
US Occupational Safety and Health Administration (OSHA uses “expert advisors” to help identify fire and other safety hazards at work sites
Expert Systems
What intelligent systems is used in the example?
MetLife insurance uses automated extraction of information from applications
Language Technology
What intelligent systems is used in the example?
Personalized, Internet-based TV listings
Intelligent Agent
What intelligent systems is used in the example?
Hyundai’s development apartment construction plans
CBR
Intelligent Systems should posses these characteristics (CHLDF)
Capability to extract and store knowledge
Human like reasoning process
Learning from experience (or training)
Dealing with imprecise expressions of facts
Finding solutions through processes similar to natural evolution
Among the recent trend in interaction with user usign intelligent systems is through ? (NSI)
natural language understanding
speech recognition and synthesis
image analysis
Most current intelligent systems based on? (RO)
rule based expert systems
one or more of the methodologies belonging to soft computing
The primary goal of Artificial Intelligence is develop software aimed at enabling machines to solve problems through ___ reasoning
human-like
AI encompasses study of the ___ to understand its structure and functions
brain
AI is in existence as a discipline since the ___
1960s
AI failed to live up to initial expectations due to inadequate understanding of ___ function and complexity of ___ to be solved
brain, problem
___ is an AI success story
Expert Systems
Soft Computing is also known as ___
Computational Intelligence
Unlike conventional computing, SC techniques can be tolerant of imprecise, incomplete or corrupt ___
input data
Unlike conventional computing, SC techniques solve problems without ___ solution steps
explicit
Unlike conventional computing, SC techniques learn the solution through ___ observation and adaptation
repeated
Unlike conventional computing, SC techniques can handle information expressed in ___
vague linguistic terms
Unlike conventional computing, SC techniques arrive at an acceptable solution through ___
evolution
The first four characteristics
- can be tolerant of imprecise, incomplete or corrupt input data
- solve problems without explicit solution steps
- learn the solution through repeated observation and adaptation
- can handle information expressed in vague linguistic terms
are common in problem solving by ___
humans
The fifth characteristic
- arrive at an acceptable solution through evolution
is common in ___
nature
The predominant SC methodologies found in current intelligent systems are (AFG)
Artificial Neural Networks (ANN)
Fuzzy Systems
Genetic Algorithms (GA)
___ is designed to solve problems in a specific domain
Expert Systems
Expert Systems is built by interrogating ___ and storing acquired knowledge in a form suitable for problem solving problems using ___
domain experts, reasoning
Expert Systems is usually used in ___ user for problem specific information and using the information to draw ___ from the knowledge base
querying, inferences
Human brain consists of billions of highly interconnected simple processing elements known as ___
neurons
___ are based on a simplified model of the neurons and their operation
Artificial Neural Networks (ANN)
ANNs usually learn from experience – repeated presentation of example problems with corresponding ___
solutions
The ____ phase of neural networks may or may not involve human intervention
learning
In ANN, the problem solving strategy developed remains ___ and ___ to the user
implicit, unknown
ANN is particularly suitable for problems not prone to ___, eg, pattern recognition, decision support
algorithmic solutions
Different models of ANNs depending on ___, learning method and other operational characteristics
architecture
ANN is good at ___ and ___ problems
pattern recognition, classification
ANN’s major strength is the ability to handle previously ___, ___ or ___ data
unseen, incomplete or corrupted
Explosive detection at airports, character and signature recognition, financial risk assessment, optimization and scheduling. These are some examples of ?
ANN
Genetic Algorithm belongs to a broader field known as ___
evolutionary computation
Solution in genetic algorithm are obtained by evolving solutions through a process consisting of? (3)
survival of the fittest, crossbreeding, and mutation
In genetic Algorithm, a population of ___ are initialized (the chromosomes)
candidate solutions
New generation of solutions produced from the current population using splicing and joining two chromosomes is called?
crossover
Fitness of newly evolved solution evaluated using a ___
fitness function
The steps of solution generation and evaluation continue until an ___ solution is found
acceptable
___ logic is two-valued – any proposition is either true or false
Traditional
Fuzzy systems handle imprecise information by assigning ___
degrees of truth
FS allow us to express knowledge in ___ linguistic terms
vague
___ systems solve problems by making use of knowledge about similar problems encountered in the past
CBR
In CBR, the knowledge used in the past is built up as a ___
case-base
___ is the process of exploring and analysing data for discovering new and useful information
Data Mining
Techniques such as ___ and ___ have made it possible to perform data mining involving large volumes of data.
artificial neural networks and decision trees
___ are computer programs that provide active assistance to information system users
Intelligent Software Agents
Intelligent Software Agents help users cope with information ___
overload
Intelligent Software Agents act in many ways like a personal assistant to the user by attempting to adapt to the ___ needs of the user
specific
___ between people and computers is an important aspect of any intelligent information system
Communication
Natural Language Processing (NLP)
Speech recognition
Optical character recognition (OCR)
Handwriting recognition
Machine translation
Text summarisation
Speech synthesis
are applications of ___
Language Technology
A Language Technology-based system can be the ____ of information systems themselves based on other intelligence tools
front-end
