Download Knowledge Representation

Survey
yes no Was this document useful for you?
   Thank you for your participation!

* Your assessment is very important for improving the workof artificial intelligence, which forms the content of this project

Document related concepts

Biology and consumer behaviour wikipedia , lookup

Neurophilosophy wikipedia , lookup

Brain Rules wikipedia , lookup

Memory consolidation wikipedia , lookup

Source amnesia wikipedia , lookup

Emotion and memory wikipedia , lookup

Holonomic brain theory wikipedia , lookup

Childhood memory wikipedia , lookup

Prenatal memory wikipedia , lookup

Eyewitness memory (child testimony) wikipedia , lookup

Atkinson–Shiffrin memory model wikipedia , lookup

Learning theory (education) wikipedia , lookup

Procedural memory wikipedia , lookup

Memory and aging wikipedia , lookup

Metamemory wikipedia , lookup

State-dependent memory wikipedia , lookup

Music-related memory wikipedia , lookup

Embodied cognitive science wikipedia , lookup

Transcript
Knowledge Representation
•
•
•
•
•
•
•
why bother ?
what is it ?
what do we represent ?
how is it represented ?
Kn Repn strategies
inferencing
example tasks
why bother ?
intelligence & knowledge
•
•
•
•
•
•
reacting to sensory infm
using tools
communication
learning
human knowledge organisation
the messed up species
psychological clues
• interlinking
• voice – person – mood
• sound/smell – memory – reminiscence
• vis-ob – properties – ownership
• cognitive limitations
•
•
•
•
levels of infm & retrieval
attention
short-term memory
field dependency
physiological clues
• perceptive pre-processing
• brain areas & malfunctions
•
•
•
•
memory freezing
the elephant's toenail
unknown ownership / name
the speechless monk
(Aitchison P.39)
who cares?
• feathers & flight
• random sorts
• chaos, complexity & emergent behaviour
why bother ?
a language example
• The old man the boats.
• I saw the racing pigeons flying to Paris.
• I saw the Eiffel Tower flying to Paris.
• The boy kicked the ball under the tree.
• The boy kicked the wall under the tree.
• Put the apple in the basket on the shelf
what is it ?
• declarative forms
data
facts
• procedural forms
processing
retrieval / linking / lumping
• inference
what do we represent ?
•
•
•
•
objects
(& their relationships)
events
(& sequences)
performance (& cause & effect)
meta kn
– extent, priority, strategy, reliability, importance
n
K
•
•
•
•
•
•
n
Rep
strategies
facts & rules
logic
semantic nets
frames
scripts
conceptual dependency
a simple semantic net
animal
if has( wings )
flying
elsif has( legs )
walking
else
crawling
moves_by
ako
brown
color
bird
ako
sparrow
has
wings
ako
budgie
ako
freddie
color
yellow
general capabilities
•
•
•
•
inheritance
defaults
demons
perspectives
inferencing
•
•
•
•
rule application & deduction
generalisation
detecting similarity
measuring differences
tasks
•
•
•
•
•
•
automatic rule generation
IQ tests
learning (near miss)
planning
agent interaction (conflict & co-operation)
human dialog management