Knowledge Representation and Reasoning

Part-II

Dr. Singara Singh Kasana

Associate Professor
Computer Science and Engineering Department
Thapar Institute of Engineering and Technology
Patiala
 WEAK SLOT-AND-FILLER STRUCTURES
 The structures that support property inheritance along instance and is a
links are called slot-and-filler structures.
 These structures are useful for support of other reasons besides
property inheritance:
1) It indexes assertions by the entities they describe. The binary
predicates are indexed by their first argument. So retrieving the value
for a particular attribute of an object is fast.
2) It is easy to describe properties of objects which requires some higher
order mechanisms in purely logic based system.
3) It is a form of object oriented programming. So it has all advantages
that such systems have like modularity, inheritance, ease of use, etc.
 These structures do not say much about the specific knowledge
these should contain. So they are called as “knowledge-poor” or
“weak-slot-and-filler structures”.
 These structures include semantic networks and frames.
2. Semantic Network Representation
 In Semantic networks, you can represent your knowledge in the form of
graphical networks.
 This network consists of nodes representing objects and arcs which
describe the relationship between those objects.
 Also, it categorizes the object in different forms and links those objects.
This representation consists of two types of relations:
i. IS-A relation (Inheritance): Child is a subclass of parent class
ii. Has–A relation(Composition)
Graph is intended to represent the data like
 Tomy is a cat
 Tomy has fur
 Cat likes milk
 Tomy is a mammal
Consider the following sentences:
1) Clyde and Nellie are elephants.
2) Nellie likes apples.
3) All elephants are mammals.
4) All elephants are large in size and grey colored.
5) All mammals and reptiles are animals.
6) Mammals have head.
Represent the knowledge in the form of semantic networks.
 animal

isa isa

has_part
reptile mammal head

isa

size colour
large elephant grey

instance_of instance_of

likes
Clyde Nellie apples
 SEMANTIC NETWORK EXAMPLE 2
Represent following KB as semantic networks.
 Tom is a cat.
 Tom caught a bird.
 Tom is owned by John.
 Tom is ginger in colour.
 Cats like cream.
 The cat sat on the mat.
 A cat is a mammal.
 A bird is an animal.
 All mammals are animals.
 Mammals have fur.
 REPRESENTING NON-BINARY PREDICATES
RELATIONS
 Semantic nets are natural ways of representing
binary predicates of predicate logic like:
 Team(Pee-Wee-Resse,Brooklyn Dodgers)
 Uniform_color(Pee-Wee-Resse,blue)

 Unary predicates can be converted in binary predicates

using instance and is a relations.
For example man(marcus) can be converted into semantic
network by representing it as instance(marcus,man)
 Three or more arity predicates can be converted into binary
form by creating new objects representing the entire
predicate statement and then representing relationships of
all old objects with this new object.
REPRESENTING NON-BINARY PREDICATES
RELATIONS CONTD..
 John gave Mary the AI book
Advantages:
 Semantic networks are a natural representation of knowledge.
 Also, it conveys meaning in a transparent manner.
 These networks are simple and easy to understand.
Disadvantages:
 Semantic networks take more computational time at runtime we need to
traverse the complete network tree to answer some questions..
 Also, these are inadequate as they do not have any equivalent quantifiers e.g.,
for all, for some, none etc.
 These networks are not intelligent and depend on the creator of the system.
 Semantic networks do not have any standard definition for the link names.
3. Frame Representation
 A frame is a record like structure that consists of a collection of attributes and
values to describe an entity in the world.
 These are the AI data structures that divides knowledge into substructures by
representing stereotypes situations.
 Basically, it consists of a collection of slots and slot values of any type and size.
Slots have names and values which are called facets.
 Frames are a variant of semantic networks
 All the information relevant to a concept is stored in a
single complex entity called a frame.
 Superficially, frames look like record data structures or
class.

class Book {
Person author;
String title;
int price;
}

 A single frame is rarely useful.

 We build a collection of frames called frame systems that
are connected by virtue of the fact that the value of an
attribute of one frame is another frame.
 In frame-based systems we refer to
 objects – Mammal, Elephant, and Nellie;
 slots – properties such as color and size;
 slot-values – values stored in the slots, e.g. grey and
large.

 Slots and the corresponding slot-values are inherited

through the class hierarchy
 FRAMES - EXAMPLE
 The previous example can be also represented as
frames:
Mammal:
subclass: Animal animal
has-part: head
isa
Elephant:
has_part
subclass: Mammal mammal head
colour: grey
size: large isa

Nellie: size colour

large elephant grey
instance: Elephant
likes: apples instance_of
instance_of

Clyde: Clyde Nellie likes apples

instance: Elephant
 FRAMES – EXAMPLE CONTD....
 The above frame system can also be represented as:
Animal

Mammal
subclass: Animal
has-part: head Nellie
Instance: Elephant
likes: apples
Elephant
subclass: Mammal
colour: grey Clyde
size: large Instance: Elephant
Example of a frame for a Student

Slots Filters
Roll No 101
Branch BE
Name Ram
Age 20
Address Patiala
AGPA 9.8
Advantages:
 It makes the programming easier by grouping the related data.
 Frame representation is easy to understand and visualize.
 It is very easy to add slots for new attributes and relations.
 Also, it is easy to include default data and search for missing values.

Disadvantages:
 In frame system, inference mechanism cannot be easily processed.
 It has a very generalized approach.
4. Production Rules
 In production rules, agent checks for the condition and if the condition exists then production rule fires and
corresponding action is carried out.
 The condition part of the rule determines which rule may be applied to a problem. Whereas, the action
part carries out the associated problem-solving steps. This complete process is called a recognize-act cycle.
The production rules system consists of three main parts:
i. The set of production rules
ii. Working Memory
iii. The recognize-act-cycle
Example:
IF (at station AND train arrives) THEN action (get into the train)
IF (on the train AND empty seat) THEN action (sit down).
IF (on train AND unpaid) THEN action (pay charges).
IF (train arrives at destination) THEN action (get down from the
train).
The working memory contains the description of the current state of problems-
solving and rule can write knowledge to the working memory. This knowledge
match and may fire other rules.

If there is a new situation (state) generation, then multiple production rules may
be fired together, this is called conflict set. In this situation, the agent needs to
select a rule from these sets, and it is called a conflict resolution.
Advantages of Production rule:
1. The production rules are expressed in natural language.
2. The production rules are highly modular, so we can easily remove, add or
modify an individual rule.
Disadvantages of Production rule:
1. Production rule system does not exhibit any learning capabilities, as it does not
store the result of the problem for the future uses.
2. During the execution of the program, many rules may be active hence rule-
based production systems are inefficient.