Unit 2

Knowledge Representation

Knowledge representation can be explained as the way artificial intelligence represents knowledge
not with the help of stored data in the system but with prior experiences and knowledge to act like
an intelligent human.

What knowledge needs to be represented?

Below are the different kinds of knowledge that need to be represented.

 Object

There are many objects present in the human world. All information we have, related to all the
objects, can be considered as a type of knowledge. For instance, a bus has wheels and a guitar has
strings, etc.

 Events

Our understanding of the world is based on the idea we have about the various incidents that have
occurred in our world. Thus, events refer to every action that happens in our world.

 Performance

The term is used to explain human behavior or the way they perform certain actions during
different situations.

 Meta-knowledge

Knowledge about things we are already aware of.

 Facts

The reality of the actual world and what we stand for.

Knowledge base
Knowledge Base abbreviated as KB is the most significant part element of knowledge-based agents.
It refers to a set of information about any discipline, field, etc. For example, a knowledge base on
road construction.

Types of knowledge representation

There are different types of knowledge which are categorized as follows:

 Declarative knowledge

It refers to the knowledge that lets us describe our world and it contains everything including ideas,
facts, objects, etc… and therefore deals with the description of things.

 Procedural knowledge

Also referred to as imperative knowledge. It refers to more complex concepts (how things work or
behave) than declarative knowledge. Therefore, this knowledge is used to complete any task with
specific rules, processes, and agendas and thus makes the system work efficiently. Also, this kind of
knowledge relies on the task we are trying to finish.

 Meta-knowledge

Meta-knowledge is a set of cognitive knowledge when combined. So, this is knowledge about
different types of knowledge.

 Heuristic knowledge

Heuristic knowledge is the knowledge imparted by experts of particular domains, topics, and areas,
which they have acquired after years of experience. This kind of knowledge enables you to take a
better approach to specific problems and make decisions.

 Structural knowledge

This type of knowledge serves as the basic form of knowledge for solving problems in the real
world and helps to establish a connection between ideas or objects and their description.

Properties of knowledge representation

There are certain properties or requirements for a good knowledge representation system. This
system should demonstrate some of the features that help us to evaluate the system. These
properties are listed below:

 Representational adequacy

Adequacy and the ability to make the AI system understand are the important assets of a
knowledge representation system, which means it represents all the knowledge necessary to
manage a particular field or domain.

 Inferential adequacy

It refers to the flexibility of the knowledge representation system to deal with the existing
knowledge to pave the way for new knowledge.

 Acquisitional efficiency

The ultimate property of the knowledge representation system is its ability to automatically acquire
new knowledge, enabling AI to integrate into its existing knowledge and, as a result, become more
efficient and productive.

 Inferential efficiency

With the presence of existing old knowledge, the system of representation cannot include new
knowledge but it can put in this knowledge efficiently and without hindrance.

The Connection between knowledge and intelligence

Knowledge of the actual world plays an important role in intelligence, as well as in the creation of
artificial intelligence. When it comes to expressing intelligent behavior in AI agents, knowledge
plays an essential part. An agent cannot function accurately, without enough knowledge or
experience of certain inputs. For example, if you have to interact with a person and you are unable
to understand his language, then definitely you will not be able to give a response or deliver any
action. This is the same as with the intelligent behavior of the agents. A decision-maker works by
discovering the environment and applying knowledge. However, without the knowledge
component, it cannot exhibit intellectual behavior.
Techniques of knowledge representation

There are mainly four ways of knowledge representation which are given as follows:

1. Logical Representation
2. Semantic Network Representation
3. Frame Representation
4. Production Rules

1. Logical Representation

Logical representation is a language with some concrete rules which deals with propositions and
has no ambiguity in representation. Logical representation means drawing a conclusion based on
various conditions. This representation lays down some important communication rules. It consists
of precisely defined syntax and semantics which supports the sound inference. Each sentence can
be translated into logics using syntax and semantics.

Syntax:

o Syntaxes are the rules which decide how we can construct legal sentences in the logic.
 o It determines which symbol we can use in knowledge representation.
o How to write those symbols.

Semantics:

o Semantics are the rules by which we can interpret the sentence in the logic.
o Semantic also involves assigning a meaning to each sentence.

Logical representation can be categorised into mainly two logics:

 Propositional Logics
 Predicate logics

Advantages of logical representation:

1. Logical representation enables us to do logical reasoning.

2. Logical representation is the basis for the programming languages.

Disadvantages of logical Representation:

1. Logical representations have some restrictions and are challenging to work with.
2. Logical representation technique may not be very natural, and inference may not be so
efficient.

2. Semantic Network Representation

Semantic networks are alternative of predicate logic for knowledge representation. In Semantic
networks, we can represent our knowledge in the form of graphical networks. This network
consists of nodes representing objects and arcs which describe the relationship between those
objects. Semantic networks can categorize the object in different forms and can also link those
objects. Semantic networks are easy to understand and can be easily extended.Play Video

This representation consists of mainly two types of relations:

 IS-A relation (Inheritance)

 Kind-of-relation

Example: Following are some statements which we need to represent in the form of nodes and
arcs.
Statements:

a. Jerry is a cat.
b. Jerry is a mammal
c. Jerry is owned by Priya.
d. Jerry is brown colored.
e. All Mammals are animal.

In the above diagram, we have represented the different type of knowledge in the form of nodes
and arcs. Each object is connected with another object by some relation.

Drawbacks in Semantic representation:

1. Semantic networks take more computational time at runtime as we need to traverse the
complete network tree to answer some questions. It might be possible in the worst case
scenario that after traversing the entire tree, we find that the solution does not exist in this
network.
2. Semantic networks try to model human-like memory (Which has 1015 neurons and links) to
store the information, but in practice, it is not possible to build such a vast semantic
network.
3. These types of representations are inadequate as they do not have any equivalent
quantifier, e.g., for all, for some, none, etc.
4. Semantic networks do not have any standard definition for the link names.
 5. These networks are not intelligent and depend on the creator of the system.

Advantages of Semantic network:

1. Semantic networks are a natural representation of knowledge.

2. Semantic networks convey meaning in a transparent manner.
3. These networks are simple and easily understandable.

3. Frame Representation

A frame is a record like structure which consists of a collection of attributes and its values to
describe an entity in the world. Frames are the AI data structure which divides knowledge into
substructures by representing stereotypes situations. It consists of a collection of slots and slot
values. These slots may be of any type and sizes. Slots have names and values which are called
facets.

Facets: The various aspects of a slot is known as Facets. Facets are features of frames which enable
us to put constraints on the frames. Example: IF-NEEDED facts are called when data of any
particular slot is needed. A frame may consist of any number of slots, and a slot may include any
number of facets and facets may have any number of values. A frame is also known as slot-filter
knowledge representation in artificial intelligence.

Frames are derived from semantic networks and later evolved into our modern-day classes and
objects. A single frame is not much useful. Frames system consist of a collection of frames which
are connected. In the frame, knowledge about an object or event can be stored together in the
knowledge base. The frame is a type of technology which is widely used in various applications
including Natural language processing and machine visions.

Example: 1

Let's take an example of a frame for a book

Slots Filters

Title Artificial Intelligence

Genre Computer Science

Author Peter Norvig

Edition Third Edition

Year 1996

Page 1152

Example 2:

Let's suppose we are taking an entity, Peter. Peter is an engineer as a profession, and his age is 25,
he lives in city London, and the country is England. So following is the frame representation for this:

Slots Filter

Name Peter

Profession Doctor

Age 25

Marital status Single

Weight 78

Advantages of frame representation:

1. The frame knowledge representation makes the programming easier by grouping the
related data.
2. The frame representation is comparably flexible and used by many applications in AI.
3. It is very easy to add slots for new attribute and relations.
4. It is easy to include default data and to search for missing values.
5. Frame representation is easy to understand and visualize.

Disadvantages of frame representation:

1. In frame system inference mechanism is not be easily processed.

2. Inference mechanism cannot be smoothly proceeded by frame representation.
3. Frame representation has a much generalized approach.

4. Production Rules
Production rules system consist of (condition, action) pairs which mean, "If condition then action".
It has mainly three parts:

o The set of production rules

o Working Memory
o The recognize-act-cycle

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. And the action part carries out the associated problem-solving steps.
This complete process is called a recognize-act cycle.

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) generates, then multiple production rules will 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.

Example:

o IF (at bus stop AND bus arrives) THEN action (get into the bus)
o IF (on the bus AND paid AND empty seat) THEN action (sit down).
o IF (on bus AND unpaid) THEN action (pay charges).
o IF (bus arrives at destination) THEN action (get down from the bus).

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.
Examples of NLP

 Machine Translation

Machine Translation (MT) is the task of automatically converting one natural language into
another, preserving the meaning of the input text, and producing fluent text in the output
language.

We can understand the process of machine translation with the help of the following flowchart −

Types of Machine Translation Systems

There are different types of machine translation systems.

  Bilingual MT System
Bilingual MT systems produce translations between two particular languages.
 Multilingual MT System
Multilingual MT systems produce translations between any pair of languages. They may be either
uni-directional or bi-directional in nature.

Approaches of Machine Translation

Machine Translation has been present in the industry for several decades. Beginning in the 1970s,
there were projects to achieve automatic translation. Over the years three major approaches have
emerged:

 Rule-based Machine Translation (RBMT): 1970-1990

 Statistical Machine Translation (SMT): 1990-2010
 Neural Machine Translation (NMT): 2014-Present

 Rule-based Machine Translation (RBMT): 1970-1990

RBMT systems are built on linguistic principles that allow words to be put in multiple locations and
have varied meanings depending on the context. The RBMT approach is applied to the following
language rules: transfer, analysis, and creation. Human language experts and programmers develop
these rules.

Advantages:

 You do not need bilingual text

 Complete control (new law applicable in all cases)
 Reusable (existing language rules can be transferred when paired with new languages)

Disadvantages:

 Need good dictionaries

 Manually set rules (requires expertise)

 Statistical Machine Translation (SMT): 1990-2010

SMT (Statistical Machine Translation) mainly uses or gets trained on existing human translation
known as bilingual text corpus. As we have seen above the RBMT system mainly focuses on word-
based translation but the SMT system focuses on phrase-based translation. The goal of phrase-
based translation is to remove the constraints of word-based translation by translating entire
sequences of words of varying lengths. The word sequences are called phrases, however, they are
generally not linguistic phrases, but rather phrases discovered using statistical approaches from
bilingual text corpora.

Advantages:

 One SMT is suitable for two languages.

 Offline translation without a dictionary: with a proper language model, the translation is
very smooth.

Disadvantages:

 Requires a bilingual corpus

 It is not appropriate for pairs of languages that have significant differences in word order.

Neural Machine Translation (NMT): 2014-Present

NMT is a popular and widely used translation service that incorporates an end-to-end approach for
automatic translation which overcomes the weaknesses of RBMT and SMT methods.

NMT uses the most recent deep learning methods to produce better translation output than other
traditional Machine Translation solutions. It is the most recent type of machine translation that
employs a neural network that is closely related to the neurons of the human brain, allowing it to
categorize data into various groups and layers.

NMT is a language translation approach that tries to incorporate the context of the sentences or
paragraphs rather than individual words. The NMT system is made up of current multilingual
databases and automated learning mechanisms that contribute to continuous improvement.

Advantages:

 End-to-end models (no pipeline of specific tasks)

Disadvantages:

 Requires a bilingual corpus

 Sometimes unable to resolve unfamiliar words.

 Database Interface

One of the major applications of Natural Language Interface is to allow users to

compose questions in Natural Language and receive the response also in Natural
Language. The idea of using natural language instead of SQL has promoted the
development of new type of processing called Natural Language Interface to
Database (NLIDB).

Definition
A natural language interface to a database (NLIDB) is a system that allows the
user to access information stored in a database by typing requests expressed in
some natural language (e.g. English), or a subset of natural language.

History
Early attempts
First attempts of language interfaces to databases appeared as early as the
sixties. An example of such a system is LUNAR, designed to provide answers to
naturally written questions about the geological analysis of lunar rocks returned
by the Apollo missions.

Another such system is LADDER which could be used with large databases and
could be configured to interface different databases. LADDER used semantic
grammars for each database.

In the mid-eighties NLIDB were a very popular area of research, and efforts were
precisely focused on portability issues and on developing generic interfaces that
could be easily adapted to various databases. A well-known system from these
times is CHAT-80, which was implemented entirely in Prolog.

Problem Faced

Despite the interest and numerous attempts, NLIDBs did not gain the expected
rapid and wide commercial acceptance. In 1985,it was foreseen that “By 1987 a
natural language interface should be a standard option for users of DBMS”.
Instead, most of the commercial products have been discontinued due to their
poor performance and the difficulty for porting and configuring them.

Some examples of such failures from prominent vendors are:

 LanguageAccess, discontinued by IBM,

 English Query developed by Microsoft
 English Wizard (by Linguistic Technology Corporation) which was
discontinued years ago.
 ELF no longer provides technical support and has not been updated for
many years.

The development of successful alternatives to NLIDBs (most notably user

interfaces) and the intrinsic problems of NLIDBs are main reasons for the lack of
acceptance.

Rebirth of Database Interface

With the advent of AI, a rebirth of interest for natural language processing occurs.
Due to the boom of mobile computing, chat and advances in speech processing,
we are now seeing bots (aka. conversational agents) becoming popular, taking
advantage of connected services to carry out tasks.

Also, ubiquity of search-like interfaces made popular by internet search engines

call for the refinement of such search experiences. Search applications have
become very interactive, with smart contextual suggestions, some degree of
comprehension, and fast as-you-type updates which provide continuous
guidance.

People are now used to formulating their information needs using keywords and
browsing through results, and iteratively converging onto what they are looking
for. Related information revealed while seeking the answer is itself often valuable
and is even expected by users. Who hasn’t changed their mind after googling
some provider and thus finding out about another? It is indeed no longer a
question of providing a unique and correct answer to a perfectly formulated and
unambiguous question, but rather act as an “information guide”, a sort of smart
assistant in the presence of unclear needs and vast amounts of data.

[su_quote]It no longer a question of providing a unique and correct answer to a

perfectly formulated and unambiguous question, but rather act as an “information
guide”, a sort of smart assistant in the presence of unclear needs and vast
amounts of data.