Category: Knowledge in AI

Knowledge Representation (KR) plays an essential role in artificial intelligence that enables systems to organize and interpret information in a way similar to human thinking.

This allows AI systems to process data, make decisions, and tackle problems by keeping knowledge in a structure form. Just as humans use language or symbols to express thoughts, AI needs structures to represent the world around it.

What is Knowledge Representation in AI

Knowledge Representation in AI uses the methods and frameworks to encode and store knowledge making it accessible to reason and make decisions. It allows machines to process and use information to understand the world, solve problems, and learn from experiences.

• KR has an impact on simplifying and structuring large amounts of complex data, which makes it easier to analyze and apply AI.
• Good knowledge representation can boost machine learning by offering improved data structures that the system can employ to learn and recognize patterns.
• Knowledge representation frameworks can be used in different problem areas, like healthcare, robotics, or finance letting AI work in various contexts.
• By arranging knowledge well, KR helps make decisions faster and with more information.

What to represent in AI

Following are the kind of knowledge which needs to be represented in AI systems −

• Object Knowledge: Information about physical objects and their properties, for example, “A laptop has keyboard, mouse, screen”. or “A tree has branches, leaves, and roots. “This helps AI recognize and classify things in its environment.
• Event Knowledge: Knowledge about actions or events, for example, “A traffic light turns red” or “A user clicks a button.” This helps AI understand cause and effect.
• Performance Knowledge: Knowledge of how to do tasks, like “How to Bake a Cake” or “How to Troubleshoot a Machine.” It gives AI the efficiency to accomplish tasks.
• Meta Knowledge: Knowledge of what the system knows, such as understanding when to update the knowledge base or determining which information is most useful in a given context.
• Factual Knowledge: Verifiable statements or facts, such as “Water boils at 100C.” It serves as the basis of AI reasoning and decision-making.
• Knowledge base: A centralized source of maintaining facts, rules, and procedures, which enables the AI system to take decisions and address problems based on relevant knowledge.

Relation between Knowledge and Intelligence

Knowledge and intelligence are important concepts in Artificial Intelligence. Knowledge gives the facts and information needed for reasoning and to solve problems, while intelligence use that knowledge to fix problems, decide things, and adjust to new situations. An AI system that has more knowledge can seem more intelligent by making smart choices based on what it knows.

Knowledge without intelligence is nothing but having a raw information without the ability to use them while intelligence without knowledge means lacking the information to make smart decisions.

• Knowledge allows AI to assess different options and make decisions on past information and learned patterns.
• Example: Online stores use what customers have looked at before to suggest products.
• An intelligence makes an AI understand and implement knowledge in many different situations, therefore making it an accomplishment machine.

Types of Knowledge in AI

Following are the various types of Knowledge −

Declarative Knowledge

Declarative knowledge refers to facts, statements, or information that describe “what is known” about a domain. It can be described as static since the information can be represented as either assertions or truths.

For example, declarative knowledge has facts like “The sky is blue”, “Delhi is capital of India”, and “A triangle has three sides”. These statements are declarative knowledge because it is a fact that can be directly expressed and documented.

• Declarative knowledge contains various facts and knowledge regarding the world.
• It answers “what” queries rather than “how” to achieve anything.
• This type of knowledge is easy to convey through statements, databases, or documents.
• Examples include scientific truths, historical events, and general knowledge.
• AI uses declarative knowledge for reasoning, decision-making, and problem-solving.

AI Application: In a question-and-answer system, declarative knowledge is utilized to answer factual questions such as “What is the capital of France?”

Procedural Knowledge

Procedural knowledge refers to the knowledge, of how to perform a task. It involves procedures, methods, or processes that needed to achieve a task or solve a problem. It focuses on step-by-step procedures rather than just facts.

For example, solving a quadratic equation is an structured process that includes determining coefficients, applying the quadratic formula, and simplifying the result.

• Procedural knowledge gives step-by-step instructions for how to do a task.
• It explains “how” to do something rather than just stating the facts.
• This knowledge is acquired through practice and experience.
• Procedural knowledge more difficult to convey explicitly than declarative knowledge.
• It is widely used in artificial intelligence applications like automation, robotics, and expert systems.

AI Application −

• In robotics, procedural knowledge is applied when programming a robot to perform duties such as assembling auto-mobiles or navigating in a maze.
• A robot chef follows rules in a step-by-step manner to prepare a meal. “How To Make Tea: Boil water. Add tea leaves. Pour into a cup. Add sugar and milk to taste.”

Meta Knowledge

Meta-knowledge is a term that refers to “knowledge about knowledge.” It allows AI to understand what it knows, how reliable the information is, and when to apply it. This kind of knowledge allows AI systems to evaluate and enhance their reasoning and decision-making abilities.

For instance, if an AI chatbot knows that its answers are sourced from a reliable database, it will become more confident in its responses.

• Meta-knowledge is the capability of AI to evaluate the correctness and reliability of its own knowledge.
• It enables AI to determine whether it should use specific rules or facts.
• Crucial in learning, over time it sharpens the artificial intelligence decision to make.
• It is very useful for debugging AI models as it helps detecting gaps or inconsistencies.
• Meta-knowledge improves problem-solving by helping AI in choosing the best reasoning strategy.

AI application, a self-driving auto-mobile knows traffic laws and has meta-knowledge to recognize false sensor data caused by fog. It can then decide whether to slow down or switch to a backup system.

Heuristic knowledge

Heuristic knowledge is the rule-of-thumb or experience-based knowledge that helps in problem-solving and decision-making when complete information is not available. It helps in decision-making when specific rules or formulae are not available.

• Heuristic knowledge often relies on intuition, experience, or common sense.
• This knowledge is used in AI to make accurate and quick decisions, particularly in complex or unclear situations.
• Heuristic information reduces computation time by directing the search process toward likely solutions.
• It is widely used in applications such as game AI, medical diagnostics, and optimization problems where exact answers are hard to compute.

AI Application: In game-playing AI like chess or Go, heuristic knowledge helps the machine to evaluate board positions and make strategic decisions.

Structural Knowledge

Structural knowledge refers to the relationships and connections between different concepts or things in a domain. It helps AI understand how things are related.

• Structural knowledge describes how things are structured and connected.
• This knowledge is often represented as graphs, trees, or networks.
• Structural knowledge is widely used in many applications of artificial intelligence, for example, in semantic networks, ontologies, and knowledge graphs.
• This type of knowledge assists AI in inferring new relationships from the existing ones which improves future decision making process quick and accurate.
• Structural knowledge strengthens reasoning and decision-making by presenting information in an organized manner.

For example, an AI-driven medical diagnosis system understands, A fever is a symptom of flu. Flu is caused by a viral infection. Antiviral drugs can treat viral infections this way it connects different entities in a domain.

Humans are intelligent and do tasks which requires creativity by using the logic to solve problems, learn form their past experiences and change when the things are new. They solve the problems or handle the situations with the knowledge stored in their minds.

On the other hand, Artificial Intelligence stores this kind of intelligence in Knowledge-Based Agents (KBAs). These agents make decisions through reasoning by storing facts, rules, and relationships in a knowledge base. Unlike humans, they can analyze vast volumes of data quickly and accurately.

Knowledge Based Agents in Artificial Intelligence

In Artificial Intelligence, Knowledge-Based Agents(KBA) use stored information and reasoning techniques to make intelligent decisions and solve problems. They are designed to do complex tasks effectively by combining logic and organized information.

• Knowledge based agents depend on a knowledge base which stores facts, rules, and information about the world.
• They use an inference system for reasoning and draw conclusions from the knowledge base.
• KBAs can adapt to new situations by adding or updating their knowledge.
• KBAs are commonly used in problem-solving, decision-making, planning tasks and in domains like healthcare, education, and legal systems.
• KBAs improve the scalability and accuracy of decision-making by handling large amounts of structured data.

Example of Knowledge Based Agents

Health Care: Knowledge-based agents analyze patient’s data and medical history to suggest diagnoses or treatments which helps the doctor make wiser decisions. For instance, AI tools can identify patterns for the diagnosis of diseases like cancer and diabetes.

Business Intelligence: AI tools analyze large amounts of data to provide insights on market trends, customer behavior, and productivity. Companies use them to plan marketing campaigns and improve efficiency.

Architecture of Knowledge-Based Agents

First, the knowledge based agent receives the data from environment through sensors then it retrieves the relevant information, rules, facts form the knowledge base to understand the context. It then processes the relevant information through inference system using techniques like forward or backward chaining. Based on this reasoning, agent decides the best action.

Once the decision is made, the agent executes the action using actuators to affect the environment. If the agent is capable of learning, it updates the new information to the knowledge base for future tasks.

What is Knowledge base?

Knowledge base is the critical component of the Knowledge-Based Agent, which includes all the information, rules, and facts necessary for reasoning and decision.

It gives the agent all the information it needs to reason, make informed judgments, and act appropriately with reference to its environment.

• Facts are information about the world. For example, “The Earth orbits the Sun once every 365.25 days”.
• Rules are logical statements that relate facts. For example, “If the temperature drops below 0C, water will freeze into ice”.
• Updates: The knowledge base can be created or updated during the process in which an agent learns or introduces new knowledge so that over a period, its decision-making skill is increased.

For example, smart thermostat saves information as “Room temperature needs to be at 22C”, “If the temperature of room has exceeded to more than 25C then turn on air conditioner”.

Inference System

The inference engine proceeds to process knowledge in the knowledge base using logical reasoning to help the agent take intelligent decision and solve the problem. Imagine this as a tool to determine “what is next” from facts and regulations.

The inference system acts like the thinking part of the agent. It begins with a set of given facts and uses reasoning to obtain new facts or decide what to do.

This system also creates new knowledge from the existing information and includes them in the knowledge base to assist future decision making.

• In practical, an inference system is constructed around algorithms and logical representations such as propositional or first-order logic.
• Inference systems are used in many applications, such as expert systems, natural language processing, and automated reasoning. This allows machines to think logically and make judgments like humans.

This technique uses two basic approaches to introduce logic −

• Forward Chaining: This is based on the assumptions given. The inference engine checks the rules, applies them, and keep inferring the new facts until it reaches a conclusion or goal.
• Backward Chaining: Start from the goal and work backward. It determines which facts or conditions need to be satisfied to achieve that goal.

For example, the thermostat checks the current temperature, applies the criteria, and concludes, “The room is 27C, so I need to activate the cooling system.” This decision is made by inference system.

Sensors and Actuators

In knowledge-based agent sensors collect environmental information such as temperature, light, and movement, while actuators use that information to perform actions such as moving, turning, or cleaning. Robot vacuums use sensors to look for dirt and actuators to clean it or to avoid obstacles.

Operations Performed by KBA

Knowledge-based agents utilize three key procedures to take intelligent actions −

• TELL: The agent updates its knowledge base with new information from the environment. For example, if a robot sees a door open, it tells the knowledge base, “The door is open.”
• ASK: The agent asks the knowledge base what decision it should take. For example, it may ask, “Should I close the door or move forward?”.
• PERFORM: The agent will follow the guidance of the knowledge base, such as closing the door or moving forward. It performs the selected option.

These processes enable the agent to learn, make smart judgments, and efficiently adapt to its surroundings.

A Simple Knowledge Based Agent

Knowledge-Based Agent (KBA) processes perceptions, reasons using its knowledge base, and makes decisions. The following pseudocode represents a basic Knowledge-Based Agent −

function KB-AGENT(percept):  
persistent: KB, a knowledge base   
          t, a counter, initially 0, indicating time   
TELL(KB, MAKE-PERCEPT-SENTENCE(percept, t))   
Action = ASK(KB, MAKE-ACTION-QUERY(t))   
TELL(KB, MAKE-ACTION-SENTENCE(action, t))  
 t = t +1return action   

How a Knowledge-Based Agents works

A Knowledge-Based Agent continuously perceives its environment, utilizes stored knowledge for reasoning, and makes informed decisions. Below is the explanation of how it operates −

• MAKE-PERCEPT-SENTENCE: The agent updates its knowledge base with whatever it observes(environment) at time t by converting the percept into a logical sentence.
• MAKE-ACTION-QUERY: The agent looks in its knowledge base to determine which action it should execute given the current time step and latest knowledge.
• MAKE-ACTION-SENTENCE: After choosing an action the agent updates its knowledge base with a logical statement that describes the action which the agent intends to execute.
• The function returns the chosen action, which the agent will carry out in the environment.

Various levels of Knowledge Based Agent

A knowledge-based agent can be understood at multiple levels, each explaining a distinct aspect of how it perceives, processed, and acted. Following are the various level of Knowledge based agent −

Knowledge level

At this level, the focus is on what the agent knows. The agent make decisions from the facts, rules, and the logical relationship found in its knowledge base.

For example, In a sprinkler activating system the knowledge base has information like “the soil is dry” as well as “if the soil is dry then turn on sprinkler.” This describes what the system knows.

Logical Level

This level describes how the knowledge is represented and processed. The agent uses logic like that of propositional or first-order logic to infer, derive new information and to make decisions. Rule: “The soil is dry, therefore turn the sprinkler on. So the logical level will conclude “Turn the sprinkler on.”

Implementation level

This is the physical or software implementation of the agent. It uses programming languages, algorithms, and hardware to implement the knowledge and reasoning processes. The system is implemented with a soil moisture sensor, sprinkler actuator, and a program written in a simple language like Python that processes the sensor data and controls the sprinkler.

Future of Knowledge-Based Agents

Knowledge-Based Agents (KBAs) are changing with the evolution of artificial intelligence, using machine learning, natural language processing, and automated reasoning to reason with complex data, learn from new data, and make intelligent decisions. KBAs are used extensively in industries to automate and improve efficiency.

KBAs will use deep learning in the future to improve reasoning and manage uncertainty, and hence become indispensable in healthcare, finance, cybersecurity, and intelligent automation.