Scribd
Upload
18 views4 pages

AIQuiz

Questions Regarding Artificial Intelligence

Uploaded by

felixrotich57
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PDF, TXT or read online on Scribd
18 views4 pages

AIQuiz

Questions Regarding Artificial Intelligence

Uploaded by

felixrotich57
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PDF, TXT or read online on Scribd
You are on page 1/ 4

a) Explain two machine learning problems that can be implemented in AI

systems.

1. Classification Problem: Classification involves predicting a category label based on


input data. This is commonly used in image recognition, spam email detection, and
medical diagnosis, where an AI system classifies data into distinct categories.
2. Regression Problem: Regression is used to predict a continuous output based on
input data. For example, AI systems can use regression to predict stock prices,
forecast weather, or determine house prices based on various features.

b) Discuss the Turing test in relation to intelligent systems.

The Turing Test, proposed by Alan Turing, is a measure of a machine’s ability to exhibit
intelligent behavior indistinguishable from that of a human. In this test, a human judge
interacts with a machine and a human without knowing which is which. If the judge cannot
reliably tell the difference, the machine is considered to exhibit "intelligence." It is significant
in evaluating intelligent systems in AI, particularly in natural language processing and
conversational AI.

c) State two types of knowledge that can be used in intelligent systems.

1. Declarative Knowledge: Facts or information explicitly stored, such as "Paris is the


capital of France."
2. Procedural Knowledge: Knowledge about how to perform tasks, such as algorithms
or procedures for problem-solving.

d) An Expert System

i) Use backward chaining to validate the goal (Jack should invest in shares).

To validate "Jack should invest in shares," we start from the goal and use backward
chaining:

1. Goal: "Jack should invest in shares" matches R5's condition ("if a person wants to
invest in stocks, then they should invest in shares").
2. To satisfy R5, Jack should "invest in stocks."
3. R2 and R3 suggest conditions to invest in stocks. R3 applies since Jack is 25 years
old, which satisfies "younger than 30 but older than 22."
4. Since R3 is true, Jack should invest in stocks, fulfilling R5, and thus, "Jack should
invest in shares."

ii) Rewrite rules 1 and 2 in propositional logic form

● R1: HasInvestment∧BasicKnowledge→InvestInSecurities\text{HasInvestment} \land


\text{BasicKnowledge} \rightarrow
\text{InvestInSecurities}HasInvestment∧BasicKnowledge→InvestInSecurities
● R2: HighIncome∧BasicKnowledge→InvestInStocks\text{HighIncome} \land
\text{BasicKnowledge} \rightarrow
\text{InvestInStocks}HighIncome∧BasicKnowledge→InvestInStocks

iii) Rewrite rules 3, 4, and 5 into FOPL

● R3: ∀x (Age(x)>22∧Age(x)<30→InvestInStocks(x))\forall x \ ( \text{Age}(x) > 22


\land \text{Age}(x) < 30 \rightarrow \text{InvestInStocks}(x))∀x
(Age(x)>22∧Age(x)<30→InvestInStocks(x))
● R4: ∀x (Age(x)>22∧Age(x)<30→BasicKnowledge(x))\forall x \ (\text{Age}(x) > 22
\land \text{Age}(x) < 30 \rightarrow \text{BasicKnowledge}(x))∀x
(Age(x)>22∧Age(x)<30→BasicKnowledge(x))
● R5: ∀x (InvestInStocks(x)→InvestInKenGenShares(x))\forall x \
(\text{InvestInStocks}(x) \rightarrow \text{InvestInKenGenShares}(x))∀x
(InvestInStocks(x)→InvestInKenGenShares(x))

e) Explain types of knowledge that can be used in intelligent systems

1. Heuristic Knowledge: Knowledge gained from experience or intuition that can guide
problem-solving or decision-making.
2. Structural Knowledge: Understanding the relationships between concepts, often
represented as graphs or hierarchies, that allows systems to organize and relate
information.

f) Distinguish between modus ponens and modus tollens using


examples in each case

1. Modus Ponens: This is an inference rule where if "If P then Q" is true and P is true,
then Q must also be true.
○ Example: If it is raining (P), then the ground is wet (Q). It is raining, so the
ground is wet.
2. Modus Tollens: This is an inference rule where if "If P then Q" is true and Q is false,
then P must also be false.
○ Example: If it is raining (P), then the ground is wet (Q). The ground is not wet,
so it is not raining.

g) Explain two applications that have been used to run or develop


intelligent systems.

1. IBM Watson: IBM’s AI platform that supports natural language processing and
machine learning, allowing organizations to analyze large datasets, perform medical
diagnostics, and develop decision-support systems.
2. Microsoft Azure AI: A suite of AI services that provides tools for image recognition,
natural language understanding, and building machine learning models for
applications in healthcare, finance, and retail.
h) Draw the truth table for the following statement: ¬(P ∧ Q) → ¬P ∧ ¬Q
P Q P∧ ¬(P ∧ ¬P ¬Q ¬P ∧ ¬(P ∧ Q) → ¬P ∧
Q Q) ¬Q ¬Q

T T T F F F F T

T F F T F T F F

F T F T T F F F

F F F T T T T T

i) Medical Diagnosis and Supervised Learning

i) How a system can be trained to do medical diagnosis through supervised learning

In supervised learning for medical diagnosis, a model is trained on a labeled dataset


containing historical medical records, where each record includes symptoms, diagnostic test
results, and the correct diagnosis. The system learns patterns in the data to predict a
diagnosis for new patients based on their symptoms and test results. Regular retraining with
updated data helps improve accuracy.

ii) Explain any three supervised machine learning algorithms

1. Decision Trees: A model that makes predictions based on a tree-like structure


where each node represents a decision based on input features, helping with clear,
interpretable results.
2. Support Vector Machines (SVM): A classification algorithm that finds the optimal
hyperplane to separate classes in the feature space, effective for high-dimensional
data.
3. k-Nearest Neighbors (k-NN): A non-parametric algorithm that classifies a data point
based on the majority class of its k-nearest neighbors, commonly used for simple
classification tasks.

iii) Developing a Case-Based Reasoning System

To develop a Case-Based Reasoning (CBR) system, follow these steps:

1. Problem Identification: Define the problem domain and identify requirements.


2. Case Collection: Gather cases, which are past instances or solutions relevant to the
problem.
3. Case Representation: Structure cases in a database with details such as conditions,
solutions, and outcomes.
4. Retrieval Mechanism: Implement methods to retrieve similar cases when a new
problem arises.
5. Adaptation and Solution: Adjust retrieved solutions to fit new problem contexts.
6. Evaluation and Learning: Evaluate the solution, and if successful, store it as a new
case for future reference.

You might also like