Scribd
Upload
8 views

Introduction To AI Module 1 Part C

Uploaded by

createmypath12
Copyright
© © All Rights Reserved
Available Formats
Download as DOC, PDF, TXT or read online on Scribd
8 views

Introduction To AI Module 1 Part C

Uploaded by

createmypath12
Copyright
© © All Rights Reserved
Available Formats
Download as DOC, PDF, TXT or read online on Scribd
You are on page 1/ 5

Introduction to Artificial Intelligence

MODULE 1 -PART C
UNIT 1: Introduction: What is AI? The Foundation of Artificial Intelligence
1. Remembering
Question 1: Define and briefly describe the following fundamental concepts related
to Artificial Intelligence (AI) and provide one real-life example for each:
i) Artificial Intelligence (AI)
ii) Machine Learning (ML)
iii) Neural Networks
iv) Natural Language Processing (NLP)
v) Robotics
2. Understanding
Question 2: Explain how deep learning uses mathematical principles to perform a
task such as image recognition.
3. Applying

Question 3: A real estate agent wants to predict the price of a house based on its size. You have
the following data:

| Size (sq ft) | Price ($) |

|--------------|-----------|

| 1500 | 300,000 |

| 1700 | 340,000 |

| 1800 | 360,000 |

| 2000 | 400,000 |

| 2200 | 440,000 |

Using linear regression, predict the price of a house that is 1900 sq ft both manually and using Python.
Question 4: An email classification system uses the frequency of certain keywords to determine if an
email is spam. Given the following data:

| Keyword Frequency | Spam (1) / Not Spam (0) |

|-------------------|-------------------------|

|3 |1 |

|1 |0 |

|4 |1 |

|2 |0 |

|5 |1 |

Using logistic regression, determine the probability that an email with a keyword frequency of 3 will
be classified as spam.

UNIT 2: The History of Artificial Intelligence, The State of the Art


Question 4: Discuss the Chronicles of Artificial Intelligence Advancements.
Question 5: Discuss the Current Advancements in Artificial Intelligence.

Question 6: A neural network is trained to classify images and achieves an error rate of 15% on
the training set and 20% on the test set. If the training set contains 10,000 images and the test
set contains 5,000 images, calculate the number of misclassified images in both sets both
manually and using Python commands.

Question 7: In a healthcare setting, a hospital administrator wants to analyze patient


satisfaction survey data to determine the overall satisfaction score based on different
criteria. The survey data is collected on a scale of 1 to 5 for cleanliness with scores (4,
5, 3, 4, 5); staff behavior with scores (5, 4, 5, 3, 4): treatment effectiveness with
scores (4, 4, 5, 3, 5) and overall experience with scores (5, 5, 4, 4, 3). Compute the
average satisfaction score for each criterion and overall.

UNIT 3: Risks and Benefits of AI

Question 8: A company uses an AI model to classify emails as spam or not spam.


Given the confusion matrix with values (TP, FP, FN, TN) of the model's predictions,
calculate the precision, recall, and F1-score. Defend this work on using manual
computation and using python commands

Question 9: Given the following data for pollutants and their respective
concentrations in a city, predict the AQI using a simple linear regression model. The
AQI is influenced by two main pollutants: PM2.5 and O3 (Ozone). Model the
relationship between AQI and the pollutants. The required data are given below:

PM2.5 (µg/m³) O3 (ppb) AQI

55 40 150

78 60 200

90 80 250

100 90 280

120 100 310

Using this data:


1. Manually calculate the coefficients a and b, and the intercept c.
2. Predict the AQI for the following pollutant levels: PM2.5 = 85 µg/m³ and O3 =
70 ppb.

Question 10: Discuss the potential risks of AI in terms of data privacy and security.
Provide examples to illustrate your points.
Question 11: Evaluate the benefits of AI in healthcare, with specific examples of how
AI is currently being used to improve patient outcomes.

UNIT 4 : Intelligent Agents: Agents and Environments

Question 12: Consider a small online store that sells three types of products: Books,
Electronics, and Clothing. The store wants to implement a recommendation system
based on customers' past purchases. The logic is as follows:
(i) If a customer has purchased a book, recommend a related book.
(ii) If a customer has purchased electronics, recommend an accessory for that
electronic item.
(iii) If a customer has purchased clothing, recommend a complementary clothing
item.
Suppose that a customer purchases "The Great Gatsby"; "Smartphone" and "T-Shirt"
in Books, Electronics and Clothing respectively, then:
a. Create a knowledge base to represent the rules and facts.
b. Determine the recommendations manually.
c. Implement the logic in Python to automate the recommendation process.
Find the output manually and using python commands.

4. Analyzing
Question 13: Compare the use of AI in recommendation systems versus autonomous
vehicles in terms of their interaction with the environment.

Question 14: A doctor prescribes a medication that needs to be administered based


on the patient's weight. The dosage is 5 mg per kilogram of body weight. If a patient
weighs 70 kg, how much medication should be administered. Compute it both
manually and using python commands.

UNIT 5: Good Behavior: The Concept of Rationality


5. Evaluating
Question 15: Evaluate the advantages and disadvantages of using AI in healthcare
diagnostics.
6. Creating
Question 16: Design a simple machine learning algorithm using logistic regression to
classify emails as spam or not spam. Outline the steps and mathematical principles
involved.
Question 17: The data below represents concentrations of pollutants and
corresponding Water Quality Index (WQI) values for several water samples:

Sample pH Dissolved Biochemical Nitrate Phosphate WQI


Oxygen Oxygen (mg/L) (mg/L)
(mg/L) Demand
(mg/L)

1 7.0 8.0 5.0 2.0 0.5 85

2 6.5 7.5 6.0 2.5 0.6 80

3 7.2 8.5 4.5 1.8 0.4 90

4 7.5 9.0 4.0 1.5 0.3 95

5 6.8 7.0 6.5 2.8 0.7 75

Using the given data, predict the WQI using both manual approach and machine
learning model implementing Python.

You might also like