Artificial Intelligence

Krishna Engineering College (ECS-801)

B.Tech.    Computer Science

Semester - 8 

Pre-Sem Exam May - 2014

ARTIFICIAL INTELLIGENCE

Time : 3 hrs

MM : 100

Q.1  Attempt any four of the following questions (5 * 4 = 20)

1. Define and describe the difference between knowledge, belief hypothesis and data.
2. Define Artificial Intelligence and write about application areas of Artificial Intelligence.
3. Define Intelligent Agents and explain the various structures of intelligent agents.
4. What is means-ends analysis? Explain it with examples.
5. Explain the various steps of natural language processing.
6. Briefly explain the parse tree & parse the sentence. "John broke the window with a hammer"

Q.2  Attempt any four of the following questions (5 * 4 = 20)

1. Explain the various uninformed search strategies in brief and compare them.
2. Explain the A* algorithm with an example.
3. Explain the RBFS algorithms in brief.
4. What is a local search? Explain the various forms of Hill climbing with example.
5. What is an adversarial search?Explain the Min-Max Search procedure with Alpha-Beta pruning.

Q.3  Attempt any two of the following questions (10 * 2 = 20)

1. Explain the First order logic, Forward & Backward chaining and Probabilistic reasoning.
2. (i)  Explain the Bayesian Networks (ii) Explain the Hidden Markov Models (HMM).
3. Consider the following sentences:

• Marcus was a man.
• Marcus was a Pompeian.
• All Pompeian's were Romans.
• Caesar was a ruler.
• All Romans were either loyal to Caesar or hated him.
• Everyone is loyal to someone.
• People only try to assassinate rulers they are not loyal to.
• Marcus tried to assassinate Caesar.
• All men are people.

          Use the first order predicate logic to answer the question, "was Marcus loyal to Caesar?"

     4.  Discuss Bayes rule ans Bayesian network with examples and solve the given medical cancer diagnosis problem:

There are two possible outcomes of a diagnosis : positive and negative. We know 0.8% of world population has cancer. Test gives correct positive result 98% if the time and gives correct negative result 97% of the time. If a patient's test returns positive, should we diagnose the patient as having cancer?

Q.4  Attempt any two of the following questions (10 * 2 = 20)

1. Explain Supervised, Unsupervised & Reinforcement learning in detail. Also explain Decision Tree in learning with example?
2. Explain the Learning with complete data. Naive Bayes models.
3. Briefly explain learning with hidden data : EM algorithm.

Q.5  Attempt any two of the following questions (10 * 2 = 20)

1. What is Pattern Recognition? Explain the process of recognition problem and write down the structure and stages of Pattern Recognition Systems.
2. Explain the Principle Component Analysis (PCA) in detail.
3. Explain the various Classification Techniques - 

• Nearest Neighbor (NN) Rule,
• Bayes Classifier,
• Support Vector Machine (SVM),
• K-means clustering.

****ATCAPH.com****