Pune University Previous Year Papers
B.E. (Information Technology) NEURAL NETWORK AND EXPERT SYSTEM
(Elective – IV) (Sem. – II)
Time :3 Hours] [Max. Marks :100
Instructions to the candidates:
1) Answers to the TWO sections should be written in SEPARATE sheet.
2) Use of logarithmic tables, slide rules and electronic pocket calculator is allowed.
3) Neat diagrams must be drawn wherever necessary.
4) Pigures to the right indicate full marks.
5) Assume suitable data, if necessary.
SECTION – I
QI) a) Describe features of the biological neural network that make it superior to the most sophisticated Artificial Intelligence computer system for pattern recognition tasks. 
b) Explain, with examples, differences between the following pattern recognition tasks: 
i) Association vs classification.
ii) Classification vs mapping.
Q2) a) Compare the performance of a computer and that of a biological neural network in terms of speed of processing, size and complexity, storage, fault tolerance and control mechanism. 
b) With the help of suitable diagram discuss functioning of a simple artificial neuron. Explain how the functionality is affected if two such neuron are connected in series. 
Q3) a) Draw and explain Rosenablatt’s perceptron model of a neuron. Write the equations which describe the operation of the perceptron model of a neuron. 
b) What does Learning laws describes? Explain Perceptron Learning Law. 
What are the main differences among the three models of artificial neuron, namely, McCulloch-Pitts, perceptron and ADALINE? 
What is conjugate gradient method? Comment on the performance of the conjugate-gradient method. 
Draw and explain architecture of Radial Basis Function (RBF) Networks.
Explain how Support Vector Machines is used for pattern classification and non-linear regression? 
Explain Radial basis function network for function approximation, with one output unit and L hidden units. Draw suitable diagram. 
What is the basic concept of Relevance Vector Machines? Explain how it is used for classification problem? 
SECTION – II
Explain the principal goal of self-organizing map (SOM). Explain three essential processes involved in the formation of the self-organizing map.
What are the salient features of the Kohonen’s self-organizing learning algorithm? 
Explain architecture of a Boltzmann Machine. Illustration it with suitable diagram. 
What is Hopfield model of a neural network? What is meant by capacity of feedback network? 
What are the main advantages in keeping the knowledge base separate from control module in knowledge-based systems? 
Identify and describe two good application areas for an expert system within a university environment. 
What is blackboard system architecture? Draw diagram to explain three functional components of blackboard system. 
What is uncertainty? Explain at least two approaches which are used to deal with uncertainly problem. 
QII) a) What is the need of knowledge system building tools for expert system?
Which features and capabilities a developer should consider when evaluating building tools for expert system development? 
b) Explain how MYCIN helps diagnoses infectious blood diseases and recommend therapy for patients. 
Q12) a) Explain how DENDRAL determine molecular structure of an unknown compound. Write prominent features of DENDRAL. 
What do you mean by knowledge engineering? Explain various stages of knowledge acquisition.