Periods Per Week (60 Min. Each) Practical 02 Tutorial - Hours Marks Evaluation System Theory 03 100 Oral - 25 Term Work - 25 Total - 150
Uploaded by
dil17Periods Per Week (60 Min. Each) Practical 02 Tutorial - Hours Marks Evaluation System Theory 03 100 Oral - 25 Term Work - 25 Total - 150
Uploaded by
dil17Class: B.E.
University of Mumbai Branch: Instrumentation
Semester: VII
Elective-I: Expert Systems (Abbreviated as ES) Periods per Week Lecture 04 (60 min. each) Practical 02 Tutorial -Hours Evaluation System Theory 03 Oral --Term Work --Total --Module 1 Contents
Marks 100 25 25 150
Hours 3
Introduction to Expert System What are Expert Systems, Features of Expert System, features of good Expert System, Role of human in Expert System, Expert System organization, Difference between expert system and conventional program, Basic activities of expert system and the areas in which they solve problems, Prospector systems-features, working Expert System Tools Knowledge representation in expert systems-using rules semantic nets, frames, Types of tools available for expert system building and how they are used, Stages in the development of expert system tools, Examples of knowledge engineering Building an Expert Systems Necessary requirements for expert systems development, Justification for expert system development, Task in building expert systems, Stages of expert system development, Choosing a tool for building expert system, Acquiring knowledge from the experts, Examples of the expert systembuilding process, Examples of expert system used in different areas Difficulties in developing an expert system Common pitfalls in-planning an expert system, Scaling with the domain expert during development process. Neural Network Introduction Biological neural networks-neuron physiology, eyes neural network, Artificial neuron models, Neural net architecture (Topologies), Learning in ANN, Characteristics of ANN
26
6 Supervised Learning Single layer networks-perceptrons, Linear separability, Perceptron training algorithm, Modifications multi-layer network-multi-layer discrimination, back propagation algorithm, Classification using back propagation algorithm, Application of back propagation algorithm, Adaptive multiplier networks-algorithm, boosting, prediction network, radial basis functions 7 5 Unsupervised Learning Winner-take all networks, Learning vector quantiser, Co-inter propagation networks, Adaptive response theory 8 5 Associative Networks Non-interactive procedures for association, Hopfield networks, Optimization using Hopfield networks, Brain state in a box network, Boltzmann machines, Hetero associators 9 6 Fuzzy Logic Propositional logic, Membership functions, Fuzzy logic, Fuzzy rule generation, De-fuzzification, Time dependent fuzzy logic, Temporal fuzzy logics, Case study-to use fuzzy logic for processes control problem Theory Examination: 1. Question paper will consist of total 7 questions carrying 20 marks each. 2. Only 5 questions need to be attempted. 3. Q.1 will be compulsory and based on the entire syllabus. 4. Remaining questions will be mixed in nature. 5. In question paper, weightage of each module will be proportional to the number of respective lecture hours as mentioned in the syllabus.
Oral Examination:
Oral examination will be based on entire subject.
Term Work:
Term work consists of minimum six experiments/assignments and a written test. The distribution of the term work shall be as follows, Laboratory work (Experiments and Journal) :15 marks Test (at least one) :10 marks The final certification and acceptance of term work ensures the satisfactory performance of laboratory work and minimum passing in the term-work
Text Books:
1. Stamatios V. Kartalopolous, Understanding Neural Network and Fuzzy Logic, PHI Pvt Ltd. 2. Kishan Mehrotra, Elements of ANN, 2nd Editon, Penram International Publishing (I) Pvt. Ltd.
Reference Books:
27
1. Donald A. Waterman, A Guide to Expert Systems, Addison-Wesley Publishing Company.
28
