Current Final Term CS607 Papers Subjective Solved

My today's paper 23/Aug/2016

80% MCQs were from Moazz & Arslan file.

Subjective questions b kuch past sy thy or kuch new....

1.What do you mean by single layer perceptron? (2)

Answer:

           Single layer perceptron can perform pattern classification only on the linearly separable patterns regardless of the type of non-linearity.

                                     ……………………………………..

 2.Write the code in CLIPS to add two digits? (2)
Answer:

            CLIPS> (+ 3 4)

                                    ……………………………………..

3.Write three Advantages of Neural Networks. (2)

Answer:

            Excellent Pattern Recognition

            Excellent classifiers

            Handles noisy data well

            Good for generalization

                                    ……………………………………..

4.Write the code in CLIPS to display "Hello World"? (3)
Answer:

            (printout t “Hello World” crlf)

……………………………………..

5.which topic in AI deals the concept of partial true & membership? (3)
Answer:

            Handling Uncertainty with fuzzy systems

                                    ……………………………………..

6.Artificial neural network how you elaborate the design phase of artificial neural network. (5)

Answer:

·         Feature Representation

·         Training

·         Similarity Measurement

·         Validation

·         Stopping Criteria

·         Application

……………………………………..

7.Different kind of memories need is expert system which memory is best and difference between this memory and knowledge base. (5)

Answer:

Working memory     

The working memory is the ‘part of the expert system that contains the problem facts that are discovered during the session’ according to Durkin. One session in the working memory corresponds to one consultation. During a consultation: 

• User presents some facts about the situation.

 • These are stored in the working memory.

 • Using these and the knowledge stored in the knowledge base, new information is inferred and also added to the working memory.

Knowledge Base 

The knowledge base is the part of an expert system that contains the domain knowledge, i.e.  

• Problem facts, rules • Concepts • Relationships

As discussed in the KRR section, one way of encoding that knowledge is in the form of IF-THEN rules. We saw that such representation is especially conducive to reasoning.

                                    ……………………………………..

8."Goal is predicate used to change the state" Do u agree this statement? give reason
Answer

No, Action is a predicate used to change states. It has three components namely, the predicate itself, the pre-condition, and post-condition predicates. For example, the action to buy something item can be represented as, 

Action:  buy(X) 

Pre-conditions: at(Place) ∧ sells(Place, X) 

Post-conditions/Effect: have(X) 

What this example action says is that to buy any item ‘X’, you have to be (preconditions) at a place ‘Place’ where ‘X’ is sold. And when you apply this operator i.e. buy ‘X’, then the consequence would be that you have item ‘X’ (postconditions).

And Goal is also represented in the same manner as a state.

                        ……………………………………..

9.Genetic algorithm and human brain sy related ak question tha 5 marks ka.
Answer
          The genetic algorithm technology comes from the concept of human evolution. Genetic Algorithms is a search method in which multiple search paths are followed in parallel. At each step, current states of different pairs of these paths are combined to form new paths. This way the search paths don't remain independent, instead they share information with each other and thus try to improve the overall performance of the complete search space.           

         ……………………………….……………………………………………………………………

24 August-2016

Cs607 Current Paper

All Mcqs was from Past Papers … 2 or 3 was new but very easy and simple

Subjective:   Totally from Current Papers:

1. “Boolean logic is a subset of fuzzy logic.” Do you agree with the statement or not? Give reason to support your answer. [Marks 2]
Answer:

Fuzzy logic is a superset of conventional (Boolean) logic that has been extended to handle the concept of partial truth -- truth values between "completely true" and "completely false". For example, There are two persons. Person A is standing on the left of person B. Person A is definitely shorter than person B. But if boolean gauge has only two readings, 1 and 0, then a person can be either all or short. Let’s say if the cut off point is at 5 feet 10 inches then all the people having a height greater than this limit are taller and the rest are short.

                        ……………………………………..

 2.    Write the CLIPS command to remove the fact from working memory? [Marks 2]

Answer:
            The retract command is used to remove or retract facts.

For example:

 (retract 1) removes fact 1

(retract 1 3) removes fact 1 and 3

                        ……………………………………..

3-Repeated

                                    ……………………………………..

4-  From learning which is best for the given statement  

 having seen many cats, all of which have tails, one might conclude that all cats have tails. [Marks 2]

Answer:

       Inductive learning takes examples and generalizes rather than starting with existing knowledge. For example, having seen many cats, all of which have tails, one might conclude that all cats have tails. This is an unsound step of reasoning but it would be impossible to function without using induction to some extent.

                                    ……………………………………..

 5- Write Three Soft-Computing Algorithms? [Marks 3]

Answer:

The most common combinations are of the pairs

 • Genetic algorithms – fuzzy systems (genetic fuzzy)

• Neural Networks – fuzzy systems (neuro-fuzzy systems)

 • Genetic algorithms – Neural Networks (neuro-genetic systems)
                        ……………………………………..

6- Write the Structure of deftemplate relation so that we can use the assert in given

CLIPS>(assert ( father ( fathersName Ahmed) (sonsName Belal) ) )  [Marks 3]

Answer:

           

The Deftemplate  construct defines a relation’s structure

   (deftemplate <relation-name> [<optional   comment>] <slot-definition>

 e.g.   

  CLIPS> ( deftemplate  father “Relation father”

                        (slot fathersName)

                                    (slot sonsName) )

                                    ……………………………………..

7- Which type of ordering in hypothesis spaces is best suited from h <?, ?> hypothesis and towards h <φ, φ> hypothesis (3 marks)

Answer:

            All the hypothesis in h can be ordered according to their generality, starting from the <?,  ?> which is the most general hypothesis since it always classifies all the instances as positive. On the contrary, we have < φ, φ > which is the most specific hypothesis since it doesn’t classify a single instance as positive.

                                    ……………………………………..

 8- One table was given and had to calculate its final fuzzy logic outcome using AND operation (3 marks)

Answer:

                                    ……………………………………..

9- Write Fuzzy inference system 5 points (5 marks)     

Answer:

            1-Fuzzification of the input variables.

            2-Application of fuzzy operator in the antecedent(premises).

            3-Implication from antecedent to consequent.

            4-Aggregation of consequents across the rules.

            5-Defuzzification of Output.

                        ……………………………………..

10- man (ahmed)

Father (ahmed, belal)

Write down the commands to add the above facts into CLIPS. Also write the command to modify the name “ahmed” to “Ahmed Ali”. [5 marks]

Answer:

We add a fact:

 CLIPS>(assert ( father ( fathersName “Ahmed”) (sonsName “Belal”) ) ) 

To modify the fathers name slot, enter the following: 

CLIPS> (modify 2 ( fathersName “Ali Ahmed”)) 

Notice that a new index is assigned to the modified fact

To duplicate a fact, enter: 

CLIPS> (duplicate 2 (name “name”) )            

                                    ……………………………………..

11. In context of Machine Learning, do you agree with the following statement? Justify your answer with reasons.

"If the program gets something right once it will always get it right. If it makes a mistake once it will always make the same mistake every time it runs." [5 marks]

Answer:

            

                                                ……………………………………..

12. Genetic algorithm is inspired by the structure and or the functional aspects of biological neuron and it’s consist of an interconnected groups of anti-neurons.

Is it true or false give reason? [5 marks]

Answer:

             Genetic algorithms are a modern advancement to the hill climbing search based problem solving. Genetic algorithms are inspired by the biological theory of evolution and provide facilities of parallel search agents using collaborative hill climbing. We have seen that many otherwise difficult problems to solve through classical programming or blind search techniques are easily but un deterministically solved using genetic algorithms.   

At this point we introduced the cycle of AI to set base for systematic approach to study contemporary techniques in AI.

………………………………………………………………………………………………………

Q: Unsupervised Methodology (5 marks)        

Answer:                                                                                         

Given a set of examples with no labeling, group them into sets called clusters.

·         A cluster represents some specific underlying patterns in data.

·         Useful for finding patterns in the large data sets.

·         Form clusters of input data.

·         Map output of clusters.

·         Given a new example, find cluster and generate into associated output.

……………………………………..

Q: Neural Networks (5 marks)

Answer:

·                     A neural network is a massively parallel distributed computing system that has the neural propensity for storing experiential knowledge and make it available for use. It resembles the brain in two respects.

·         1-Knowledge is acquired by neural network through learning process (called training).

·         2-Interneuron connection strength known as synaptic weights use to store knowledge.

·         Knowledge in artificial neuron network is implicit and distributed.

……………………………………..

Q: elaborate “similarity Measurement” in the context of ANN? (2 marks)

Answer:

            A measure to tell the difference between the actual output of the network while training and the desired labeled output.

            The most common technique for measuring the total error in each iteration of neural network is Mean Squared Error (MSE)

                                    ……………………………………..

Q: A student wants to use forward chaining. Elaborate possible issue with forward chaining. (2 marks)

Answer:

      Undirected Search:

            In this approach, it has no way of distinguishing important and important facts. So the equal time spent on trivial as well as crucial approach.

      Conflict Resolution:

            In this approach, this is the question of what to do when the premises of two rules match the given facts. Which should be fired first? If we fire both, then we may conflict facts. E.g:

IF you are bored

AND you have no cash

THEN go to a friend’s house

IF you are bored

AND you have cash

THEN go watch a movie

If both rules are fired, you may be conflicting recommendations to the working memory.

           

                ……………………………………..

Q: write difference between Find-S and candidate elimination? (2 marks)
Answer:

            FIND-S outputs a hypothesis from H that is consistent with the training examples, but this is just one of many hypotheses from H that might fit the training data equally well.

            The Candidate-Elimination algorithm represents the version space by storing only its most general members (denoted by G) and its most specific members (denoted by S)

                                    ……………………………………..

Q: write main two branches of “Problem” (2 marks)

 Answer:

There are two main branches of problems:

• Tractable

 • Intractable 

Those problems that can be solved in polynomial time are termed as tractable, the other half is called intractable. The tractable problems are further divided into structured and complex problems. Structured problems are those which have defined steps through which the solution to the problem is reached. Complex problems usually don’t have well-defined steps.

                                   

                                    ……………………………………..

Q: Write command to add 34 in clips

Answer:

CLIPS> (+ 3 4)

                                    ……………………………………..

Q: Write command add fact

Answer:

CLIPS> (assert ( man ( name “Ahmed” ) ) ) 

CLIPS>(assert ( father ( fathersName “Ahmed”) (sonsName “Belal”) ) )

                                    ……………………………………..

Q: What does means Vague?

Answer:

Ours is a vague world. We humans, talk in terms of ‘maybe’, ‘perhaps’, things which cannot be defined with cent percent authority. But on the other hand, conventional computer programs cannot understand natural language as computers cannot work with vague concepts. Statements such as: “Umar is tall”, are difficult for computers to translate into definite rules. On the other hand, “Umar’s height is 162 cm”, doesn’t explicitly state whether Umar is tall or short

                                    ……………………………………..

Q: Write connectionlist.

Answer:

which the focus of algorithms is on training rather than explicit programming. Tasks for which connectionist approach is well suited include:

 • Classification

            • Fruits – Apple or orange

 • Pattern Recognition

 • Finger print, Face recognition

 • Prediction

• Stock market analysis, weather forecast

                        ……………………………………..

Q: When we develop expert system what step we need

Answer:

Expert systems may be used in a host of application areas including diagnosis, interpretation, prescription, design, planning, control, instruction, prediction and simulation. 

The general stages of the expert system development lifecycle or ESDLC are 

• Feasibility study

• Rapid prototyping

• Alpha system (in-house verification)

 • Beta system (tested by users)

 • Maintenance and evolution

                                    ……………………………………..

Q: How you will describe 'Deductive Learning' if you will have to tell someone? [2 marks]

Answer:

Deductive learning works on existing facts and knowledge and deduces new knowledge from the old. This is best illustrated by giving an example. For example, assume:

 A = B

 B = C

 Then we can deduce with much confidence that:

 C = A

                                    ……………………………………..

Q:  Clustering Algorithms [3 marks]

Answer:

           

       The famous clustering algorithms are:

1.    Self-organizing maps (SOM)

2.    k-means,

3.    linear vector quantization,

4.      Density based data analysis, etc.

……………………………………..

Q: How do a Working memory work in expert system? [3marks]
Answer:

            The working memory is the ‘part of the expert system that contains the problem facts that are discovered during the session’ according to Durkin. One session in the working memory corresponds to one consultation. During a consultation: 

• User presents some facts about the situation.

• These are stored in the working memory

. • Using these and the knowledge stored in the knowledge base, new information is inferred and also added to the working memory.

                                    ……………………………………..

Q: Fuzzy Logic work in real life..example.. 

Fuzzy logic represents partial truth

How does it work? Reasoning in fuzzy logic is just a matter of generalizing the familiar yes-no (Boolean) logic. If we give "true" the numerical value of 1 and "false" the numerical value of 0, we're saying that fuzzy logic also permits in between values like 0.2 and 0.7453.  “In fuzzy logic, the truth of any statement becomes matter of degree” We will understand the concept of degree or partial truth by the same example of days of the weekend. Following are some questions and their respective answers: 

– Q: Is Saturday a weekend day? 

– A: 1 (yes, or true)

 – Q: Is Tuesday a weekend day?

– A: 0 (no, or false)

 – Q: Is Friday a weekend day?

– A: 0.7 (for the most part yes, but not completely)

 – Q: Is Sunday a weekend day?

– A: 0.9 (yes, but not quite as much as Saturday)

OR:

I explained Fuzzy Logic first then gave example of Automated washing machine which is decided How much cloths have dirtiness, and how much time is required to wash them.

                                                ……………………………………..