CS607 Final Term Solved Subjective with References by Moaaz

CS607- Artificial intelligence · Solved Subjective · From Final term Papers · 10 July,2013 · PSMD01 · 2 mark Questions:

Q1. What is hypothesis space?

Answer: The hypothesis space used by a machine learning system is the set of all hypotheses that might possibly be returned by it. It is typically dened by a hypothesis language, possibly in conjunction with a language bias.

Q2. What is machine learning?

Answer: (Page 160) Generally speaking, the goal of machine learning is to build computer systems that can learn from their experience and adapt to their environments. Obviously, learning is an important aspect or component of intelligence.

Q3. What is a strip?

Answer: (Page 197) STRIP is one of the founding languages developed particularly for planning. Let us understand planning to a better level by seeing what a planning language can represent.

Q4. What is differing between height and depth of a tree?

Answer: Height and depth of a tree is equal but height and depth of a node is not equal because the height is calculated by traversing from leaf to the given node depth is calculated from traversal from root to the given node.

Q2. What do you know about Mamdani’s fuzzy inference method?

Answer: (Page 153) Mamdani's fuzzy inference method is the most commonly seen fuzzy methodology. Mamdani's method as among the first control systems built using fuzzy set theory. It was proposed in 1975 by Ebrahim Mamdani as an attempt to control a steam engine and boiler combination by synthesizing a set of linguistic control rules obtained from experienced human operators. Mamdani's effort was based on Lotfi Zadeh's 1973 paper on fuzzy algorithms for complex systems and decision processes.

Q3. What is the basic key of candidate elimination algorithm?

Answer: (Page 173)

Q4. Predicate action?

Answer: (Page 198) Action is a predicate used to change states. It has three components namely, the predicate itself, the precondition, 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)

Q1. Differ between supervised and unsupervised learning? (5 marks)

Answer: Unsupervised Learning

• The model is not provided with the correct results during the training.
• Can be used to cluster the input data in classes on the basis of their sta?s?cal proper?es only.
• Cluster significance and labeling.
• The labeling can be carried out even if the labels are only available for a small number of objects representative of the desired classes.

Supervised Learning

• Training data includes both the input and the desired results.
• For some examples the correct results (targets) are known and are given in input to the model during the learning process.
• The construction of a proper training, validation and test set (Bok) is crucial.
• These methods are usually fast and accurate.
• Have to be able to generalize: give the correct results when new data are given in input without knowing a priori the target.

2. What are the linear separable line and non linear separable line?5

Answer: (Page 184) There is a whole class of problems which are termed as linearly separable. This name is given to them, because if we were to represent them in the input space, we could classify them using a straight line. The simplest examples are the logical AND or OR. We have drawn them in their input spaces, as this is a simple 2D problem. The upper sloping line in the diagram shows the decision boundary for AND gate, above which, the output is 1, below is 0. The lower sloping line decides for the OR gate similarly.

A single perceptron simply draws a line, which is a hyper plane when the data is more then 2 dimensional. Sometimes there are complex problems (as is the case in real life). The data for these problems cannot be separated into their respective classes by using a single straight line. These problems are not linearly separable.

Another example of linearly non-separable problems is the XOR gate (exclusive OR). This shows how such a small data of just 4 rows, can make it impossible to draw one line decision boundary, which can separate the 1s from 0s.