• Tidak ada hasil yang ditemukan

CSE -2010-2011 (L-4,T-2 ).pdf

N/A
N/A
Protected

Academic year: 2023

Membagikan "CSE -2010-2011 (L-4,T-2 ).pdf"

Copied!
37
0
0

Teks penuh

A multiple-step income statement (ii) An owners' equity statement. ii) Quick (acid test) ratio, (iii). The following information is taken from the accounting records of Salton Manufacturing Company for the last year ended 30 June 2011. i) Prepare a schedule of cost of goods sold (ii) Prepare income statement. a) What are the different methods of allocating overhead costs of service departments. The following data relates to 20 11. Fixed costs of operating the computer facility of the company Practical capacity.

Required: Allocate costs between two departments using the double-rate allocation method. a) What is the break-even point. Prepare an income statement if the selling price increases by Tk. 2.00 per unit, fixed costs increase by Tk. iv) Calculate the BEP in units if selling price increases by 10% and variable costs increase by 15%. Calculate the margin of safety in units and values. vi) Calculate the number of units sold if the target profit after tax is Tk. vii) Number of computer units sold if the target profit is Tk.

Variable operating cost per unit Total fixed operating cost. i) Prepare income statements under the variable costing method and the absorption costing method. ii) Calculate the percentage of operating income under each method. iii) Explain the difference in using the two methods. The insurance expires in the amount of Tk. 500 supplies were used during the month. The following information was obtained from the accounting records of Salton Manufacturing Company for the last year ended June 30, 2011.

Rent for factory construction. i) Prepare a schedule of cost of goods sold (ii) Prepare income statement. a) What are the different methods of dividing the general cost of service departments.

HUM 371

T.•2/CSE Date: 26/12/2012

SECTION-A

5) (lOX)

What are the factors that affect the supply curve of a good? cardinal and ordinal utility. utility theory .of consumer behavior. BANGLADESH UNIVERSITY OF ENGINEERING AND TECHNOLOGY, DHAKA L-2/T-2 B. Figures in the margins indicate full marks. Give the Ford-Fulkerson algorithm for finding the maximum flow in a flow network and analyze the time complexity of the algorithm.

Also show the order of the matrices for multiplication by the. minimum number of scalar multiplications.

CSE 207

Consider the following approaches to service the maximum number. i) Select the task with the earliest start time (ii) Select the task with the shortest duration. iii) Select the job with the largest number of non-conflicting jobs. For each of the approaches mentioned above, give an example where the machine does not perform the maximum number of tasks. The table contains information about the start and end times of jobs waiting to be serviced by the machine.

Apply the greedy earliest start time approach in Table 1 and show each step of your simulation. BANGLADESH UNIVERSITY OF ENGINEERING AND TECHNOLOGY, DHAKA L-4/T-2 B. Figures in difference indicate full marks. HMM, eg, number of hidden and visible states. is used to determine the proximity between two groups. used in hierarchical clustering 'cluster mean'.

It depends on the order of the maxmim algorithm. c) Under what conditions and why is the k-means halving algorithm guaranteed to produce.

CSE 473

For each iteration, state the value of the weight vector, the decision boundary g(x), and the samples that are. What happens if its value is large. the distance between the word 'poem' and its misspelled version 'poten'. For each polyhedron formed by the lines, determine the vertex of the hypercube to which it will be mapped by the first detections of the hidden layers (with the gradual activation function) of the MLP.

Give an example of how each polyhedron should be assigned one (out of two) class if. i) a two-layer network is sufficient to achieve perfect classification and (ii) a three-layer network is required. BANGLADESH UNIVERSITY OF ENGINEERING AND TECHNOLOGY, DHAKA L-4/T-2 B. The figures in the margin indicate full marks. Describe how this algorithm would be useful in obtaining a linear-time algorithm for triangulating a simple polygon without gaps.

Describe Chan's algorithm for convex hull in two dimensions and derive the running time. a) Prove that for a set of points in two dimensions, the minimum spanning tree is a subset of its Delaunay triangle. Use this theorem to give an O(n2)-time incremental algorithm for computing the order of lines in the plane. Given n points in one dimension, their height-balanced binary search tree, and a search range (low, high]), describe an O(k + n long n)-time algorithm for reporting the all points.k in the interval.

SECTION-B

CSE 463

Write the string production rule using the symbols 'F' , '-' and, '+' ; where the symbols keep their usual meaning.

CSE 409

L (a) Discuss three computer applications to which machine learning approaches appear. appropriate and three that they seem inappropriate for. 7) (c) Explain with a suitable: example how we can avoid over- and under-fitting. can we use such a dimension to measure the complexity of a learning machine. a) Consider the following set of training examples:. i) what is the entropy of this collection of examples with respect to the objective function classification. ii).

FIgure for Question 4(91
FIgure for Question 4(91

CSE 471

How can you modify the training rule to ensure convergence of Q-learning in a non-deterministic environment. How does reinforcement learning problem differ from other function approximation tasks. a) Suppose two different algorithms Ml and M2 were evaluated using cross-validation technique, where the total number of samples in the data set is 500 and k is 10. The difference in sample error in the two algorithms was found and is shown in the following table:.

Now calculate the value of the 95% confidence interval for the true difference in true error between the two algorithms. What are the benefits of lazy. teaching methods over enthusiastic methods. when we try to estimate the future accuracy of a hypothesis based on only a limited set of data. e) What are the limitations of the Find-S algorithm. There are FOUR questions in this section. a) Give the definition and examples of LCS and CLCS.

Give a randomized algorithm to determine the min-cut where the probability that no min-cut is found in a graph is less than !,f: is a very small positive number.

CSE 461

There are FOUR questions in this section. a) Discuss three computer applications for which machine learning approaches occur. appropriate and three for which they seem inappropriate. How can we use such a dimension to measure the complexity of a learning machine. a) Consider the following set of training examples:. i) what is the entropy of this set of examples with respect to the target function classification. ii). What are the types of problems that can be solved with neural networks. advantages and disadvantages of such networks.

Use the CANDIDATE-ELIMINA nON algorithm to learn the version space of the target concept based on the following training examples:. How does reinforcement learning problem differ from other function approximation tasks. a) Suppose two different algorithms MI and M2 were evaluated using cross-validation technique where the total number of examples in the data set is 500 and k is 10. Now calculate the value of 95% confidence interval for the true difference of true error between the two algorithms.

Gambar

Figure for Question No~4(c)
FIgure for Question 4(91

Referensi

Garis besar

Dokumen terkait

Quantity Unit Description Unit Cost Total Cost 1 1 unit PaBX/PBX/Telephone Private atomatic branch exchange system, at least 4 port for direct line, at least 10 post for local lines,

Republic of the Philippines : Appropriation : Source of Fund : : Target Start Date : : Total Project Duration : Project Description : MOBILIZATION/DEMOBILIZATION SQ.M - LOT -