29. What is the postfix of the expression A + B * C / (D - E)?
(A) A B C * D E - / +
(B) A B C D E - / * +
(C) A * B C / D E - +
(D) + A / * B C - D E
(E) + /A * B C - D E

1. A screw factory would like to identify flawed screws by using a machine learning (ML) algorithm. Suppose there are total P flawed screws among all the produced screws, but only n out of all the flawed screws are reported by the ML algorithm. Which one of the following measurements is used to describe the ratio of identified flawed screws (n/P)? (A) F1-score (B) False positives (C) Precision (D) Recall (E) True positives

2. Given the binary tree on the right-hand side. What is the result of post order traversal? (A) 8, 31, 35, 47, 61, 83, 84, 91, 96, 49 (B) 31, 8, 47, 35, 61, 83, 91, 96, 84, 49 (C) 31, 61, 8, 47, 83, 91, 96, 84, 35, 49 (D) 8, 31, 35, 47, 49, 61, 83, 84, 91, 96 (E) 31, 61, 91, 8, 47, 83, 96, 35, 84, 49

3. Which of the following statement is INCORRECT? (A) In designing a machine learning system, signal processing methods can be a very valuable data pre-processing tool. (B) When the training and testing errors of a neuron network is becoming larger and larger, we should terminate the training process to avoid the overfitting problem. (C) The key of human intelligence comes from the connection weights between the neuron. (D) Generally speaking, one of the difficulties encountered by most of the optimization methods is in finding global minimum. (E) Compared to batch learning, the sequential learning method can more accurately estimate the gradient of the error function.

4. The sequence below shows the page number used from left to right. How many times of page fault will occur if the Least Recently Used (LRU) page replacement policy is used and the frame size is 5?  1 3 6 7 2 4 6 2 5 6 9 1 2 1 4 3 4 5 7 8 (A) 5 times (B) 6 times (C) 7 times (D) 8 times (E) 9 times

5. Which of the following statement is INCORRECT? (A) The basic objective of auto-encoder is to reproduce its inputs by using a smaller set of features. (B) In generative adversarial network (GAN), the discriminator often has access to both the synthetic and real sample. Its task is to try to differentiate them as accurate as possible. (C) Training recurrent neural networks (RNNs) is much easier than training convolution neural networks (CNNs). (D) In GAN, the generator has to provide information to the discriminator so that the discriminator can try to do a better job for classifying real and fake samples. (E) For CNNs, the convolution is essentially a filtering operation.

6. Suppose a hashed file is constructed using the division hash algorithm of “key-value mod size”. Assume that indices of the hash table are labeled 0⋯6. The sequence of inserting the following key-values into the hash table is 24, 30, 3, 18, 15, 21, 9. To read the keys out from the table following the order of 0⋯6 of the hash table's indices, which of the following might be the possible sequence? (A) 24, 30, 3, 18, 15, 21, 9 (B) 3, 9, 15, 18, 21, 24, 30 (C) 21, 15, 9, 30, 3, 24, 18 (D) 3, 9, 18, 21, 30, 24, 15 (E) 21, 15, 30, 9, 24, 3, 18

7. (A) 228 (B) 289 (C) 258 (D) 226 (E) 232

8. The technique that allows a computer with small RAM to execute a program of much bigger size is _______. (A) paging (B) virtual memory (C) multi-disk (D) spooling (E) none of the above

9. Numerical system conversion. Assume (654)r=(333)10 , please find the radix r. (A) r=5 (B) r=6 (C) r=7 (D) r=8 (E) r=9

10. Which of the following 8-bit pairs, 2’s-complement numbers will result in “overflow” when the numbers of the pairs are added? (A) 00000001, 11111111 (B) 00000001, 10000000 (C) 10000001, 10101010 (D) 00111111, 00111111 (E) 00010001, 10001000

114年 - 114 學士後醫學系招生考試試題：計算機概論與程式設計#137105

113年 - 113 學士後醫學系招生考試試題：計算機概論與程式設計#119574

112年 - 112 學士後醫學系招生考試試題：計算機概論與程式設計#137106

111年 - 111 學士後醫學系招生考試試題：計算機概論與程式設計#137107

111年 - 111 國立中山大學_學士後醫學系招生考試：計算機概論與程式設計#113521

111年 - 111 學士後醫學系招生考試：計算機概論與程式設計#107630

110年 - 110 學士後_醫學系︰計算機概論與程式設計#100246

109年 - 109 高雄醫學大學_學士後醫學系招生考試：計算機概論與程式設計#86938