東京大学 新領域創成科学研究科 メディカル情報生命専攻 2017年8月実施 問題9

Author

zephyr

Description

A binary heap tree is a binary tree with integer labels such that the value of a parent node is no less than the value of any child node. We denote the height of a tree by .

(1) Choose all binary heap trees from the following.

(2) Suppose we have a binary tree of height 2. Show an algorithm that makes a binary heap tree by exchanging the labels of a node and its child at most once.

(3) Suppose we have two binary heap trees and . A tree has a root node concatenated with and ; the root nodes of and become the two children of . Show an algorithm that creates a binary heap tree from a tree in worst-case time complexity .

(4) Suppose we have a binary tree . Show an algorithm that makes a binary heap tree by iteratively exchanging the labels of a node and its child in worst-case time complexity .

(5) Show an algorithm that sorts in descending order an array of integers using the operations you showed in (4). Show the worst-case time complexity of the algorithm with regard to the number of elements in the array.

---

二叉堆树是一种二叉树，其整数标签的值满足父节点的值不小于任何子节点的值。我们用 表示树 的高度。

(1) 从以下选项中选择所有二叉堆树。

(2) 假设我们有一个高度为 2 的二叉树 。展示一个算法，通过至多一次交换一个节点和其子节点的标签使 成为一个二叉堆树。

(3) 假设我们有两个二叉堆树 和 。一个树 有一个根节点 ，连接了 和 ； 和 的根节点成为 的两个子节点。展示一个算法，使得从树 创建一个二叉堆树 的最坏时间复杂度为 。

(4) 假设我们有一个二叉树 。展示一个算法，通过迭代交换一个节点和其子节点的标签使 成为一个二叉堆树，其最坏时间复杂度为 。

(5) 展示一个算法，使用你在 (4) 中展示的操作按降序排列整数数组。展示该算法相对于数组中元素数量的最坏时间复杂度。

Kai

(1)

Choose all binary heap trees from the following.

A binary heap tree is a binary tree where the value of a parent node is no less than the value of any child node.

Let's analyze each tree:

• Tree (A) has only one node, so it is a binary heap tree.
• Tree (B) has a root node with value 5 and one child node also with value 5. This satisfies the binary heap property.
• Tree (C) has a root node with value 3, and two child nodes with values 1 and 4. Since , this does not satisfy the binary heap property.
• Tree (D) has a root node with value 9, child nodes with values 1 and 5. The node with value 5 has children with values 3 and 2. This satisfies the binary heap property.

Therefore, the binary heap trees from the given options are (A), (B), and (D).

(2)

Algorithm to make a binary tree of height 2 a binary heap tree by exchanging the labels of a node and its child at most once:

1. Input: A binary tree of height 2.
2. Output: A binary heap tree.

Algorithm:

1. Let be the root node.
2. Let and be the children of .
3. Compare with , then compare the larger child with .
4. If is less than the larger child, exchange the labels of and the larger child.

This algorithm ensures that the tree satisfies the binary heap property by performing at most one exchange operation per child.

(3)

Algorithm to create a binary heap tree from a tree concatenated with two binary heap trees and :

1. Input: Binary heap trees and , root node .
2. Output: A binary heap tree .

Algorithm:

1. Create a new tree with root node .
2. Set the left child of as the root of and the right child of as the root of .
3. Perform a heapify operation on :
   • Compare with its children.
   • If is less than any of its children, exchange the labels of and the larger child.
   • Perform the same heapify operation on the child node that was exchanged.

The worst-case time complexity of this algorithm is because the heapify operation will at most traverse the height of the taller tree.

(4)

Algorithm to make a binary tree a binary heap tree by iteratively exchanging the labels of a node and its child:

1. Input: A binary tree .
2. Output: A binary heap tree.

Algorithm:

1. Start from the last non-leaf node and move upwards to the root node.
2. For each node, perform the "heapify" operation:
   • Compare the node with its children.
   • If the node is less than any of its children, exchange the labels of the node and the larger child.
3. Repeat the process until the root node is processed.

The worst-case time complexity of this algorithm is because each level of the tree may require exchanges proportional to the number of nodes at that level.

(5)

Algorithm to sort in descending order an array of integers using the operations shown in (4):

1. Input: An array of integers.
2. Output: A sorted array in descending order.

Algorithm:

1. Build a max-heap from the array using the algorithm from Question 4.
2. Repeatedly extract the maximum element from the heap, replacing it with the last element of the heap, and place the maximum element at the end of the array.
3. Perform heapify on the root node to restore the heap property.

Steps:

1. Construct the max-heap:
   • For each node starting from the last non-leaf node to the root, perform heapify.
2. Sort the array:
   • Swap the root of the heap with the last element of the heap.
   • Reduce the heap size by one.
   • Perform heapify on the root.
   • Repeat until the heap size is 1.

The worst-case time complexity of this sorting algorithm is , where is the number of elements in the array. This is because building the heap takes time, and each of the extractions takes time.

Knowledge

堆 二叉堆 Heapify

难点思路

问题 (4) 的算法是关键，需要理解 heapify 操作以及其时间复杂度分析。

解题技巧和信息

1. 构造二叉堆时，使用自底向上的 heapify 操作，确保每个子树都是一个堆。
2. 在堆排序中，最大元素总是堆顶，通过交换堆顶和末尾元素并缩小堆大小，可以逐步将数组排序。

重点词汇

• Binary Heap 二叉堆
• Heapify 堆化
• Tree Height 树高

参考资料

1. Introduction to Algorithms, Cormen et al., Chap. 6. Heapsort
2. Data Structures and Algorithm Analysis in C, Mark Allen Weiss, Chap. 5. Priority Queues