Sponsored
Sponsored
To maximize the like-time coefficient, we can sort the satisfaction levels in non-descending order and calculate the possible like-time coefficients by considering prefixes from the end of the sorted array. This allows us to selectively choose higher satisfaction dishes to maximize the total like-time coefficient.
Time Complexity: O(n log n) due to sorting.
Space Complexity: O(1) for using a constant amount of extra space.
1#include <stdio.h>
2#include <stdlib.h>
3
4int compare(const void* a, const void* b) {
5 return (*(int*)b - *(int*)a);
6}
7
8int maxSatisfaction(int* satisfaction, int size) {
9 qsort(satisfaction, size, sizeof(int), compare);
10 int total = 0, sum = 0;
11 for (int i = 0; i < size; ++i) {
12 sum += satisfaction[i];
13 if (sum > 0) {
14 total += sum;
15 } else {
16 break;
17 }
18 }
19 return total;
20}
21
22int main() {
23 int satisfaction[] = {-1, -8, 0, 5, -9};
24 int size = sizeof(satisfaction) / sizeof(satisfaction[0]);
25 printf("%d", maxSatisfaction(satisfaction, size));
26 return 0;
27}This C program sorts the satisfaction array in decreasing order and iteratively calculates the like-time coefficient. If adding the current satisfaction level to the sum results in a non-positive value, the computation stops as it indicates that further additions would only decrease the total result.
An alternative approach involves using dynamic programming to maximize the like-time coefficient. We maintain a DP table where `dp[i]` represents the maximum like-time coefficient attainable using the first `i` dishes. This approach computes overlap coefficients dynamically using state transitions calculated for each dish considered.
Time Complexity: O(n^2)
Space Complexity: O(n)
1using System;
2
class Program {
static int MaxSatisfaction(int[] satisfaction) {
Array.Sort(satisfaction);
int n = satisfaction.Length;
int[] dp = new int[n + 1];
for (int i = n - 1; i >= 0; i--) {
for (int t = 0; t <= i; t++) {
dp[t] = Math.Max(dp[t], dp[t + 1] + satisfaction[i] * (t + 1));
}
}
return dp[0];
}
static void Main() {
int[] satisfaction = {4, 3, 2};
Console.WriteLine(MaxSatisfaction(satisfaction));
}
}The C# program adopts dynamic programming to make state transitions represent time indices for prepared dishes, updating choices dynamically to achieve the maximum like-time coefficient.