Сложно Решено: 0 Процент решений: 0%

Ranked Pairs Winner

Find the ranked pairs winner of an election by locking head-to-head victories that do not make a cycle.

Чему вы научитесь

  • Build a head-to-head preference table from ranked ballots
  • Sort victories by margin with exact tie rules
  • Lock edges only when they keep the graph acyclic

Условие

Показать перевод

Ranked pairs picks a winner by looking at every head-to-head matchup. Every voter ranks all the candidates, and candidate a beats candidate b when more voters rank a above b than b above a. The margin of that victory is (voters preferring a) − (voters preferring b). Matchups where both sides have the same count are left out.

  1. Sort the victories by margin, largest first. Equal margins are ordered by the winner's number, smaller first, and then by the loser's number, smaller first.
  2. Go through the victories in that order and lock each one in as an arrow from winner to loser, unless that arrow would close a cycle with arrows already locked (that is, unless the loser can already reach the winner by following locked arrows). Such a victory is skipped.
  3. The winners are the candidates that no locked arrow points to.

Implement ranked_pairs_winner(candidates, ballots).

Input

candidates: The candidates' names. Candidate i is candidates[i].

ballots: One row per voter: the candidate numbers (0-based) from most preferred to least preferred. Each row lists every candidate exactly once.

Output

Return the names of all candidates with no locked arrow pointing to them, in the order they appear in candidates. Usually this is a single name; it can be several when some matchups were tied, and with no ballots it is every candidate.

Ограничения

1 ≤ length of candidates ≤ 9

0 ≤ number of ballots ≤ 200

Примеры

Пример 1
Ввод:
candidates = ["Ann", "Bob", "Cal"]
ballots = [
    [0, 1, 2],
    [0, 1, 2],
    [0, 1, 2],
    [1, 2, 0],
    [1, 2, 0],
    [2, 0, 1],
    [2, 0, 1],
    [2, 0, 1],
    [2, 0, 1]
]

Ожидаемый вывод:
expected = ["Cal"]
Ввод:
std::vector<std::string> candidates = {"Ann", "Bob", "Cal"};
std::vector<std::vector<long long>> ballots = {
    {0, 1, 2},
    {0, 1, 2},
    {0, 1, 2},
    {1, 2, 0},
    {1, 2, 0},
    {2, 0, 1},
    {2, 0, 1},
    {2, 0, 1},
    {2, 0, 1}
};

Ожидаемый вывод:
std::vector<std::string> expected = {"Cal"};
Ввод:
string[] candidates = new string[] {"Ann", "Bob", "Cal"};
long[][] ballots = new long[][] {
    new long[] {0, 1, 2},
    new long[] {0, 1, 2},
    new long[] {0, 1, 2},
    new long[] {1, 2, 0},
    new long[] {1, 2, 0},
    new long[] {2, 0, 1},
    new long[] {2, 0, 1},
    new long[] {2, 0, 1},
    new long[] {2, 0, 1}
};

Ожидаемый вывод:
string[] expected = new string[] {"Cal"};

Пояснение

Cal beats Ann 6 to 3 (margin 3), Ann beats Bob 7 to 2 (margin 5) and Bob beats Cal 5 to 4 (margin 1). Lock Ann over Bob, then Cal over Ann; Bob over Cal would close a cycle, so it is skipped. Nothing points to Cal: ["Cal"].

Пример 2
Ввод:
candidates = ["Ann", "Bob"]
ballots = [
    [0, 1],
    [1, 0]
]

Ожидаемый вывод:
expected = ["Ann", "Bob"]
Ввод:
std::vector<std::string> candidates = {"Ann", "Bob"};
std::vector<std::vector<long long>> ballots = {
    {0, 1},
    {1, 0}
};

Ожидаемый вывод:
std::vector<std::string> expected = {"Ann", "Bob"};
Ввод:
string[] candidates = new string[] {"Ann", "Bob"};
long[][] ballots = new long[][] {
    new long[] {0, 1},
    new long[] {1, 0}
};

Ожидаемый вывод:
string[] expected = new string[] {"Ann", "Bob"};

Пояснение

Ann and Bob are tied 1 to 1, so no arrow is locked and both are returned: ["Ann", "Bob"].

Как подступиться

Сначала попробуйте сами, затем открывайте подсказки по одной.

Подсказка 1

For every ballot, each candidate is preferred over everyone ranked after them.

Подсказка 2

Before locking winner to loser, search from the loser along locked arrows; if you reach the winner, skip it.

Подсказка 3

At the end, a winner is a candidate with no incoming locked arrow.

Ожидаемая сложность

O(b × c^2 + c^4) time, O(c^2) extra space

Ссылки

Это задание адаптировано для CodeUtility. Следуйте условию и ограничениям выше.

Лучшее принятое решение каждого участника. Более быстрые, экономные и высоко оценённые решения стоят выше.

Принятых решений пока нет. Решите задачу и займите первое место.

Учитесь у других

Решения сообщества

Пока нет опубликованных решений. Решите задание, чтобы поделиться первым.