I was wondering if we could implement Prim's algo in O(n log n) (n is number of vertices)
№ | Пользователь | Рейтинг |
---|---|---|
1 | tourist | 4009 |
2 | jiangly | 3839 |
3 | Radewoosh | 3646 |
4 | jqdai0815 | 3620 |
4 | Benq | 3620 |
6 | orzdevinwang | 3612 |
7 | Geothermal | 3569 |
8 | ecnerwala | 3494 |
9 | Um_nik | 3396 |
10 | gamegame | 3386 |
Страны | Города | Организации | Всё → |
№ | Пользователь | Вклад |
---|---|---|
1 | Um_nik | 164 |
2 | -is-this-fft- | 160 |
2 | maomao90 | 160 |
4 | atcoder_official | 158 |
4 | cry | 158 |
4 | awoo | 158 |
7 | adamant | 155 |
8 | nor | 154 |
9 | TheScrasse | 153 |
10 | Dominater069 | 152 |
I was wondering if we could implement Prim's algo in O(n log n) (n is number of vertices)
Название |
---|
I believe the the best we can get is $$$O(m\cdot\lg(n))$$$ for general graphs. I'm assuming the problem you are solving doesn't give you edges explicitly but instead defines their weight based on some vertex properties. Like $$$cost(u, v) = a_u \oplus a_v $$$(You can find this problem here). Or if a vertex is connected to a range a contiguous range of vertices. Or if the vertices are actually points on 2d-plane and we use euclidean distance.
tl;dr; No, unless graph has special properties.
You're right, iam solving problem which the vertices are points on 2d-plane. You have any link about this ?. Tkx.
If you are using euclidean distance on points on the plane, you know that the minimum spanning tree will be a subset of the delaunay triangulation, which has O(n) edges, and can be found in O(n*log(n)). So you can do this:
Of course, finding the delaunay triangulation is an enormous amount of work to actually implement...