24.腹腔双套管灌洗引流的目的是:( )

腹腔双套管灌洗引流的目的是：( ) A. 引流作用 B. 冲洗作用 C. 减少胰液对机体的损<|endoftext|>Transitive Reduction of Binary Relations Consider a directed binary relation $R$ on the set $X$. For this relation, I would like to know what elements in $X$ are transitively related to each other. I could compute the transitive closure of $R$ and then do a search for elements which are connected to each other in the graph representation of the transitive closure. Is there a better way? In particular, can I avoid computing the entire transitive closure? pyotiny 2013-01-07: This can be done more efficiently than computing the entire transitive closure. Since you only care about whether two nodes are in the same equivalence class, you could use a union-find data structure to track this. Initially, each element is in a different equivalence class, and each element is its own parent. Then you perform a depth-first search of $R$, calling the Union function whenever you find that an element is in the same equivalence class as another. The Union function will combine the two equivalence classes into one. Let $n$ be the number of elements in $X$, and let $m$ be the number of pairs in $R$. Using the union-find data structure takes $O(n \log n + m)$ time. The time for union-find is dominated by the $O(n \log n)$ term, because the $O(m)$ term can be absorbed by the $O(n \log n)$ term by using the fact that $m = O(n^2)$. (If $m = O(n)$, then we can use a different data structure such as a hash table instead of union-find to get $O(m + n \log n)$ time.) I believe that the time can be further improved, to $O(n \alpha(n) + m)$ time, where $\alpha(n)$ is the extremely slow-growing inverse Ackermann function. Hopcroft's 1974 paper "An $n \log n$ algorithm for minimizing states in a finite automaton" gives a plan for doing this on the middle of page 262. However, even though the author promises to give more details later, I could not actually find those details later in the paper! The details can be found in Tarjan's 1983 paper "Data structures and network algorithms" on page 113, but it requires a lot of background from the rest of the paper, so it's not an easy read. I believe the details can also be found in Sleator and Tarjan's 1983 paper "A data