MarkAllenWeiss數(shù)據(jù)結(jié)構(gòu)與算法分析課后習(xí)題答案8

1、chapter 8: the disjoint set adt8.1 we assume that unions operated on the roots of the trees containing the arguments. also, in case of tics, the second tree is made a child of the first. arbitrary union and union by height give the same answer (shown as the first tree) for this problem. union by siz

2、e gives the second tree.8.2 in both cases, have nodes 16 and 17 point directly io the root.8.4 claim: a tree of height h has at least 2 nodes. the pn>of is by induction. a tree of height 0 clearly has al least i node, and a tree of height i dearly has al least 2. let t be the tree of height h wit

3、h fewest nodes. thus at the time of 廠 s last union, it must have been a tree of height h i, since otherwise t would have been smaller at that time than it is now and still would have been of height h，which is impossible by assumption of t minimality. since 7s height was updated, it must have been as

4、 a result of a union with another tree of height "-l by (he induction hypothesis, we know that at the time of the union, t had at least 一憲 nodes, as did the tree attached to il, for a total of 2" nodes, proving the claim. thus an 況-node tree has depth at most log jvj.8.5 all answers are 0(

5、af) because in all cases a(m.n) = i.8.6 assuming that the graph has only nine vertices, then (he union/find tree that is formed is shown here. the edge (4.6) does not result in a union because at the time it is examined. 4 and 6 arc already in the same component. the connected components arc 1.2,3.4

6、,6| and8.8 (a) when we perfonn a union, we push onio a slack the two roots and (he old values of their parents. to implement a deuiiion. wc only have to pop the slack and restore the values. tliis strategy works fine in the absence of path compression.(b) if path compression is implemented, the stra

7、tegy described in pari (a) does not work because path compression moves elements out of subtrees. for instance, the sequence union(1.2). union(3.4). union! j.3). find(4). dcunionu.3) will leave 4 in set 1 if path compression is implemented.8.9 we assume that the tree is implemented with pointers ins

8、tead of a simple array. thus find will return a pointer instead of an actual set name. we will keep an array (o map set numbers to their tree nodes. union and find are implemented in the standard manner. to perform remuvetxk first perform a find(x) with path compression. then mark the node containin

9、g x as vacant. create a new one-node tree with x and have it pointed to by the appropriate array entry. the time to perform a remove is the same as the time to perform a find, except that there potentially could be a large number of vacant nodes. to take care of this, after n remove s are performed,

10、 perform a find on every node, with path compression. if a find(x) returns a vacant root, then place x in the root node, and make the old node containing x vacant. the results of exercise 8.11 guarantee that this will take linear time, which can be charged to tlie n remove s. at this point，all vacan

11、t nodes (indeed all nonroot nodes) are children of a root, and vacant nodes can be disposed (if an array of pointers to them has been kept). this also guarantees that (here are never more than in nodes in the forest and preserves the ma(m. n) asymptotic time bound.s. 11 suppose there are u union s a

12、nd f finds. each union costs constant time, for a total of ". a find costs one unit per vertex visited. we charge, as in the text, under the following slightly modified rules:(a) the vertex is a root or child of the root(b) otherwiseessentially，all vertices arc in one rank group. during any fin

13、d, there can be at most two rule (a) charges, for a total of 2f. each vertex can be charged at most once under rule (b) because after path compression it will be a child of the root. the number of vertices that are not roots or children of roots is clearly bounded by independent of the unioning stra

14、tegy, because each union changes exactly one vertex from root to nonroot status, and this bounds the number of type (b) nodes. thus the total rule (b> charges are at most u. adding all charges gives a bound of2/ + 2u，which is linear in the number of operations.8.13 for each vertex r. let the pseu

15、dorank r、be defined as |jog s、j. where sv is the number of descendents (including itself) of v in lhe final tree, after all union s are performed, ignoring-43-path compression.although the pseudorank is not maintained by the algorithm, it is not hard to show that the pscudorank satisfies the same pr

16、operties as the ranks do in union-by-rank. clearly, a vertex with pseudorank r、has at least 2 descendents (by its definition，and the number of vertices of pseudorank r is at most n/2r. tlie union-by-size rule ensures that the parent of a node has twice as many descendents as the node, so the pseudor

17、anks monotonically increase on the path toward the root if there is no path compression. the argument in lemma 8.3 tells us that path compression docs not destroy this property.if we partition the vertices by pseudoranks and assign the charges in the same manner as in the text proof for union-by-ran

18、k. the same steps follow, and the identical bound is obtained.8.14 this is most conveniently implemented without recursion and is faster because, even if full path compression is implemented nonrccursivcly. it requires kvo passes up the tree. this requires only one. wc leave the coding to the reader since comparing (he various union and find strategies is a reasonable programming project. tiic worst-case running time remains (he same because lhe properties of the ranks are unchanged.