0% found this document useful (0 votes)
40 views16 pages

Daa Unit 3

3design analysis of algo notes

Uploaded by

vinodnag128
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PDF or read online on Scribd
0% found this document useful (0 votes)
40 views16 pages

Daa Unit 3

3design analysis of algo notes

Uploaded by

vinodnag128
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PDF or read online on Scribd
You are on page 1/ 16
= Qytt-T BAstc TRAVERSAL AND SEARCH TECHNIQUES. —_=—_-— —- Techniques for Binawy Tees — 2 a Binary +xee -tvaversal + there ave rary Operations that’ we want +0 perform on nas trees (Node in the tree Teavessal techniques: There are 3 -tupes ag tvaversal techniques ‘in binany tree hey aye it " Hraversing a tree oy vistting each exactly once” I Tordes traversal (Left Root, aight? R preorder tvaversal (Root , Abft, xighd d-postosder tyaversal: (1e¢t vvights 000) Algovithm streenade = vetord: “Rime complenity? 0C5D 4 type data: tveenode x dchiid : “Even ode « vchid > & > ‘Aigowithm Tn0vdey (4) Mt isa binary tree ig 440 then Inosdev (4 > hekind)5 7 visit, . “Trower. (4 > schind, ye > Algofith Preorder Ris 440 then t Visit 05, Pre Ordes (4 > Lehind)s 4 Pie ordey (4 sscniid), ee rn RZ “ts algostthin Post Oxder(e) “s Y \ it + #0 then a alt ( eae ony “e t postOsdey (> leis; a postovdey Gesntds | MELO t i saint) Example FoR tnovder (boot) lefty O04 rg : stepli- visit the Left node first Steps on the lept Sulbree of each node, apply inorder vue Skepai- vidit the oot node SHU ON the signe subtree af each node , apply Inordes ate - Path= D-B-f-E-H-G@-T-8-c Example for pre owes (root, ieee ged A Root Stepl- Viclt the left Node first Seepar on the left Subtsee of each node apply pee owes wale Sip ON the Vom subtyce af each node apply the preorder sute A-B- D-€-F-G—H-T-¢ Example for post oxder Cegg. réigyrt 600%) Steply vist the leet nme -fiyst step on the lest subtree of each node, apply postordes vule- Sdepar On the Fight subtree ef each node, APpely post ovde vule | Skepyr visit the yoot Node Geoph | ravers¢ sal tobehques: - “the proces, is cated D>we have OF -traverting OM the nos ox verties on a gyaph aph traversal’ sto -Lraverat| BES — Breadth first search PES depth Pixs search * Breadth 645 teclhriques on graghs t_ search Tt ts one of the — cimpkst algorithms fos ae) os vetting each veatex Wn a graph: Tn this method cath rele on the Same level Is checked before the seach prieeds +m the rest level- BES makes use of A queue Aw stove visited See enero the PAH Lyon ale easiest vietted vertices Algowitinn Algorithm Bre (v) Ut=ve visited vd 215 Yepeat for all veekces w adjacent frm 4 do © te Wtehted fed then Add wtvas [uw Xs unexplosed visted [wl i=ty y 4 if aq ts empty then seturns I[mouexplored vente>- Delete the next element Us fw 44 N get First unexplored Vener oni Cale) + xyes ~derth fist seaoclh . bene anc ne. aa cane [the pre explore each possible path to 4s conclusion [path Is +¥ed In othes words as a fay oS you ean Cif you dort have a node -to visit), Otheswise, go back and try anothey wotys Straply Tt can le called ay & hack-tyacking” jah os 2) vl a el) sy gees Alggtih? me act! aie 1g Ey xa pnvore Ms cl) Q ce eo (ws Xara ysl 4 fe swe BEE co 2 phy opting pt, visitieg ane output lp=0 ee So ptbervaing, the eaing + and these adjacent nodes 08 | Quine : [9 the role that te ‘nsnited fst ‘into he widted hows then the lp ail be- lp: 0) ~Sthe adjacent nodes OF" ane shred tp she queue « | hould not be visted ia a Ques Tv]a}b igs on the above ker gee js. steed hon a 5 done on ane wsited ned @ sin ed ‘i queue queue should be [3 rm) node ts 3" ghaald be vised ofp! 013 a nodes gf's) ate OLAG «ag and H ade Hot MisHed $+ oo (SRERT] Wee 'E wh be vetled Olps 013 6 Padinert nodes one 434.5 BEE | en "owt be Ved Adjacent nodes vig st lp: 0 13b2 queue : OP tol sb 24 g ™ theve ane no adjacent nodes that ane not visited Taken graph tm toxm of tree Ae last 5’ th be Visited o- 3 ' — 03,6 2 - 0,124 6—ha,ys Raa u8 O-saB | DES example | : | ina By applying DFS 6n +he +taken graph [Dvistting “6 node then +he output whl be 0/p The adjacent nodes for | added to the 10 Now, node 1 uit be © ane 143.6, = stack F MOW, AtHer Visiting | oP rod > Adjacent NOdes py | Oe 3,6. one of the adjacent Nodes should be taken dek us take ‘6’ and now 6% > pttey should be ex plore explosing . the Me as wow Stack 7 , t > PFter a is Viddted, node “wit be exploved as acljacent Nodes ef Q are Of soles > Now let “3 + ty Ke exploied but 3 adjacehe nodes , so adjacent nodes fy y | e' wit be added 45 the Aes 3.41 Bg does not ha back +0 a and anothey taken anti wt comtaing BD 4 wetuyns Ra is ve any adjacent node of adjacent Nnocks = oe NS ad Connected Components *- ¥ &/ —— & ot “this Te an application of PF ‘ g Used for network ~ establishment mal \ fh | | ’ 1 | Super graph DA subgraph ts a super graph and no additional edge to any verter of the Super graph ts known as connected component: connected | | i L SUper qraph Peres ee oe ea corrected components in the above Superga = efter traveling Goi) vidting te aith the vertices In the Super guaph, -the murnbey af Connected Components obtained! 4s the vesutt Approach: Before vieiting the Vevtices wil be Bere ('o’)- | wow, one vertex should be visited at frst. then that vertex will be 1" AS Adjacent vestices of | ane ah 34 we Visit tw! then tt becomes I’, Again the adjacent vertices o4 Uo ane land s bub | fs aheady visited and 3 shoud be visited and ‘it becomes *V’ : 7 Adjacent of 3 ave g anda, as 4 fs alveady visited! , a should be visited then vertex >’ becomes ‘I’. — PS all the vertices ane visited ‘they donot have additional edlges - > the same procedure takeS place Until ay yhe nodes ane \ 4 didted im the given 5 . rt epengey Algosttbin for connected components: See eee Allgovithm DFS(v) ¢ int Count =o; Veta 1 Ifn 15 no-ef Todes In super graph For lint vel) vee ni vad) £ Flag (vi=05 12 (visited Cv)s 20) then Drs (wi flag); count ++) Drs Cugiagny? Huts a adjacent vertex Flag [uJ13 for each verter uw adjacent from v do if(flag [uj==0> Des (wiFlagd» rrekuan count) 4s Contains ~ Liag! flag tg used to check whether he elempn any value (ox) not Bi-connected components:- Sr fs also applicahon of DFS | > an Wis section, ux study two concepts | Avieatation point Aptecomected components tqragh) Dpvfewlation pointy det = (vier be a connected undiverted graph - |rnen an covketilation foint of quagh |G! ts aventex usho onticalotin Pott of. graph “AAs a vewter whose removal disconneds the ie graph G's Te 4s also Known as "cut point, | i) Th a “4 point In the graph to oe node qe destwoyed (or) {oma dame ocurs then tt doesnt share any ‘nfosination « | Ms pdine ts Known 05 Avticulation point 1D Bitonmected graph A graph 'G fs. said in be bi-camected | # He Conteins no amkialation Point Approach 0) eppy DFS traversing (discovery me of a node) (Dattey visiting 1 dhe discovery stme oe 1 at be | ancl 4 ae | Pdjacent vertices of lane agra TF a 1S visited then | diseovag Hime Of a node wil be a. this wil) be continu Funk att nodes are Visted and the nodes will be giver | [Value of they Af scovery Ame of tasele- (la sy 5 6 # B Fw ND S Cin 2 g a to 3 & 4H + 5) Discovery dime - i | The graph wit be: Back edges CBmoten edges? Vo 3 4 s 64 aos wb least cost ! & { 2d Bed 2 ye Wy S Lute ( ce 4 ) Exe pwcedure ee ee ee : “as jek > * ®@ 3 6 & 5 a 19 —> Discovery tyne | or ea 2 FF > 2 2 F 4 least cost nodes c wv + L chia Pareee i, rode Aerud] > aps Cpanent] 8 S513 desqy Backtracking t- General meth eHculet on Point Mate d¢se) dea ae2v 3s dese) Aw 2 df) R pax >a ATBJ2 dF) A(m eases) oie apa ey ILO] > despey ACiod > fsa} osx ye4e J ov algovithm | 2 DEN the backtracking method» Barktracking “6 a metho Lochen a pablem “having — murtl ple Sowtions to -find atl the Solutions we reed to Use backtracking » Dan the ack-toackt ne Uple (Arar hare” the Sows as Some finite set oi- an L> some of 4the pwvblens chose backtvacting ane D N- queens peoblems RD sum of Subsets > Geaph coloring *) Hamntitonian cycle | © O71 kemapsiek prob lem OA critesion function ¢ (arta -” tnetnods the desived — sotutlon ean) whee the 4 on Mayra’ in) soiutsons fs expressable are chose -fon 35 (or wrinimniyes” oF socishes: ane best done using Applications of Backtracking principle of backtyack back-tyacking' «then there are possible for sottstying 3 would -form ai) these \ and save “those whith vied has the abitity +0 yield mm trials approach has to Satisfy LSsuppose mi 1S eran ne tuples force approach one with P» Me Mss. Sunckion p- The brute n- tuple — evaluate each the optimum: the backtrack algooithin the same anaes uftth fo fewes thar > the problem solving using pack-backiing Sek Of constraints a complet D > the constraints ave dassified Into two categon"s explctt x Tompticit ticit 2 Lyexpticit censtyalnts are the ules that restrict the possi ble Value of each xi, and depend on the pasticulas instance 1 of the pooblem. All tuples that satisfy the explicit consteaints define a possible Solution space for T- Examples are att non negative wreat numbers) Ex: Graph coloving s sum of subsets nir=0 wr Siz ( gico (oS WO Sie pony €x! Jjeeni 4=Ur OF Sie Satdic=a 2 ut) Exr 8- Queen psoblem 5 Emplick constraints ove the vules that cletermine which of dhe tuples In the Solution — Spare of 1 salisty the cxitexion Chounding fanckon) heorratie STEPS - 1. Build pula the Gnitial node a8 the active node Ce-node) a. with the. £-node , track the net node ustth Depth Fivst seanc method 3. cheex constraint , 4¢ the concivaint 15 satisfied then state nodes os attve nodes Ce-node) ne 2e constvoine WS not sassfied, ckclove tre node as a clead note Co-node> 5: TF 4the obtat obtained d-node, batktratk -to the -£-npde on tty \ “hen atk 49 another branch: an unt he disired — sdution “6 obtained | and $0 “-flagodsthen fos Backtracking + Recussive mn BoekAacking C4) f for Ceath a(k1€ TEAL | lgowih deal, eat EMS i 1s (BEGUD Hal sr ALK) 0) then 4 Ve (ata, “eafra ts a path to on answer node) then write (a(l* KD i¢ (Kem) then Backtvactinfk D5 | 5 i; 4 | g-Queens_prdbolems- iis 45 peoblero based on chess game with ants peolem is a queens on the ches® board in such stated nat awsang® the a way that no +WO queens can ditack each other. Here attack means no £090 queens can be placed tn the same yw same same dfagonat- when we plate the queens k the dfagonat confltcts by | couumn os on the fn the chess board we need 0 che | using the follovaing fosmula+ rape bed wv tlekd the above conditfon can also be checked as linil= |e-a] | Example: Fiest consider 4x4 chess board tp place + queens - | sowuk{or’ pw , ye start whith empty chess board» place Queen 4 on ah 2 first possible location ie 1 on the -fivst mw & -figgt Coun - - - a 2 uy of stepar then place Qa after 4ayfing Some possiite loraiions of Ga Stepae ae \ le L an nis tg the dead end, pecarse thivd queen aa ae 7 he next column as there fs M0 acceptable position for 03, | Hence apply BT algovithm and place the second queen at Cardi) location Beco ee nest araeen Mat) 21-0) Doe noa a ae aad once dead end aS. next quem 'u? cannot be placed ott pexmissible loccthon: “Here we need to back 4ratk atl the way to @ and move 7t to (uad and place ‘oa at Car). queen's’ at (3,0 and queen 'y’ at ad owt s corwrrme | “The above one is the Sowtion for 4 -queen problem. | 2 Ba as oy | Pt 1 4 2 a 3 | ata xe + Byax2 + UG xBKM) = BG > [ie in & (| 64) g GY BR AVA 1 \%X \* @ \N bd hs 86 oH O | ?@ | al W Pal * 5 © OOO OF Fee owanryation of aH ates. k ve of the u-quee i | "| : ns Solution spate. No | numbered as 1D Clepth Fixst search , “ Ls a queens pobler also Same os above queens problem: Guin 0 subsets gsvblem’- ns numbers called walq nts and we have © we Ore geen Ip poctive m. this find all combinations of these numbers who sum 3 is called Sur lem - Example for 5ay $3, $4 950561 $5) 10 Lad 1 197808 m= 2 $6 os 8h= . . ; 5, 10118 = 20 [ ile golubions | Gap toh = 20 constraint The sum af the ahaa gelected solutions shoutd not euceed “the value of ‘at oy not fnetuced - 2 TH this problem: he solutions MAY (om) ™ Lime. complerity :2™ | enple:- | From example’ fi $a/84,5H-85:564 oF Bio tay B15) A038 12 3 4 ® , 1 0? | subset = $1a5U- ord sowwtion!— I= fio,aoy. forsardarstsss.Sey D {5,lrlar 115 .a0% oe poe ee ee ee | i Geragh _colouving Graph coltusting is Q pwbler of. cauing each vertex t a qrath fn a such a way thet no two adjacent vertices have Same colour and ™ colours ave used 40 louy the vewtices- Therefore this poblem fs also called as M™ colouring problem - constiaint No two adjacent vertices and nelq greasing ventiees chould not be coloured with same colour. | Fkample— Hence we vequhe 3 coloUxs tp colous+the graph, then the chapmatic number of given graph (3 8- We Use: back -racking technique +0 Solve -this prblem - consider the graph: Dgraph gq constels of vertices nn) aa) Rae D there ove 3 colours used reds green and Wues te cuit numiey them out as j—veds a-green » 3—blue te me Cannge astgn the colour cithey 4 oy 2 ors for c’ becuse the graph catouring papblem constraint #8 that ho Ewe adjacent vestees have the Same colour y? IF we assign colour 4 to © then adjacent vertex ofc tbe A is having colour: 4. sinflanly g 7s having 2 and having 3: | therefore we reed 0 bacetrack1Ng \ | oO | -ime complevity - U (c") Ly 6 Tete no: % COONS using ™ Hamiltonian cyele- consider any starting vertex & travelling froma starting Vortex visiting evey node atleast once and ot last tHavelling back to the ‘it forme a cycle and this cqele formed #8 Known as Hamiltonian dle startin vertex, then Lo) -Hamtitonian path (of) -HamiHonton graph: |? Back-rocking method is wed In his problem +o find ou the mut ple Solutions | posstble solutions: ni NN: of vertices | €xample > | \ a . yam above example , n=b hence there will be 6! ie. 420 possible Sowtions Time complexity - o(n)) where nts no: of veatices.

You might also like