twork bo used to find people who have recently chy 4 pelmary e-mail adeegy hohave recently changed jow can a graph that models o-mall m. network be used to find electronic wall mating Ts sed to et the same message to many different e-mail cab hh model th: describe a graph model that represents whether each per- onata party knows the name of each other person is arty. Should the edges be directed or undirected? Should nultiple edges be allowed? Should loops be allowed? Jescribo a graph model that represents a subway system 1 a large clly. Should edges be directed or undirected? hould multiple edges be allowed? Should loops be al- owed? ‘or each course at a university, there may be one or more ther courses that are Its prerequisites: How can a graph used to model these courses and which courses are pre- equisites for which courses? Should edges be directed or indirected? Looking at the graph model, how can we find courses that do not have any prerequisites and how can we find courses that are not the prerequisite for any other urses? Describe a graph model that represents the positive rec- ommendations of movie critics, using vertices to repre- ~ [rassuauatumonts must be executed before Sg 1s executed A es roman 4n Example 8? (Use the precedence graph $8. Construct aprecedonco graph for the following program: Sixt 0 Spx 34. Describe a discrete structure based on a graph that can be used to model airline routes and thelr fight times. (Hint: Add structure toa directed graph.) 85. Describe a discrete structure based on a graph that can be used to model relationships between pairs of individuals Ina group, where each individual may either like, dislike, or be neutral about another individual, and the reverse relationship may be different. [Hint: Add structure to a directed graph. Treat separately the edges in opposite di- rections between vertices representing two individuals.) 38, Describe a graph model that‘can be used to represent all forms of electronic communication between two people {na single graph. What kind of graph is needed? i, Graph Terminology and Special Types of Graphs Introduction ee We introduce some of the basic vocabulary of graph theory in this section. We will cabulary later in this chapter when we solve many different types of problems. One sucl determining whether a graph can be drawn in the plane so that no two of its edges cross. Unis Acothee examples Miociding whether there is a one-to-one correspondence between the vertices of two graphs that produces a one-to-one correspondence between the edges ofthe graphs. We EFINITION 1 i in models. also introduce several important families of graphs often used as examples and in m Several important applications will be described where these special types of graphs arise. Basic Terminology ¢ Pee First, we give some terminology that describes the vertices and edges of undirected graphs. i » in an undirected graph G are called adjacent (or neighbors) in G if w eee a ia oF an edge ¢ of G. Such an edge ¢ is.called incident with the vertices u and v and ¢ is said to connect u and v.ITION 3 ——— FNAMPLET nbors of a vertex v of G = (V, E), d led the neigh. kr vubeot oF V, on denote by ea the set of all vertices In G that ey vertex in A. So, N(A) = Uyeg NW). To keep track of how many edges are incident to a vertex, we make the following de ted graph |s the number of edges incident with it, oc de @ rtex in an undit ibetateen a tien s twice to the degree of that vertex. The degree of that a loop at a vertex contri vertex v'Is denoted by deg(v). What aro the degrees and what are the neighborhoods of the vertices in the graphs G and 1 displayed in Figure 1? In G, deg(a) = 2, deg(b) = deg(c) = deg(f) = 4, degid) = 1, deg(e) = 3, and deg(s) = 0. The nelghborhoods of these vertices are N(a) = (b, f), N(b) =(a,c.p, f, N(e) = {b, dye, f), N(d) = {c), N(e) = (b,c, fh, NC) = {a,b, c,e}, and N(g) =. In H, deg(a) = 4, deg(b) = deg(e) = 6, deg(c) = 1, and deg(d) = 5. The neighborhoods of these vertices are (a) = (b, d, e}, N(b) = {a, b,c, d,e}, N(c) = (by; N(d) = (a, be) and N(e) = (a, b, d). 4 O&O ey FIGURE 1 The Undirected Graphs G and H. A vertex of degree zero is called isolated. It follows that an isolated vertex is not adjacent to any vertex. Vertex g in graph G in Example 1 is isolated. A vertex is pendant if and only if it has degree one, Consequently, a pendant vertex is adjacent to exactly one other vertex. Vertex d in graph G in Example 1 is pendant. Examining the degrees of vertices in a graph model can the model, as Example 2 shows. si "There is an edge between two vertices in a niche overlap graph if and only Species represented by these vertices compete. Hencé, the degree of.a vertex in a niche overlap graph is the number of species in the ecosystem that compete with the species represented by this vertex. A vertex is pendant if the species competes with exactly one other species in the? -suoppanp aacy safipa papastp raya sudesd 203 ABojoran2), jes poyoaup a sofpa yemp 1985 a4 S129! -gasBop ppo JO S021H9A Jo - tue aq yoru 93a) “ppo ave uns sy gunn to oa iy aso wa ne on ‘uo suiiay om amp Jo wins ayy ‘axouLLOy HM ba ysey atp Jo apas purey-qyst4 90) ba ysey ay Jo aps purey-ySts amp UF WHO SIG OM “14.3 Joy wana Sf (4)Bap asmesag] taps ax ype -caifap “% + (Bop CK = (Bop {= ‘a= ‘Apanoadsar (aA) = 9 ydexd payparpun we uy ‘ALeANe aafap wana Jo soafian J 195 9m 04 2A Pur 1A I9] food wat, 995% ‘aaxZap ppo Jo saaquian Jo 195 amp PUE 2 -saiSop ppo jo soquan jo soquun uano we sey yds paraasypum ty ‘ duns su 2, wiesoau, se vanyS s} yptyss Jo ouo ‘sdouenbasuoa Aue sey 12ey 2 -uana sj ydea8 paasrpum we jo seaman atp Jo soadBap oun Jo wins amp yeqy sMOYS T WOIORKLL > “0g = i ‘asoyesay,‘s08pa Jo Joqumm ay) sj m/ asay xis oa1Bop Jo yova soayuon oy ypa ude e uy azoxy axe soBpa Aurew MO} Pe € = Cp )(osaid are sdoo] pue sepa ajdprnur 3j wana soydde syy yey) 2)0N) £2) we <4 Yp “ofp CX = 2 E=(w)boP ug 989 w qa qdes8 poyauqpun ue oq (7 ‘A)= O19] WAYXOTH.L ONIMVHSONVH SELL “spuey om Surajonuy 2yeyspuey pu syuodpus omy Suyaey o8po we uaosjoq AFoqeuE ayy Jo asnesoq ‘(eurtu9| Bupjeyspuey ay} se uALoUY UdYO osye s} pie) waloat, Bupyeyspuey a pay soupeuos 5] yoni ‘| wo9q, uy ynse2oxp Omey ay “Sopa Jo zaqummU omy earn 4 Seaton om Jo s99d8op om) Jo wns oi ex) suet ss, sBapHaA ((enba Ajqissod) om) Atoexe {us wpau sepa ue asneD9qsILOA OM Jo soap a4 Jo ums M9 01 OM somnqiarG Lae > gym sajadusoo wiayssona sty) uy soysads Arana asneoo sooyuian pajejos| ou axe atau 'sa}2ads 1auj0 om) yse0 ayy une Kuo sajeduron astowr ayy asmevaq 1 amout atp "yded deyon0 ayo stp uy woyPadpoom stp pus nee des deyono von gas ou sta ie 9pm sry ty ydead Qs a}adwoo sojaads sano ype pure : a ‘oun jweptiad v Xq paquasaidar s9pads Atuo ah) 24 :soppods Jou snoy yyys sojodwo> jaunnbs axp asneane ‘woos ato "wo}sAS009 a4 ut sate ul et SPads ayadir09 yu saop saad sm 9 poreost st sojsadse 59 sudeip Jo ICL reoads pue ABojomuuoy ydeay ¢gp054 10/Graphs DEFINITION 4 DERINITION 5 a EXAMPLE 4 ‘When (x, v) is an edge of the graph G with directed edges, 1 is sald to be adjacent to vandy is said to be adjacent from u. The vertex u is called the initial vertex of (u, ¥), and vis calle, the terminal or end vertex of (u, v). The initial vertex and terminal vertex of a loop are the same, Because the edges in graphs with directed edges are ordered pair, the definitio ofa vertex can be refined to reflect the number of edges with this vertex as the i as the terminal vertex, mn of the de, initial verte ad Ina graph with directed edges the in-degree ofa vertex v, denoted by deg(v), 1s the number of edges with as thelr terminal vertex. The out-degree of v, denoted by deg (), 1s the number of edges with v as thelr inital vertex. (Note that a loop ata vertex contributes | t¢ both the in-degree and the out-degree of this vertex.) Find the in-degree and out-degree of each vertex in the graph G with directed edges shown tn Figure 2. THEOREM 3 FIGURE 2 The Directed Graph G. oO Solution, The in-degrees in G are deg~(a) = 2, deg-(b) = 2, deg-(c) = 3, degé(d) =2, deg” (e) = 3, and deg~(f) = 0. The out-degrees are deg* (a) = 4, deg*(b) = 1, deg* (c) =2, deg* (d) = 2, deg* (e) = 3, and deg*(f) = 0. < Because each edge has an initial vertex and a terminal vertex, the sum of the in-degrees and the sum of the out-degtees of all vertices in a graph with directed edges are the same. Both of these sums are the number of edges in the graph. This result is stated as Theorem 3, Let G = (V, E) bea graph with directed edges’Then » SL deg= (vy = 7 deg" (») = 12}. rev vev ‘There are many properties of a graph with directed edges that do not depend on the direction of its edges. Consequently, itis often useful to ignore these directions. The undirected graph that results from ignoring directions of edges is called the underlying undirected graph. A grap! with directed edges and its underlying undirected graph have the same number of edges. Some Special Simple Graphs We will now introduce several classes of simple graphs. These graphs are often used as examples and arise in many applications,102 Graph Terminology ant [XAMPLES Complete Graph =——— that contains exactly CoP raph o a actly one edye by mt n=1,2,3.486 Ide between « pair of distinet ver ree Plu by Types of Graphs 955 vertices, denoted b oled by Ky, Is a simple graph ‘ach palr of distinct vertices. “The oe Ke fot sim ane ph for which there Isat least o an edge Is called noncomplete. at Teast one HIGURE3 The Graphs K, for 1 3, ~~ and connect this new vertex to each of the n vertices in C,, by new edges. The wheels Ws, Wa, Ws, and We are displayed in Figure 5. a ANS FIGURES ‘The Wheels Ws, Wa Ws, and Wo. ' orn-cube, denoted by Qn, is agraphithat has vertices EXAMPLES n-Cubes aa dimension Wo ‘Vertices are adjacent If and only ifthe bit strings tar ey preant differ in exactly one bit position. We display Cr. Q2, and 3 are 6. they eres ean construct the (n+ D-cube Oust FOr the w-cube Q, by making two ote that Ye facing the labels on the Tertices with a 0 in one copy of Qn and with a 1 in the coples of On, prefacing Mees connecting two vertices that have labels differing ony in other copy of Qn and adn cr rom Oe by drawing two capes of Oss op and the frst it. In Figure 6 23 oie beginning of the label of each vertex in the bottontface and propensity Lng each vertex in the top face, Tete, by face we mean & face of Vathbapontng te lao a ning be nh te in - en drawing the projection o ath soptes of Qz as the top and boom ees ‘ofa cube and then drawing the pro} resulting depiction in the plane.) : }FF se 10 Graphs vast DEFINITION 6 IXAMPLE 9 _ UNAMPLE 10 eee EXAMPLE 1 uu to ut w u wo 4 | 101 —-+ owl | t ‘ ut ow 00 wv ° VIGURE 6 The n-eube Q,, 1 = 1, 2,3 Bipartite Graphs ‘Sometimes a graph has the property that its vertex set can be divided Into two disjoint Subsea, such that each edge connects a vertex In one of these subsets to a vertex in the othe For example, consider tho graph representiag marrlages between men and women in al where each person represented by a vertex and a mariage ls Yepresented by an ee ins graph, each edge connects a vertex in the subset of vertices Tepresenting mal the subset of vertices representing females. This loads us to Definition and a vertex A simple graph G Is called bipartite if Its vertex set V can be Partitioned into two disjoint sets Vi and Vz such that every edge in the graph counects a vertex in Vj and a Vertex in Vy (so that no edge in G connects elther two vertices in V1 of two vertices in V3). When this condition holds, we call the pair (1, Ve) a bipartition of the vertex set V of G. In Example 9 we will show that Co {s bipartite, and in Example 10 we will show that xy 5 not bipartite, Co Is bipartite, as shown in Figure 7, because its vertex set can be partitioned nto the two ses Vi = (v1. v3, v5} and Ve = (vz, v4, vp), and every edge of Cy connects a vertex In Vy anda vertex in Ve. 4 Xs is not bipartite. To verify this, note that if we divide the vertex set of A's into two disjoint Sets, one of the two sets must contain two vertices. Ifthe graph were bipartite, these two vertices could not be connected by an edge, but in ’y each vertex is connected to every ater vate an edge. Are the graphs G and 11 displayed in Figure 8 bipartite? a ’ 2 > a . )< fe ‘ fe ‘ @ e q ‘ a HIGCRE 7 Showing That Co Is FIGURES The Unalrected Graphs G and H. Bipartite.THEOREM 4 EXAMPLE 12 10.2 Graph Terminology and Spectal Types of Graphs 657 Soduuion’ Graph G As bipa « and fees woe bipartite bocause its vertex set Is the anton of (wo disjolnt sets, (a. bd) aniiles gf a} and each edge connects a vertex in one of thes subsets o 8 vstex in the other ay vate in fe 7 tobe bipartite tis not necessary thal every vertex in (a, bd) beadjacent Graph His an fg). For instance, b and g are not adjacent.) that edges do not babel ere rs pete set cannot be partitioned into two subsets so condong beet ae fromthe same subset. (The reader should verfy thls by ‘Theorem 4 provides a useful criterion for determining whether a graph ts bipartite, A simple graphs bipartite fan only iis possible to assign one of two different colors to each vertex of the graph so that no two adjacent vertices are assigned the same color. Proof: First, suppose that G = (V. £) isa bipartite simple graph. Then V = Vi U Vze where Vi and Vp are disjoint sols and every edge in E connects a vertex in Vy and a vertex In Va; If we assign one color to each vertex In Vi and a second color to each vertex In Vz, then no (wo adjacent vertices are assigned the same color. ‘Now suppose that itis possible to assign colors tothe vertices of the graph using Just ie colors so that no two adjacent vertices are assigned the same color. Let Vi be the set of vertices assigned one color and Ve be the set of vertices assigned the other color, Then, Yi and Vz are disjoint and V = Vi U Va. Furthermore, every edge connects a vertex in Y1 and a vert in Vz because no two adjacent vertices are ether both in V1 oF both in Va. Consequently, G 4 is bipartite. We illustrate how Theorem 4 can be used to determine whether a graph is bipartite in Example‘t2. Use Theorem 4 to determine whether the graphs In Example LL are bipartite, fist consider the graph G. We will ty to assign one of lwo colors, say red and blue, to each vertex in G so that no edge In G connects a red vertex and a blue vertex. Without Joss of generality we begin by arbitrarily assigning red to. Then, we must assign blue to ¢¢ 77 aud g, because each of these vertices {s adjacent to a. To avold having an edge with two bl , we must assign red to all the vertices adjacent to elther c, ¢, f, or g. This means that eo aired to both Fe aandad (and means that a must be assigned red, which It already has been}, We have now assigned colors to all vertices, ‘with a, 6, and d red and, ¢, f, and g blue. Checking all edges, we see that every edge ‘connects a red vertex and a blue vertex. Hence, by jh G Is bipartite. ee ie) to ‘sien ‘lther red or blue to each vertex in 17 so that no edge in #7 connects a red vertex and a ‘blue vertex. Without loss of generality we arbitrarily assign red to a. ‘Then, we must assign blue (ob, and J, becanse each is adjacent to a. But this is not possible Decause e and f are adjacent, so ‘both cannot be assigned blue. This argument shows that we cannot assign one of (Wo ‘colors to each of the vertices of H so that no adjacent vertices are fssigned the same.colo. I follows by Theorem 4 that #/ Is not bipartite, B Solurion: We ‘Theorem 4 is an example of a result In the part of graph theory known as graph colorings. Graph colorings isan Important part of gr: h theory with important applications. We will study graph colorings further in Section 10.8. oo iher useful criterion for determin, whether 9 graph is bipartite is based on the notion ofa path, a tople we study In Section 10.4- “A graph is bipartite if and only if tis not possible to start at a vertex and return to this vertex by traversing an odd number of distinct edges. We will make this notlon more precise ‘when we discuss paths and clreuits In graphs In: ‘Section 10.4 ena Fyorcise 63 in that section).DEFINITION 7 DEPINITION 8 EXAMPLE 18 10.2 Graph Terminology and Special Types of Graphs G03 in exactly one bit. The hypercube nese ond the aie hy rete network balances the number of direct connections for each mmunleate. Many computers h Intermediate connections required so that processors can com- algorths hare beets ave been but using a hypercube network, and many parallel resents the hypercub ied that use a hypercube network. The graph Qj», the m-cube, rep- e network with n = 2" processors. Figure 14 displays the hypercube net work for eight process aneelg) aM rocessors.(Fgure 1 displays a diferent way to draw Q than was shown In New Graphs from Old Sometimes we need only part of a graph to solve a probl a solve a problem. For instance, we may care oul ote torpatt of a large computer network that involvés the computer centers In New York, Denver, Detrol, and Atlanta Then we can ignore the other computer centers and all telephone ines not linking two of these specific four computer centers. In the graph model for the large network, we can remove the vertices corresponding to the computer centers other than the four of interest, and we can remove all edges incident with a Vertex that was removed. When edges and vertices are removed from a graph, without removing endpoints of any remalning edges, a smnaller graph is obtained. Such a graph is called a subgraph of the original graph. A subgraph of a graph G = (V, £) is a graph H = (W, F), where WC V and F CE. A subgraph H of G is a proper subgraph of Gif H # G. Given a set of vertices of a graph, we can form a subgraph of this graph with these vertices and the edges of the graph that connect them. Let G = (V, E) be asimple graph. The subgraph induced by a subset W of the vertex set V is the graph (WV, F), where the edge set F contains an edge in E If and only if both endpoints of this edge are in W. ‘The graph G shown in Figure 15 is a subgraph of Ks. I'we add the edge connecting cand eto G, we obtain the subgraph induced by W = (a, b,c, e). < REMOVING OR ADDING EDGES OF A GRAPH. Given a graph G = (V, £) and an edge svn. we can produce a subgraph of G by removing the edge e, The resulting subgraph, denoted by G — e, has the same vertex set ¥ as G. Its edge set ls E ~ e. Hence, G-e=(V,E-(e) Similarly, if isa subset of E, we can produce a subgraph of © by removing the edges in £° from the graph. The resulting subgraph has the same vertex set V as G. lis edge sets EE" a © _ FIGURE 14 A Hypercube Network for Eight Processors. FIGURE 15 A Subgraph of Ks.oot 10/Graphs DEFINITION 9 EXAMPLE 19 a i a . L a . GG, @ a 1GURE 16 (@) The Simple Graphs Gs and Gx; (b) Thelr Union Gy G, larger graph when thi ddan edge e toa graph to produce a new sis ein. We denote by G +e the new graph produced bya connecting two previously nonincident vertices, to the graph G Hence, i Gte=(V,EUle}). The vertex set of G + eis the same as the vertex set of G and the edge sets the unions, edge set of G and the set (e). EDGE CONTRACTIONS | Sometimes when we remove an edge from a graph, ‘Wwe do ag ‘want to retain the endpoints of this edge as separate vertices in the resulting subgraph. In su a case we perform an edge contraction which removes an edge e with endpoints w nd vat merges u and w into anew single vertex w, and for each edge with u or v as an endpoint epi the edge with one with w as endpoint in place of w or v and with the same second enipi Hence, the contraction of the edge e with endpoints u and v in the graph G = (V, E) produces new graph G’ = (V’, E') (which is not a subgraph of G), where V! = V — (u,v) U {w] and contains the edges in which do not have either u or v as endpoints and an edge connecting» ‘o every neighbor of either u or vin V. For example, the contraction of the edge connectnghy vertices ¢ and c in the graph Gy in Figure 16 produces a new graph G', with vertices a and w. As in Gy, there is an edge in G', connecting a and 6 and an edge connecting a and 4. There also is an edge in G} that connects b and w that replaces the edges connecting b andcal connecting bande in Gy and an edge in G’, that connects d and w replacing the edge connecting ande in Gy. REMOVING VERTICES FROM A. GRAPH When we remove a vertex wand all edges incident to it from G = (V.E), we produce a subgraph, denoted by G — v. Observe ti Gate ven Eo where Eis the set of edges of G not incident to v. Similar i ¥ is a subset of V, then the graph G — V’ is the suby — V’, E’), where £’ is edges of G not incident toa vertex fa YP enOR Ui Ves Bs Where GRAPH UNIONS ‘Two or more graphs can be combined in vari Thi ot ; in various ways. The new gr" give amna pe verlces and edges ofthese graphs is called the union ofthe graphs We¥! sive a more formal definition for the union of two simple graphs. The union of two simple graphs Gy = (V4, E; He = (M1, £1) and Gz = (V2, Ez) is the simple grap vertex set Vi U Ve and edge set £1 U Es. The union of Gi ori Gis denoted by G1 US Find the union of the graphs Gy and Gp shown in Figure 16(a).satan "Th voutex sot af the wuilon Gy Ud Gy [ 1). "The ahs no of the undone en ote Wo verte sets, namely, last evdties Alapayodt In Uyure 10), jones inochmthaiaaie a tit the number of vertices, the number of gv Jeyune of teh vouten Otho glen undlrocted i mal Heaate axl portant verte, en W et 6 {7 t 1 h ‘ . 4 Findthe sum ofthe degrees ofthe vertices of each graph Inferclses 1~Band verity that {t equals twice the number ofedyes inthe graph, 5. Canasimple graph exist wlth 15 vertices each of degreo fe (Show thatthe sum, over the set of people at a party, of ‘heunber of people a person has shaken hands with, is ‘ven Assume that no one shakes his of her own hand, tiers 7-9 determine the number of vertices and edges inte degree an out-degree of each vertex forthe Wrendeted multgraph, 1 (04 Graph Taratuotoyy and Spectah Typer of Graphs O05 As thd uilow of the (wo adye sets, ‘The unton is ———— 10, Vor each of the graphs tn Be Hudeyrees of tho vertices directly: Show that Ive 11, ileal the nero edges nh pp + Construct the underlying w wih rected edges a Pg 28? te graph 12, What doos tho degree of a vertex represent inth equaine {ncoahlp graph, whore vertices represent al the people ‘nthe wort? What does the neighborhood veroretg {raph represent? What do lsolated and pendant vertices & this gen ferent In mat ‘twas estimated that ‘vorage dogreo ofa vertex in this graph s 1000, sa isi maa nO Wi he degree of vertex represent nan academe collaboration graph? What doos the nelghburheod ats Vertex represent? What do fsolated and pendant vertices reprosent? 14, What does the degree ofa vertex in the Hollywood graph reprosent? What does the nel, lborhood of a vertex repre- ‘sont? What do the Isolated and pendant vertices represent? 10 out-degree of a vertex ina Il graph, as described 1p Example 4 of Sec- spresent? What does the degree of a vertex in tho undirected vorsion of this graph represent? 16, What do the in-degree and the out-degree ofa vertex in the Web graph, as described in Example 5 of Section 10.1, represent? 17, What do the ln-degree and the out-degree ofa vertex ina directed graph modeling a round-robin tournament rep- resent? 18 Show that in a simple graph with atleast two vertices there must be two vertices that have the same degree. 19, Use Exercise 18 to show that in a group of people, there must be two people who are friends with the same number of other people inthe group. 20, Draw these graphs. . » Ki @ Kea 3a aM Boy In Exercises 21-25 determine siete the geoph ‘sbi. fou may find it useful to apply Theorem 4 and ansi {sony dtemllng withers posble asin her red or blue to each vertex so that no two adjacent vertices are assigned the same color. 2g y esi A . ‘ e666 10/Graphs af ‘ « d Ree , : ‘ 25, » i a 20, For which values of » are these graphs bipartite? a) Ky) Cy OW, dQ 27. Suppose thal there are four employees in the computer support group of the School of Engineering of a large university. Each employee will be assigned to support ‘one of four different areas: hardware, software, network- ing, and wireless. Suppose that Ping! qualified tosupport hardware, networking, and wireless; Quigley s qualified to suppor software and networking: Rulz Is qualified to support networking and wireless, and Sitea is qualified to support hardware and software. a) Use abipartte graph to model the four employees and their qualifications. ) Use Hall's theorem to determine whether there is an assignment of employees to support areas so that each ‘employee is assigned one area to support. ©) Ifanassignment of employees to support areas so thal cach employee is assigned to one support area exis, find one. 28. Suppos” that anew company has five employees: Zamora, Agraharam, Smith, Chou, and Macintyre. Each employee will assume one of six responsibities: planning, public- ity, sales, marketing, development, and industry relations. Each employee is capable of doing one or mote ofthese jobs: Zamora could do planning, sales, marketing, or in- dustry relations; Agraharam could do planning o devel- ‘opment; Smith could do publicity, sales, or industry re- lations; Chou could do planning, sales, or industry rela- tious; and Macintyre could do planning, publicity, sales, or industry relations. a) Model the capabilities of these employees using a bi- partite graph ‘b) Find an assignment of responsibilites such that each employee is assigned one responsibility. 30, *31. #32, 33, a . Suppose that there are five young women and fy ) Is the matching of responsibilities you found in, (b)a complete matching? Is ita maximum matching? ve ‘men on an island, Each man is willing to many, somest the women onthe island and each woman swing marry any man wo swing o mary het Suppose? Sandeep is willing to marry Tina and Vandana; Banya wiling to marry Tina, Xia, and Uma: Teas vig Tia and Zelda; Anil is willing to many Van? and Zelda; and Emilio willing to mary naan eit Use Hall's theorem to show there ls no matching of ‘young men and young women onthe islandsuch young man s matched witha young woman be wig to marry. Suppose that there are five young women and six yo men on an island. Each woman is willing to mary sont ofthe men on he island and each man wllngto many any woman who Is willing (o marry him. Suppose thy ‘Anas willlg to mary Jason, Lamy, and Mal: Barbar, is willing to many Kevin and Larry: Caroli Wg tg marry Jason, Nick, and Oscar; Dlan ls willing to many Jason, Lary, Nick, and Oscar; and Elizabeth swiling i mary Jason and Matt 8) Mode the possible marlages on the Island using a ~ bipartite graph. ) Finda matching ofthe young women and the young men on the island such that each young woman is matched witha young man whom she {s wllng to many. «Is the matching you found in part (b) a complete matching? Is ita maximum matching? ‘Suppose there is an integer k such that every man on a desert island is willing to marry exactly k of the on the island and every woman on the island is marry exactly & of the men. Also, suppose that a man is willing (o marry a woman if and only if she is willis to marry him. Show that itis possible to match the men a! women on the island so that everyone Is matched with someone that they are willing to mary, In this exercise we prove a theorem of Bysteit Ore. Suppose that G = (V, E) is a bipartite graph with bipar- titlon (Vj, V2) and that A C Vj. Show that the maximum number of vertices of Vj that are the endpoints of a \, matching of G equals |¥j|—maxacvjdef(A), where def(A) = [Al ~ [(A)]. (Here, def(A) is called the de- ficiency of A.) [Hint; Form a larger graph by adding max,cy,def(A) new vertices to Vz and connect all of them to the vertices of V1.) For the graph G in Exercise 1 find a) the subgraph induced by the vertices a, b, c. and / b) the new graph G; obtained from G by contracting the ‘edge connecting b and f. Let be a postive integer. Show that a subgraph induced bya nonemplysubsel ofthe verexseof Ky isa compe graph.say veces and OW AY cediges do these graphs i aS My ak ©) On he we de eigen oe gph Cn Examp re egreesequances for cach of the graphs In fad te pes sequence of each of the following its ba om Mia Os ofthe bipartite graph K, i sequence of the bip: o sh sa er poste Integers? Explain your answer. vege equence of Ky, Where 1a posve 98 Mgr Exlan your answer. weeny edges docsa graphhavelf its degree sequence rs 3 2.27 Draw such a oy a ny edges doesa graph have ifs degree arr I? Draw such a graph. nue dv called graph Ii the degree gue of simple graph, 'E Deernin whether exch of these sequences Is graphic. Fortase tat are, draw a graph having the given degree 44 Determine whether each of these sequences is graphic. Fer those that are, draw a graph having the given degree sequence 5,3,2 5.4,3,2,1 ©) 4,4,3,2,1 wg (28553 932200 HLL 4 Soppse ta dy, dp... dy iS graphic Sequence. Show at there isa simple graph with vertices vy, v2... i) = dj fori =1,2,...,n and »y is adja- ist atasequence di, dp, ..., dy of nonnegative inte- £3 in nouincreasing or ray der is a graphic sequence if and a the sequence obtained by rade the terms the ee de Ma days — Wy days «on dp SO gq ‘betce, increasing order is a graphic "MRS are in noni Use Exercise AS to ¢ i it ig tg Uta ecursive algorithm for de- : ct is raphe Nonincteasing sequence of positive a ery non ges wi mnincreasing coches map Tht lps te af OPH, that Is, an undirected graph Hf aitng allowed. (int: Construct sacks Land ” id edign ay 1290S as possible a each vere vt aes connecting vertices of odd ¥ this construction works.) with a least one vertex does Kz 40. How many subgraphs with atleast one vertex does Ks ave’ 50. How many subgraphs with at least one vertex does Ws have? 51, Draw all subgraphs of this graph. a 6 € a ‘52. Let G be a graph with v vertices and e edges. Let M be the maximum dogree of the vertices of G, and let m be the mintmum degree of the vertices of @. Show that a) 2e/v> m. b) 2e/veM. A simple graph Is called regular if every vertex of this graph has the same degree. A regular graph {s called n-regular if every vertex In this graph has degree n. 58. For which values of n are these graphs regular? a) Kn b) Cy © Wn 4 On ‘54, For which values of m and n Is Ky, regular? ‘55, How many vertices does a regular graph of degree four with 10 edges have? In Exercises 56-58 find the union of the given pair of simple graphs. (Assume edges with the same endpoints are the same) 56, 2 co V e e d 7 ad 57.4 b> a ff 8B x ae c da © g @ 5B a ba é e af & 59, The complementary graph G of a simple graph G has the same vertices as G. Two vertices are adjacent in G if and only if they are not adjacent in G. Describe each of these graphs. _ ae a) Kn b) Kin) Ge d) On 60. IfG isa simple graph with 15 edges and G has 13 edges, how many vertices does G have?