[ unite int | Wl | _ Graph Theory Q.1 Define graphs with examples. Ans. : Graphs : A graph is an ordered pair (V(G), E(G)) where i) V(G) is non empty finite set of elements known as vertices or nodes. V(G) is called the vertex set. ii)E(G) is a family of unordered pairs (not necessarily distinct) of elements of v, known as edges or arc or branches of G. E(G) is known as edge set. Graphs are so named because they can be represented diagrammatically in the plane. It is denoted by V(G, E). a) Each vertex of G is represented by a point or small circle in the plane. In practical examples vertex set may be the set of states or cities or objects etc. b) Every edge is represented by a continuous curve or straight line segment. Edges may be the route among states or cities or relation among objects etc. Diagrams of road maps, electrical circuits, chemical compounds, job scheduling family trees, all have two objects common namely vertices and edges. Let us consider the following examples of graphs with V(G) and E(G). - 1) (a 3} 2 {4 2" i | 1 VG) = (1, 2,3, 4} VG) = (62345, 6) B-)Graph Theory Discrete Mathematics ge 1, 6), : 4) E(G.) = {(, 2) (1, 5) Cs bs an a ; (2, 5), (2, 3) G4) 4 5), a 4.6.6, 9) 4 pe 3) y 8 V(G3) = {a, b, ¢, d} V(Gp) = (1, 2,3, 4, 5, 6} E(G3) = {@, b),@d) (b,c) EGy) = {(1, 2), (1, 6), (2, 3), , @), (d, a} (2,5), (2, 5), 3, 4), (4, 5), 5, 6)} 5) ¥ V(Gs)= {a, b, ¢, d} EGs)= {(@, b), (@, ©), (b, &) (, €), (b, 0), (c, ©) (¢, d) (d, d), (4, e)} ob 7 6) V(Go ) = {a, b} E (Ge) = {a, b) (a, a) (b, b) = fey, 2, €3) i) Tf x and y are two vertices of a graph G and unordered pait {% y} = (% y) = e is an edge then we say that edge ¢ joins x and y ot € is incident to both vertices x and y. | In this case, vertices x and y are said i example (1), ¢ = (2, 3) «. e is incident at one incident on e = (2, 3). ii) Two vertices x and y are said to be adjacent to each other if the paif (x, y) is an edge of G. to be incident one e.g. In 2 and 3 and vertices 2, 3 are A Guide for Engineering StudentsDiscrete Mathematics 3-3 Graph Theory oO raph Theory Ife = (% y) is an edge of G then x and y are said to be end vertices of ¢ and we can say that ¢ is incident at x and y. iii) Two edges e; and e5 are said to be adjacent if they have a common vertex i.e. Ife, and e2 are adjacent then e; = {x, y} and e = {y, z}. iv) An edge joining a vertex to itself is called a loop. E.g. In example (5) there are 2 loops (c, c) and (d, d). y)A pair of vertices of a graph is joined by two or more edges, such edges are called as multiple or Parallel edges, In example (4) (2, 5) (2, 5) are multiple edges, 3.2 : Types of Graphs Q.2 Explain different types of graphs. 1) Multigraph : A graph in which a pair of vertices is joined by two or more edges is called a multigraph or multiple graph, ie. A graph having multiple edges is called a multigraph, In examples (1), (2), 3), (6), graphs are not multigraphs and graphs in examples (4), (5) are multigraphs. 2) Pseudograph : A graph having loops but no multiple edges is called a Pseudograph. Graphs in examples (3), (5) and (6) are pseudographs. A graph having only loops is called a Haary graph. For example : pee fh A Fig. Q.2.4 Graph G is a pseudograph is well as Haary graph. Graph G, is Pseudograph but not Haary graph. Graph G3 is neither Pseudo not graphs. 3) Simple graph : A graph without loops and multiple edges is called a Simple graph. Graphs in examples (1) and (2) are simple graphs. hs in examples (3), (4), (5), (6) are not simple graphs, “Null graph : A graph G (V, E) is said to be null graph if E is a empty set. Null graph on n vertices is denoted by Ny. Geconas A Gulde for Engineering Studentsee. eOry Discrete Ma fathematics ° My Ny 5) Finite graph : A graph G (V, ) in 1 sets is called a finite graph. Otherwise infinite 8 6 Directed graph : A sr GV, B) is said 19) elements of E are an or rs of vertices. £8 E= {(a, b), (b, &) @ OF aye EG) rected is called a > Ny ne vy (x) and E (x) are finite finite graph. directed graph if the aph dered pail Here (a, c) # (c, a) ( ‘A graph which is not di Non-directed graph or graph. 7) Weighted graph : A graph G (V, E) in which some weight is assigned to every edge of G, is called weighted graph. : 5 Fig. Q.2.3 Weighted graph number of lied the indegree of v. It is denoted by Fig, 0.2.2 Directed graph 8) Degree of a vertex : a) In a directed graph G the edges ending at vertex v is cal deg G* (v) or d* (v) b) Outdegree : In a directed graph G, the number of edges beginning at vertex v is called the outdegree of v. It is denoted by deg G™ (v) of d-(v). c) The number of — a edges incident at a ES g b ¢ vertex v of a graph ° G with self loops counted twice is a 6 2 a! & 6, called the degree of the vertex v. It is a Fig. Q.2.4 denoted by d(v). A vertex of degree one is called pendent vertex. A vertex of d 0 legree zer' is called isolated vertex. An edge inci e pendent edge. ige incident at pendent vertex is called A Guide for Engineering StudentsDiscrete Mathematics 3-5 Graph Theory eg. In graph Gi, Vertices | Indegree_| Outdegree a 0 2 b 2 1 c 2 1 d 1 1 In graph G2, d@) = 2, db)=2+2=4, de)=4, de)=l dd) =3 , dh=0 .. f is an isolated vertex. e is a pendent vertex. An edge {c, e} is a pendent edge. 9) Order and size of graph : The number of vertices in a finite graph G js called the order of G. The number of edges in a finite graph G is called size of the graph. A graph of order n and size m is called (n, m) graph. ' If G is a (p, q) graph then G has p vertices and q edges. 10) Degree sequence of a graph : Let G be a graph with vertex set V = {V¥1,V2/V3-+-Vn} and d; = deg (v;) then the sequence (dy, d2,43,.--dn) in any order is called the degree sequence of G. Note: 1) Vertices of G are ordered so that degree sequence is monotonically increasing. 2) Two graphs with same degree sequence are called to be degree equivalent. e.g. d(vy) = 4, d (v2) = 3, d(w3) = 2, d(vs) = 5, d (v6) = 3), d(vq) =1 “Its degree seq. is (4, 3, 2, 1, 5, 3) By relabelling vertices we may write dey y ‘Bree sequence as (1, 2, 3, 3, 4, 5). Fig. Q.2.5Graph Th, Discrete Mathematics sao 2 E) be any graph then Q.3 Handshaking lemma : Let G (Vs 5 of G. 4 (v) = 2q where q denotes the number of edge vev Aus. : Proof: Let us argue by induction on q. Suppose 4 = 0 ie. G has no edge i.e. E is an empty set. So d(v) = 0, V ve V- “Did (v) = 2q = 0. Let G be a giaph with q > 0 edges. Choose any edge e = {u, v} of g Consider the graph G obtained from G as follows : i) The vertex set of G, is same as the vertex set of G i.e. VGi)=V@=V=p ii) The edges of G, are all edges of G except e. In other words, G is obtained from G by deleting the edge e. d(x) + dg, (w+ dg, (v) = 2-1) xeV xeuy > d(x) + dg, (u)-1+dg(v)-1 = 2q-2) xeV xe uy Dd) = 2q Hence the proof, xeV The result is so named because it impli iplies that hands, the total number of hands shaken must be involved in one handshake. ~ if several people shake even as two hands are Note : If $d (v) = Odd : 2» number then there does not exist any graph with this degree sequence, : ee A Guide for Engineering Students0 i 0 0 The agi ay = Number of edges joining v; and v; a "cy matrix of the following graph is ee.vraph Discrete Mathematics» 3-8 Grap) Th, Vy V2 V3 Vy V5 vf. 1000 |" Vs vo]1 0201 | @) = 2 f an v3]0 2010 wr vy ]0 0102 Pan Fig. Q.4.2 2. Incidence Matrix: Let G be a graph with n vertices and nm cg without self loops, The incidence matrix is denoted by X (G) or 1 (@) ang defined as X@ = xy = Uxilnem Where Lif" edge is incident on i vertex vj. = 0 otherwise, X G) is an x m matrix whose 0 rows corre m columns corres spond to the n vertices ‘and spond to m edges. The grap! are given below h and it’s incidence Matriy ep 2&3 &4 5 eg ey “fl 000017 X@="2]1 116000 Ys 10 011104 Y4]J0 001119 YVslO 100000 Sx7 Properties of Incidence Matrix 1) It contains only 0 and 1, a tow is equal to the degree of the Corresponding vertex, . 4) Two identical columns corres; spond to the parallel edges in graph. 5) A row with all 0’s tepresents an isolated vertex, 6) A row with single 1 represents a pendent vertex, A Guide for Engineering Suder®porte Materials 3-9 Graph Theory re incidence matrix of a graph with loop is give 8S follows + d: ie ds 0 ; Fig. 44 0 & e: 1 1 y 3x3 3, Adjacency Matrix of a Diagraph (Directed graph) Let G be a directed graph with n vertices and without parallel edges. The adjacency matrix is denoted by were, A) = [aglan 1 if there is an edge directed from v; to vj = 0 otherwise In network flow, adjacency matrix is also known as connection matrix or masition matrix. The adjancency matrix and diagraph are given below. V1 V2 V3 V4 V5 V6 vy,f0 10000 v2/0 01000 AM)=v,/0 10000 vs]0 00000 : _s G Fig. O48 ve ll 000 0 Oleug oo 4. Incidence Matrix of Diagraph : The incidence matrix of a diagraph ‘ith n vertices m edges and no self loops is a matrix, XG) = [xy] nvm where de enn, ;th Xj = 1 if j edge e; is incident out of i vertex vj. 1 -1 if j® edge e; is incident into of vertex vj. 0 otherwise. A Guide for Engineering Studentsa ee Gi Discrete Mathematics 3-10 FAP Theoy, A graph and its incidence matrix are given below : eS ee Yif-1 0 0 0 4 41 Soa, | Yetl 10 0 0 0 op ms X@ea=y, 0 -1 1-1 0 0 Ne: valo 0-10 0 0 ns Ys}O 0 0 4 0 ~1 Fig. Q.4.6 ° YeolO0 0 0 0 ~1 9 6x6 Note +i) Sum of elements in each 4 &8 Show that ima graph the number of vertices of odd degree ig even, ‘ ‘olumn of incidence matrix is zero, Ans. : Let G be a ( — q) graph. By handshaking lemma Daw,)= 2q vieV Now separate out Vertices of even degree and odd degree ~ Yee = Yas » dy) =2q vieV xeVv yeV evendegree odd degree Dai) = 24 = 240) = Even number vev : xeV odd degree even degree . The sum of vertices of odd degree is even, Hence the number of vertices of Q.6 Show that the maximum number of edges in a simple graph 5 ices ig BO@-1) with n vertices is ——. odd degree is even, Ans. : Let G be a graph with n vertices -. By handshaking Jemma . Y 4 = am = vev ‘Sm edges A Guldg for Engineering Srudt™®ey piserete Mathematics 3-11 Graph Theory Let x € V -.x must be adjacent to remaining (x — 1) vertices nd @yea- lL ¥xev . ‘. Equation (1) = (9-1) + (M- 1) +... n times = 2 m n(a-1)=2m n(n-1) m= 7 Hence the maximum number of edges in any simple graph with n vertices ,. n(o- 1) 2 Q7 Determine the number of edges in a graph with 6 nodes, 2 of degree 4 and 4 of degree 2. Draw to such graphs. IS [SPPU : Dec.-09] Ans. : Let G be the required graph with 6 nodes and m edges. .. By handshaking lemma Yaw = 2m veG d(v)+d(v2)+dv3)+d(va)+d(vs)+/ (vg )=2m 4444+24+2+2+2 =2m 2m = 16 m = 8 Hence 8 edges are required. Two such graphs are given below. Fig. Q.7.4 A Guide for Engineering Studentscere) aa = raph Theory D Ma tc 3-12 G iscrete Mathematics i h that 2 h with 12 nodes suc! of it possible to construct a graph ; has node fi we degree 3 and the remaining have degree we BSISPPU : Dect) Ans. : Let G be the required graph with 12 vertices. By handshaking lemma. Yaw) = 2m ve VG) G+3)+Gt4+4e4ege4eatgt44+4)=2m 6+40°= => m = 23 +. It is possible to construct such graph. & 9 Is graph exist for the degree sequence 4, 4, 3, 3, 2, 2, 1. + Now apply handshaking lemma Ye () = 2m= Even 4444343424241 = Bven 19 = Even which is impossible ". Such graph does not exist, Q.10 How many simple labelled Sraphs with n vertices are there ? TSTSPPU : May-10] Ans.: We know that a simple j graph with n verti tices has maximum possible number of edges aac) = m (say), To construct a simple graph with © edges and n vertices, can be done in ‘e | Ways. . ™C, ways where m= 2") and 9 cee Hence the total number of ways to construct such graphs js given by [F} (THE (8) =) va A Guide for Engineering Studentswhematles 3. ae Math ZB Graph Theory ast show that a simple graph of order 4 and size 7 does not exist. Let G be a simple graph with 4 vertices, n(n-1)_ 4x has at most =o “ = 6 edges, Ags. then G put given that G has 7 edges which is contradiction. «there cam not be a simple graph with 4 vertices and 7 edges, 42 Explain i) Regular graph ii) Complete graph iii) Bipartite ph iv) Complete bipartite graph. ‘gs. 1) Regular Graph : A graph G is said to be r-regular graph if every Vertex of G has degree r. j) Regular graph of degree zero is called null graph. iA regular graph of degree 3 is called cubic graph, eg. i) o—— I 5 ° o——o | Ng: 4 +regular graph 4+ regular graph | 2- regular graphs 3+ regular graph Fig. Q.12.1 ?) Complete Graph : A simple graph G in which every pair of distinct ane axe adjacent is called a complete graph. If G is @ complete graph ‘ Vertices then it is denoted by Ky- “— graph, there is an edge between every pair of distinct aad K,, every vertex is adjacent to remaining of each vertex is n — 1. Ths K n — 1 vertices so a iS a(n — 1) - regular graph. ‘A Guide for Engineering StudentsDiscrete Mashematics sol Sr The n(n-}) edges, Ky has exactly Consider the following examples : Fig. 12.2 M) Bipartite Graph: 4 graph G (v, E) is said to be bipartite Staph if KS Vertex set can be partitioned into two disjoint subsets say V) and V3 sock that vy) Uv) =v and ‘1 OV2 = and every edge of G joins a Vertex of v; to a vertex of vo Ih Bipartite graph, vertices of V1 should not be adjacent. It is free from loops. , Following Saphs are bipartite graphs “a v2" {1.2.3} ce | MS 23) | *[email protected]} Fig. 0.42.3 'V) Complete Bipartite graph : A bipartite graph ue By ey v= yaaa ‘1 OV) = 03s said to be complete Bipartite M each vertex o; is joined to ey, ve . oe 1 'S Joined to every vertex of V2 by a unique If ivy] = m, |v] = then the complete bipartite denoted by Kan seh G Wi Uvy, Bis A Guide for Engineering Sodenpiserete Mathematics 3-15 Graph Theory piseete Mahe _Grai Theory | | 3 | | | | 1 | | 1 5) \ t } p> A Kaa Kaa Fig. Q.12.4 The graph K, ,, is called as star graph. Examples ¢ | Kya ' Koo Ka Q.13 Is there exist any complete bipartite graph with 7 vertices and 14 edges ? Ans. : First find all possible bipartitions of 7. They are 6 + 1, 5 + 2, 423 We know that, if G(vj Uv>,E) is a bipartite graph then the number edges in G is equal to |v | - |v] FS A= bil: bal Here 'B = 14 But 6.1 = 6, 5.2.= 10, 4.3 = 12 Therefore the complete bipartite graphs with 7 vertices has 6 or 10 or 12 edges only. Therefore any complete bipartite graph with 7 vertices and 14 edges. Q44 Explain isomorphism of graphs with examples. US[SPPU : Dec.-12] Ans. : In real life we come across so many similar objects or figures With respect to size, shape or orientation. Similarly there are a few Concepts in graph theory which deal with the similarity of graphs wart, reas of vertices or number of edges, number of regions and so on. Ae all such similarities the most important one is an isomorphism of iphs, Gicone®ie Discrete Mathematics 3-16 Graph Theory Definition: Let G, (V,E)) and Gz (V2, Ez) be two graphs, G any Gy are said to be isomorphic graphs if i) There exists a bijective function 9: > Vp ii) There exists a bijective function y : E, — Ey such that e = (x, yy is an edge in G, iff (6 (%), 6 (y)) is an edge in Gy. The pair of functions 6,and y is called an isomorphism of G, and G4 is denoted by G, = Gy. i Suppose two graphs G, (Vj, E,) and G2 (Vy, Ep) are isomorphic eraphs, Then it is clear that D M=M | ie. G and G) must have same number of vertices, i) |E\|=|Ep| ie. Gy and Gy must have same number of edges, ¥8) G, and G) must have an equal number of vertices with the same degree, iv) Gi and G2 must have an equal number of loops. v) vi) vii) Gi and G, must have same number of pendent. Gy and G, must have same number of pendent edges, If u and v are adjacent in G are also adjacent, In general it is easier to Prove two graphs are not isomorphic by proving that any one of the above property fails, Consider the following example then the Corresponding vertices in G, Fig. 14.1 These graphs have i) The same number of vertices. ii) The same number of. edges, iii) An equal number of vertices of degree k. These conditions are necessary for two Braphs to be isomorphic but sufficient. ~ i‘ A Guide for Engineering Std"a piscrete Mathematics 3-17 Graph Theory A, h G, d (8) = 3, d (&) = 1, d @ = 1 and x and y are adjacent to vertex X pgsph G2. d@pa3dQypyal | ere is only one pendent vertex adjacent to x). Hence adjacency is not ved. Therefore G, is not isomorphic to G2 ie. G, = G). Note : Isomorphism of graphs is an equivalence relation. Examples = Q.45 Draw all isomorphic graphs on vertices 2 and 3. Ans. : i) For 3 vertices. . o——o % Yo V3 a OF @; are isomorphic graphs | gs and > are isomorphic graphs $—F_ard_ g—9_zte isornorphie graphs g Fig. Q.15.1 (a) Q.46 Draw all non isomorphic graphs on 2 and 3 and 4 vertices. Aus. : All non-isomorphic graphs on 2 vertices are Ra a isomorphic graphs on 3 Fig. Q.16.1 (a)Graph Theo -18 ny Discrete Mathematics 3-1 Al non. Bonar graphs on 4 vertices, ee = ia 6 kone Fig, 16.1 (b) 9.17 Draw all non isomorphic graphs on 5 vertices and 5 edges, Ans. : The following are non isomorphic graphs with 5 vertices and 5 edges. 7 ~ ) | “ag atria Q.18 Find whether the following cr Pairs of graphs are isomorphic or i g vy Y% = WN & ~Ans.: i) Both the gtaphs have 4 i? Vertices and 5 edges, | Both have 2 vertices of degree 3 and 2 vertices of degree 2, OBE d (y, assy, Fig. Q.18.4 V2» V3s V4} such that Cry boy, doy; $ preserves adjacency and non-adjacency of vertices, A Guide for Engineering Studerwe pert Mathematics 3-19 Graph Theory yor = E; Is bijective. Gis isomorphic to Graph G,. i) Fig. Q.18.1 (a) ‘As G has 4 edges and G> has 5 edges, G, and G, are not isomorphic graphs. ii) G, And G> are not isomorphic graphs because in G, vertices vy ad v3; of 4 degree are non adjacent while in ’ G,, the vertices x and y of degree 4 are adjacent. Fig. Q.18.1 (b) Q49 Identify whether the given graphs are isomorphic or not. _SSISPPU : Dec.-12] Ams.: In graph Gy, there are 2 vertices of degree 3. But in G, thee is only one vertex of o degree 3, So G, and G» are not isomorphic graphs, Fig. 1st ip es —- ox “Fig. Q.19.4 (a) _ =3.20 Discrete Mathematics Graph Gy has 9 edges nd Gy has 8 edges. Y He. Q.20 Show that the following graphs are isomorphic 7 © geaphs, | ‘Therefore Gy and Gy are not isomorphic graphs USP [SPPU ; Mayatg oy Fig. .204 | Ans. + All graphs Gj, Gy and Gs have 10 vertices and 15 edges, AI! these graphs are Suregular graphs, Also they preserve adjaceny Hence all graphs are isomorphic, Isomorphism is’ given by hoof 2 + g¢ 3 5 hoa sy Sj 630 T3b 8 ce 94d “Oe In the similar way, we ean show that Gy and G5 are isomorphic graphs. 2.21 Are the graphs isomorphic ? Why ? | ‘Ans. Given graphs Gy and Gy have 8 vertices and 10 edges. Fig Q2ia Both the graphs have 4 vertices of de Also the adjacency is preserved. 6: v (G1) > VG) is defined as Bree 2 and 4 vertices of deste! 8910920 33d 4058 155,856 457, ( A Guide for Engineering Sities pcre MO Graph they ois bijective. ., G, and Gp are isomorphic graphs, a2 Explain how fo obtain new graphs from old graphs with examples. KS ISPPU : May-07, Dec.-09, 12] ans. A good mathematical theory must contain sufficient models and examples. Moreover it must have objects from old ones. In this section we derive new graphs from old graphs. 1) Subgraphs : Let G (V, E) be any graph. A graph H (Vy; be subgraph of G if VC V and E, cE, We also say that G is a supergraph of H. eg. number of methods to generate new + E}) is said to Graphs H, Hy, H3 and Hy are subgraphs of G. But graphs Hs and He ae not subgraph as (4,7) € E (Hs) but (4, 7) ¢E (G) and (6 Ne E (Hg) but (6, 7) ¢ E (G). Properties : }) Each graph is a subgraph of itself. 2)A subgraph of a subgraph of a graph G is a subgraph of G. 3)A graph G - {v} is a subgraph of G which is obtained from G by Temoving the vertex v € G and also the edges which are incident at v. 4) ee (G) then G - e is a subgraph of G obtained from G by deleting the edge e, ECODEY A Guide for Engineering StudentsDiscrete Mathematics 3-22 In above example H, - {5} is and H, - (4, 5) is given by py 2) Edge Disjoint Subgraphs : Two subgraphs Hy and Hy of the graph are said to be edge disjoint subgraphs of a graph G if there is no edgy common between Hy and Hp ie, E (Hy) 0 E (Hy) = 6 It may tay common vertex. 3) Vertex Disjoint Subgraphs : Two subgraphs H, and H, of the era G axe said to be vertex disjoint subgraphs if there is no vertex commes between Hy and Hy ie. V(Hy) 9 V (Hy) = @ Note : 1) Alll vertex disjoint subgraphs are edge disjoint subgraphs. ‘4 Spanning Subgraph : Let G (V, E) be any graph, A subgraph H oft Staph G is said to be spanning subgraph if V (G) = ). Example : Let G be the following graph : 244 : 3 eo ee ¢ 8 7 6 He Fig. Q.22.4 et A Guide for Engineering S™Mathematles : Discrete Jed Graph Theory Graphs Hy, U2 Are subgraphs of G, Hy and Hy are edge disjoint subgraphs but not vertex disjoint subgraphs. Hy and Hy are vertex disjoint subgraphs as well edge disjoi Fut as edge disjoint subgraphs Hs and Mg are spanning subgraphs of Gas V (is) = V (ie) = VG). §) Factors of a Graph : Let G be any graph, A kefuctor of a graph G is defined to be a spanning subgraph of the graph with the degree of each of its vertex being K. K-factor is a K-regular graph, When G has a I-factor, say Gj, if the number of vertices are even and edges of G are point disjoint, In particular, Kq4 1 can not have a Iefuctor but Koy can have 1-factor of graph. Example 1) 4 4 s ‘4-factor of G 2iactor of G Fig, Q.22.2 xX (a De 4 3 40-———03 a 3 4s af 1=factor 2-factor 3. factor 1 Fig. Q.22.3 Complement of a Graph : Let G be a simple graph. The complement Of G is denoted by G is the graph whose vertex set is the same as the A Guide for Engineering Stutents| Graph Theory Discrete Mathematles 3-24 TOT 5 " 7 a £G and i h two vertices are adjacent if and only jp . ; they are not adjacent in G. if it is isomorphic to jt, A graph is said to be self complementary graph if it is is norp! i st complement. eg. » —) it) G is isomorphic to G. Note : “. Gis self complementary graph. 1) For any graph G, (G) = G 2) The complement of the nul il graph on n vertices is the complete graph Kn on n vertices and vic e versa. 3) Ky is self complementary graph. Examples ; Q.23 For the following Braphs, determine whether H (V', E’) is # subgraph of G where DV'= A,B,C), = (4, B), (A, BR} fi A Guide for Engineering StuderFes Discrete Mathematics 3-25 Graph Theory iy v’= (BCD) E’ = (B, ©), (B, D) 1) V' = {A B, C, D}, E'= (A, O} DS [SPPU : May-07, Dec.09] Ans.: i) H is not a subgraph of G = nd ecause F € V (H) Fig. Q.23.1 put Fe V (G), so V (HK) ¢ VG) ii) Here VC VG), E’ CE (G), so H (V, E’) is a subgraph of G. iii) Here V Cv (G), but E’ ¢ E (G) Therefore H (V, FE) is not a subgraph of G. Q.24 Draw all self complementary graphs on 5 vertices. CS [SPPU : Dec.-12] Ans. : The: following graphs are self complementary graphs on 5 vertices. Here G; = Gy and G) = G; = G, as well as Gy are self G, complementary graphs. oe Q.25 Explain operations on graphs. Ans.: We define some standard operations of graphs like intersection, union, Ringsum etc, . A) Intersection of Two Graphs: The intersection of two graphs Gi (V,,E,) and Gy (V,, E>) is a graph G (V, E) whose vertex set is =\ OV) and edge set is ESE, ME . The intersection of G, and Gy is denoted by G9 Gp. eg, CODES, A Guide for Engineering Sudentsv . | Graph Th } Discrete Mathematles 3-36 €Ory Vp = (Vis Var V3 V5) } Vm (Vqy Ve Van Vad aa Ep = (e1s¢5,&6) Ey (ey, e203 ea) Therefore G = G) AG) (vy E) where Ce 1. | Ve Vv = Wi Varah | a | Ey AE: = (i) | 8 : (S108) Fig, 0.25.2 B) Union of Two Graphs : Let G, (Vj,E1)} G2 (V2,E2) be two graphs. The union of Gy and G» is denoted by G; U G2 = G (¥, E) ang it is a graph whose vertex set is V = Vj Uy and Edge set is E=E, UE; Consider the graphs G, and Gy as shown in above example : | My te ve The union of G, and G, is given by G(v, E) | e{ lez ¥ where V= Vj UV) = (ys Vas V3 Vas V5} yas * V= Vi U Ve = {@1,€2,€3,€4s &55 &6} SS Ne * Both graphs G, and G are subgraphs ae C) The Ring Sum of, Two Graphs : The ring sum of two graphs G) (V;, Ey) and G, (Vp, Ep) is denoted by G = G, ® G) (V, E) whos vertex set is V = Vj U Vp and the edge set consists of those edges which are either in E, or in E> but not in both ie. E = (Ey UE) ~ (Ey AEy) The ring sum of above graphs G, and G; is given by G (V, E) = G, ® G { V= (Vy, V2, V3, V4 Vs} = VW UV) \ E = (Ey UE2)~ (Ey NE) a = {c2, 3, €4, €5, 6} ~ D) Sum of Two Graphs : The sum of two Se vertex disjoint graphs G, (Vj,E,) and Gp (V),E,) is denoted G,+G = G (V, B) is defined as the graph whose vertex set i8 V (G, UG») and consisting of edges which are in G, or Gp toget™ 5 ‘A Guide for Engineering Suitpve nema’ 3-27 Graph Theory ages ob ed by joining each vertex of G t ip te n 1 1 t each vertex of wi qhus Gi + G2 i8 nothing but the graph G,UG, in which each ee of G, is joined to each vertex of Gy by an edge, x eg If Gy GS yen, Gen, | $2 LNN Gi +G, G,+G, Fig. Q.25.5 Note : The sum N,, +N, of null graphs is nothing but the complete bipartite graph Kin, n- 2) Product of Two Graphs : Let G, (Vj,E)) and Gp (V;,E>) be two vertex disjoint graphs then the product of G, and G) is denoted by GxG = G (V, E) is a graph whose vertex set is V = Vx V and two edges (61, X) and (y, y2) are adjacent if x; = yy and xp is adjacent to yz in Gy or Xy = yy and x; is adjacent to y; in G). eg. If Fig. 0.25.7 "Decomposition : A graph G is said to have been decomposed into ‘S° sbaraphs H and K if H UK = G and H OK = Null graph i. each SS of G occurs either in H or in K but not in both. But vertices may in both. In this context isolated vertices are not considered. ‘A Guide for EnginesDiscrete Mathematics Fig. Q.25.8 / G) Fusion of vertices : A pair of vertices a and eee Gam said to be fused if a and b are replaced by a single n ‘ane pee Such that every edge that was incident on either a or b or Fi ident the new vertex c. The fusion of two vertices do not change the Humbe oj edges but reduced number of vertices by 1. eg. e 2 8 i GS Graph after fusion | Graph after fusion i of bande te and a Fig. Q.25.9 Fig. 0.25.10 Q.26 Paths’ and circuits with examples, 1) Path: An alternating Sequence of vertices and edges Vor e1~€2~€3— Vg —en Vy beginning are ending wit vertices in which each edge is incident with the two vertices immediately Preceding if and following it is called a path, The vertices vy and v, are called terminal vertices and vy, V3, ... are called its interior vertices. e.g. Let G be the following graph. Following are some examples of path i) Vi-e-v2-v; Fig. Q.26.4 a-1 3 37 V4~€4~V5 19 ~v3 ~e) - v5 ili) ve-es “V5 ~€19-V3 ~eg ~ v6 iv) vj ~e6-ve a ae Queene A Guide for Engineering Stipiserere Mathematles 3-29 Graph Theory Graph Theory here are $0 many paths between every distinct h G. Depending upon the nature of terminal wypes at path. : {pith in which terminal vertices are equal is called a closed path. A closed path is known as circuit. A path in which’ terminal vertices are distinct, is called an open path. pair of vertices of given l vertices, there are two in above examples, paths in (i) and (jv) are open paths and (ii) and Gi ae closed paths. 4, Simple Path : A path in a graph G is said to be a simple path if the edges do not repeat in the path. Vertices may be repeated. eg. i) ¥p—€1—V¥2—€2 —V3 —e19 —Vs is a simple path. vp #1 V2 ~€2 ~V3 C19 — V5 —€g ~Vy —€7—V@ is a simple path in which v3 is repeated. itl) v3—€3—~V4—€4-Vs—e\g-V3-eg—ve is a simple path in which v3 is repeated. ‘ ivv3 —€3 V4 —€4 — V5 —€j9 —V3 ~€3 — vq is not a simple path as an edge e3 is repeated. 2, Elementary Path : A path in a graph G is said to be elementary path if vertices do not repeat in the path. Every elementary path is a simple path, eg.i) vy—e;-V9—e) —V3 is an elementary path. ii)vy-e) ~vy —e) —V3 —eg —eg —€7 —V2 is not an elementary path. But it is simple path. 3. Simple and Elementary Circuits : A closed path is known as circuit, A simple path which is closed is called a simple circuit of graph. Ih other words, A circuit in a graph G is said to be simple circuit if all edges of a circuit are distinct. A circuit in a graph G is said to be elementary circuit if all vertices of a circuit are distinct except the terminal vertices ie. the first and last Vertices. The number of edges in any circuit (or path) is called the length of the circuit (or path). Th above graph G. 88 i) vy) ~v9 ep —vg ey —V5 C9 V2 -€1-V1 8 a cirouit with e; repeated twice and v2 is also repeated twice. Qeony ~ ‘A Guide for Engineering Students3-30 Graph Discrete Mathematics V2 —€7 ~V 6 ~ C6 ~ ii me V5 09 V2 £7 V6 ~& ii) v) -ey -v2.—€2 V3 ~ 810 “9 nes simple circuit but not elementary re as V2 is rep a 4 ili) v)-e)-v2-e7-Vo-e6~ YI S an elementary circuit, Q.27 Define connected and disconnected graphs. id to be connected graph if there exists... Ans. 3 A graph G is sai Ne. A graph which i not connected jg at between every pair of vertices. the disconnected graph. A disconnected graph consists of two or more parts called components 5 blocks if each of which is connected and there is no path between tio vertices if they belong to different components. Q.28 Explain edge and vertex connectivity. Ans. : Edge Connectivity : A set of edges of a connected graph G whose removal disconnects G is called a disconnecting set of G. A cutse; is defined as a minimal disconnecting set ie. A minimal set of edges whose removal from G gives a disconnected graph is called a cutset. If a cutset of a graph contains only one - edge, then that edge is called as an isthmus or bridge. The number of edges in the smallest cutset of G is called the edge connectivity of G. It is denoted by 2. (G). e.g. Consider the following graph G. Cutsets of G are as follows : i) f{e4, 5, 6}, ii) (ey, €3, eg), iii) {e;,e2}, * A set {e;, 2, €3} is not a cutset because its subset {e), e>} is a cutset. The edge connectivity of graph G is 2. ie. AG) =2. Fig. Q.28.2 Consider the following graph Gy. Graph Gy has edge connectivity 1 as G, - {e,} is a disconnected graph. “ e; is an isthmus or Bridge. G — ep is also disconnected graph. -. €> is also isthmus. Ay eu A Guide for Engineering Stude"® _piserete Mathematles - . Graph Theory 2) Vertex Connectivity : The vertex c ivi . connected graph G is defined as pemncetivity k(G) of a simple smallest number of vertices whose removal . y, disconnects the graph. ; In graph G, the sets {v>, vs, v4}, ‘: e (v2, V3» V4> V5}, {¥2, V3} disconnect graph G. The smallest set is {v2, v3} v3 % ek@=2 Fig. 0.28.3 1) A graph G is said to be k-connected if its vertex connectivity is k. 2)A graph G is said to be seperable graph if its vertex connectivity is one. 3) A vertex v of a connected graph G is said to be cut vertex if G - {v} is a disconnected graph. 4)k (G) $A(G) $8 ie. one connectivity < edge connectivity < minimum degree of a vertex in G an ’@s [=] n where e = Number of edges in G. n = Number of vertices in G. Q29 Explain shortest path algorithm and Dijkstra's algorithm. ES [SPPU : May-05, 07, 14, 15, Dec.-06, 07, 12, 13, 14, 15] each edge e of a graph a real Ans. : Suppose there is associated to ight of e, A weighted graph is a number w (e). w (e) is called the wei graph in which each edge has a weight. The weight of graph G is the sun of weight of all edges of G. Weighted graph has many applications in communication networks. Given @ railway network connecting several sities, determine a shortest route between two cities. We consider the weighted graph where the vertices are the towns, rail Toads are the edges and the weight represent the distance between directly linked cities, Therefore weights are non negative integers. The problem is tivity two given cities. Of all paths, find their mies and i ficient. So we required different edges) this may not be efficient. 1 if gee aes problems, Cee slgorithm was found by Dijkshtra 59) and is known as Dijkshtra’s algorithm. “A Guide for Engineering Students| Dp. 3-32 Graph Theor, A)Dijishue’s algocithm to find the shonest path fom the vertex a vertex z of 2 graph G. Let G (wv, E) be 2 simple graph and a, z< Vy. uppose is the label of Bi S length of Si L(x) is the Ibel of the vertex which represents the 21 shortest pech from the vertex 2. Weight of an edge ey = (3, v,) Consider the following steps : Step 1: Let P be the set of those vertices which have permanent and T = set of all vertices of G. Sal @ = O0L@=< P=0 adT labels eTadxeza Step 2: Select the venex yin T which has the smallest label. This labe] 'S called the pemenent lebel of v. Also set P as PU {s} and T as T- fs}. If v =z then L (2) is the length of th: Z and stop the procedure. Step 3: ify Z, then revise the labels of the vertices of T. ie. The Vertices which do not hav Permanent labels. The new label of x in T is en by L(x) = min {old L(x), Liv) = wy, x)} where w (v, x) is the wei ht of the edge joining v and x. If there is no edge joining v and x then tke w (v, x) = shortest path from the Vertex a to Step 4: Repect the steps 2 Examples : Q.30 Use Dijkstra’ and z. and 3 until z gets the permanent label. 's algorithm to find the shortest path between 2 FSISPPU : May-05, 14, 8 Marks, Dec.-06, 6 Matis! Fig. Q.30.1piscrete Mathemats 3-33 Graph Theory Ams. = step 1+ P=, T={abode, fz} L {a} = 0 L{x} =o, WxreT, x#a Step 2+ v = a, the permanent label of a is 0 P = {a}, T= {b,c, de, f, z} L {b} = min {old L (b), L (a) + w (a, b)} = min {co , 0 + 22} = 22 ‘L {c} = min {0 , 0 + 16} = 16 L {d} = min {, 0+ 8} =8 L {e} = min {0 , 0 + 00} = 00 L {f} = min (2, 0 + 2} = 00 L {2} = min {0, 0 + 0} = 00! i L {4} = 8 is the minimum label. Step3: v = d, the permanent label of d is 8. P = {ad}, T= {b,c ¢, fz} L {b} = min {old L (b), L @) + w G, b)} = min {22, 8 + 0} = 22 L {c} = min {16, 8 + 10} = 16 L {e} = min {e, 8+ o} = L {f} = min {~, 8 + 6} = 14 L {2} = min {0, 8 +o} = 5 L {f} = 14 is the minimum label. Step 4 ; v = f, the permanent label of f is 14. P= fad, f}, T= {bc ¢ 2} L {b} = min {old L (6), L(+ wb, H} = min (22, 14 +7} =21 L {c} = min {16, 14 + 3} = 16 L {e} = min {oo, 14 + =} = 0 A Guide for Engineering Students —-34 Discrete Mathematics 3 L {2} = min {e, 14+ 9 }= 23 “ L {c} = 16 is the minimum label. Step 5: v = ¢, the permanent lable of c is 16. P= {ad,f,c}, T= {bez} L {b} = min {old L (b), L ( + w (£ b)} = min (21, 16 + 20} = 21 L {e} = min {o, 16 + 6} = 22 L {2} = min (23, 16 + 10} = 23 . L {b} = 2 is the minimum label. Step6: vy = b, the permanent label of b is 21. P= ad, fc, b}, T= fe, z} L {} = min (old L ©, L &) + we, b)} = min (22, 21 +2} = 22 L @} = min (23, 21 +2} = 23 * L {c} = 22 is the minimum label, Step 7: v =e, thé permanent label of ¢ is 22. P= fa, d, fc, be}, T= {2} L@ = min fold L @,L ©) + woe, 2} min’ {23, 22 + 4) = 23 which is the minimum label Step 8: v = 2, the permanent label of 2 ig 23. Hence the length of the shortest path from a to Z is 23, The shortest path is adfe or adfbz, Fig. Q.30.1 (a) A Guide for EngineeringMathematics 3-35 Graph Theory perce MOO A35__ Graph Theory Fig. Q.30.1 (b) Q31 Find the shortest path from a-z in the given graph using Dijkstra's algorithm. IS [SPPU : May-07, Dec.-07, 15] Fig. 0.31.1 Aus. : Step 1 : StP=9, T= {a,b,c de, fz} L {a} = 0 Li} =, vxeT, x#a Stp2: v= a, the permanent label of a is 0. P= {a}, T= {b,c,de £2} 1 } = min {old L (), L @ + wb} = min {00,0 +2} =2 14) = min fo, 0+1}=1 L (a) = min {~,0+4}=4 LG} = min foo, 0+ 00) = eo L {f} = min (0, 0+ =} =o LG) = min foo, 0 + 00} = o0s.L {c} = 1 is the minimum label. ‘A Guide for Engineering StudentsGraph Theory Discrete Mathematles 3-36 Step 3 : v = c, the permanent label of c is 1. P= {ac}, T= {b,d,e, f, z} L {b) = min (2,1+2}=2 L {d} = min (4,1+2}=3 L {ec} = min (9, 1+5}=6 L {f} = min {0 1+7}=8 L {2} = min {o, | + 0} = 00+, L {b} = 2 is the minimum label, Step 4: v= b, the permanent label of b is 2. P= {acb}, T={de, fz} HAG) = min 3,240)=3 L fe} = min 6,243} =5 L (8 = min (8,240) =8 L {2} = min {e, 2+ 09} =o LL {@ = 3 is the minimum label. Step 5 : v =, the permanent label of d is 3. P= {ob,d), T= fe £2} Le) =mng3+e-5 Ley = min {8,3 + 4} =7 L {2} = min {00, 3 + 09} = coy, L {e} = 5 is the minimum label, Step 6: v =e, the Permanent label of e is 5, P = {a,¢, b, d, e}, L {f} = min (7,5 +} =7 “LQ = Step 7: T {6 2} L {2} = min {e, 5+ 1} =6 6 is the minimum label, Vv = gz, the Permanent label of z is 6. Hence the length of shortest path from a to z is 6. The shortest path is a b e Ze Fig. Q.31.1 (a) Q.32 Find the shortest path between a-z for the given graph : usin8 Dijkstra's algorithm. SS'[SPPU : Dec.-12, 13, 14, May-15, 8 Mars] AGuide for Encineering StuielFig. Q.32. Step 1: Set P= ©, T = {a, b,c, dye, fg, z} L{a}=0. L(x} =o, VxeT, x#a Step 2: v = a, the permanent label of a is 0. P= {a}, T= {bode fg, z} L {b} = min {old L (6), L (a) + w @, b)} = min {0+ 2} =2 L{c} = min {@,0+4}=4 —L {d} = min {0 +5} = Life} = L {fh =L (g}=L Z = +L {b} = 2 is the minimum label. The permanent label of b is 2. Step3:v = b P= {ab}, T={cde fg, 2} L {c} = min {L (0), L (b) + w {b, c}} = min (4,2 +1) =3 L {4} = min {5,2 + oy =5 L {e} = min {0,2 +3} =5 L {f} = min (2+ 1} =3 L {g} = L {z= * Lc} Step 4: y L {f} = 3 are the minimum labels. 0 corf Vv = f, permanent label of f is 3. P= {ab,f}, T= {c, deg, 2} L(e} = min G,3 +2) =3 L {d) = min (5,3 +1} =4 Le) = min 6,342) =5 L {g} = min {3 +4} =7 1) = min (0,3 +7) = 10 “ (©) = 3 is the minimum label. A Guide for Engineering StudentsGraph Theo, +38 Discrete Mathematics 3 Step 5: v = ©, permanent label of c = 3. P = {a,b, f,c}, T= {d,¢, g, 2} “L (a) = min (4,342) =4 L {ce} = min 5,3 +o} =5 L {g} = min (7,3 + 0} =7 L {2} = min (10, 3 + 0} = q9 L {d} = 4 is the minimum label. Step 6: v = d, permanent label of d= 4, P= {a, b, fc, d}, T= {e, g, 2} L {e} = min (5, 4 + 29} L {g} = min {7, 4 +2} =6 L @) = min (10,440) =10-, L {e} = 5 is the minimum label. Step 7: v = ¢, permanent label of ¢ is 5, P= {a,b fc, 4,6}, T= {g, 2} L {8} = min (6,5 + 00} =6 L {2} = min {10,5 +3} =g L {g} = 6 is the minimum label, Step 8: v = g, permanent label of g is 6. P= @bfodeg), T= {2} L {z} = min {8, 6 + 1} = 7. which is the minimim label, Step9:y = 2, permanent label of z is 7, Hence the length of shortest path from a to z is 7. The shortest path is a b f dgz ne 9 i Sent § J Fig. 0.32.2 s A Guide for Engineering Stude™secrete Mathemtatles 3-39 Di 3 Define Eulerian path ang circuit, a A path is called an Euler = exactly once in the path, icuit of a graph which contains eve, ds AoE Eulerian cirenit, TY edge of graph exactly once is ¢f Graph Theory an path if every edge of graph G A graph which has an Eulerian circuit is called as Eu lerian graph. The problem of find Eulerian path is the same as the Problem of drawing © Paper and without Tetracing an a network without lifting the pencil off th edge: Consider the following graphs : Fig. Q.33.4 In graph G,, Eulerian circuit is e - €2-€3 -e4 -e5 -e +G, is an Eulerian graph. In graph G>, Eulerian circuit does not exist, ““G) is not Eulerian graph, In graph G3, Eulerian path ise; >e, >€3; ey >e, >es but G; does not have any Eulerian circuit. “G3 is not Eulerian graph. Sy is also not an Eulerian graph. ‘xistence of Eulerian paths and circuits in a graph depends upon the “Eee Of vertices heorem 4 : An undirected graph possesses an Eulerian path iff it is Th “ted and has either zero or two vertices of odd degree. 180) Conn 2? An undirected graph possesses an Eulerian circuit iff it is "ected and it Orem re) its vertices are all of even degree. Laer! SrA directed. graph possesses an Eulerian circuit iff it is and incoming degree of every vertex is equal to its outgoing A Guide for Engineering StudentsDiscrete Mathematics Examples : onditions Kp,q the complete bipary,, Find under what conditi Q.34 a) Jerian circuit. graph will have an Eul 2 TS |SPPU : Decoy b) What is the complement of Kn, n? i eS. Ans. : a) In K,, , consider the following cas Case 1: m=n and both m and n are even : In this case, degree of each vertex is even, Hence by . a theorem I, Kin will have | Fig, 0.34.4 an Eulerian ‘circuit, For example Ky,» and Kg Case 2: Ifm=n and m,n are odd : Re In this case degree of each vertex is odd. Hence Eulerian circuit will Not exist. K. Case 3: Ifm ¥n but m and n are even « t at In this case, degree of each vertex is even. So Fig. Q.34.2 there exists an Eulerian circuit, ee Case 4: Ifmen and either m is odd orn is odd or both are odd : Then graph will have Vertices of odd degree, i i ig. Q.34. does not exist. e.g. K3, 3, Fig. 34.3 ) The complement of K,, wn #5 the two graphs Ky, and Ky. 9.35 Define Hamiltonian graphs, Ans. : In this Section, we introduce a class of Braphs which possess ¢ strinking similarity to Eulerian graphs.’ CS |SPPu : Dec.-04, 09, 10, 12, 13) A circuit in a connected graph G is called a Hamiltonian circuit if? contains every vertex of G exactly 0; nce, A graph which has a Hamiltonian circuit is called a Hamiltonian Graph. nS A Guide for Engineering Stupcrete Mathematics 3-41 Graph Theory consider the following graphs : iy 2 3 g d x ane | " € 2 é 3 4 b a u go 2 6 G, & & Fig. Q.35.1 In graph Gj, Hamiltonian circuit is 1-2-3-4-5-6-1 <.G, is a Hamiltonian graph. In graph Gp, Hamiltonian path is a-b-c-d but Hamiltonian ‘circuit does not exist. =G, is not Hamiltonian graph. In graph G3, Hamiltonian circuit is x-y-z-u-v-x :.G; is Hamiltonian graph but it is not Eulerian. In graph Gs, Hamiltonian path is a-b-c-d-e but Hamiltonian circuit does not exist. =-G, is not Hamiltonian but Eulerian graph. Important theorems : Theorem 1: Let G be a simple connected graph on n vertices. If the sum of the degree of each pair of vertices in G is (n — 1) or large then there exists a Hamiltonian path in G. Theorem 2: If G (v, E) is a simple connected graph on n vertices and n . . A 4()=3 sve V then G will contain a Hamiltonian circuit. This condition is sufficient condition but not necessary. Theorem 3: Let G (v, E) be a connected simple graph. If G has a comlonian circuit then for every proper non empty subsets of v, the : ponents in the graph G-S is less than or equal to the number of ettices in g, Theorem 4: a Hamiltonian graph contains no cut vertices and hence i 2connected, grap! nce is iconEy = A Guide for Engineering Studentsoe Graph 7; iscrete Mathematics 3-42 heoy Q.36 Show that the complete bipartite graph K,,,,, Js Hamilto, for m= n and for m+, Ky,, is not Hamiltonian graph. Ans. : In a complete bipartite graph Kn form =n ie. Ky 4, degreg of each vertex is n. Therefore d (v) 25 for all Ve v (Ky, n) Mian By theorem 2, G contains a Hamiltonian circuit. Hence K,, , is a Hamiltonian graph. | If m = n, Let Vi and V, be the Partitions of the vertex set of Kan Where \vil=m and \v2|=n. Without loss of generality assume that m 3 and degree of each vertex is n-1l.Asn>3, d(v) = n-1255VveV«,) Therefore by theorem 2, Hamiltonian graph. Hamiltonian circuit contains all vertices of Braph and length of circuit s the number of vertices present it. Hence in Ky, the length of t¥ Hamiltonian circuits in Ks- Kn has a Hamiltonian circuit, Hence Ky is# a! Hamiltonian circuit is n and there are op The complement of K,, is the null graph on n vertices, Q.38 Find the Hamiltonian path and circuit in K4,3 ? BS [SPPU : Dec-tl ing suet A Guide for Engineering St!piscrete, Mathematics ha Ans? ‘The complete bipartite graph a js given by jn Ka,3» 4 #3 Hence it does not contain amiltonian circuit. Here degree of each vertex is either 3 or 4. For x, y any two vertices in Kq3 dx) + dy) = 7.1 = 6 Hence by theorem 1, the graph Ka has a Hamiltonian path. It is given by x ay bzew. 9.39 Give an example of the following graphs Graph Theory Fig. Q.38.1 a) Eulerian but not Hamiltonian. b) Hamiltonian but not Eulerian. ¢) Eulerian as well as Hamiltonian. d) Neither Eulerian nor Hamiltonian. Ans. : a) Eulerian but not Hamiltonian graph. Eulerian circuit: abcdeca No Hamiltonian circuit. b) Hamiltonian but not Eulerian Hamiltonian circuit : abcdea No Eulerian circuit because d(b) =3, © Eulerian and Hamiltonian Braph, Renittonian circuit : Mba, 1.2.3-4-1 Biletion circuit : a, 1-2.3.4.1 abeon themattes wed Ph Theory Jathem ~ ——— ad) Neither Bulerian nor Soe Hamiltonian Y No Hamiltonian cireuit and no 30.4 (c) Bulerian circuit, Fig. 0.30.4 (¢) Q.40 Determine, if the following kraphs are having the Homittonian clrewit or path, Justify youy answer, IS[BPPU ; Dacay a, Fig. Q.40.1 Ans. : In graph Gy, there are n = 7 vertices, 44) = 4 and all Temaining vertices jg 2. So to draw Hamiltonian circuit, we have to visit vertex 4 twice. Which is ROt possible in Hamiltonian path. G, has no Hamiltonian circuit. Hamiltonian path is 1-2-3-4.5.6-7, In graph G3, In graph Gp, a@=$ 35 xKE VG) ~. By theorem 2, 3 a Hamiltonian circuit +. Hamiltonian circuit is 1-2-3-4-5-6-1, Q.41 Which of the following have a Euler cir cuit or path or Hamiltonian cycle ? Write the path or circuit, ISPSPPU : Dec.-10] a A Guide for Engineering StudeGraph Theory Fig. Q.41.4 gas: In graph G,, degree of each vertex is an evenso3 an Eulerian Gqeuit which is a-b-c-d-e hn gaph G, there are 5 vertices and degree sum of every pair of vertices is 4 or greater than 4. Hence there exists a Hamiltonian path in G, which is given by a-b-c-d-e. But there is no any Hamiltonian circuit as vertex c jg a vertex. In graph G2, degree of each vertex is 5 which is odd integer, so here is no Eulerian path in Go, degree of each vertex is 5 > 5. Hence ere exixts 2 Hamiltonian circuit which is given by a-b-c-d-a. . in G;, Eulerian path is a-b-c-d-e-b No Eulerian circuit as d (a) = 1. Esniltonian path is a-b-c-d-e. ‘No Hamiltonian cyle because b is a cut vertex. 3 Travelling Salesman Problem (TSP) 242 Define the Travelling Salesman Problem (TSP). Ans. : A salesman is required to travel a number of cities during a tip. Given the distance among cities, in what order should he travel so that he taels as minimum distance as possible 2 This is known as Travelling Slesman Problem (TSP). 1 tems of graph theory, the TSP is to a Hamiltonian circuit with the ae weight. In the case of K,, the problem can be ed theoretically by listing all the possible 3 tonian circuits and select one which has least | .Beht But this method is highly impractical for the | agh, SaPhs. In fact no efficient algorithm is there 10 rmanglSP Tt is. therefore desirable 10 obisin © Fig, a.42.4 ly good but not an optimal solution. E E k ‘A Guide for Engineering StudentsBig Criph 7, Math h is to first find 2 Hari i oe aoe % chr Haniionan ese of Teser eg. The sple etis : follows : Let C be the Hamiltonian circuit of a graph G. Let further uv and xy be two no-adjacent edges of C such that ay vertices u, v, x and y occur in that order in C. If ux and vy are edges such that v Untwywye-w (uy) + wy (xy then replace the edges uv and xy in (by and vy). The new would still be Hamiltonian cycle and w (C) < w (C). This process ¢ continued until one gets a Teasonably good Hamiltonian cycle, Q.43 Explain nearest neighbour method. Ans. : In this method, we Start with any arbitrary vertex an vertex which is nearest to it, Continuing this way end coming back te starting vertex by travelling through all the vertices exactly once, get Hamiltonian cycle or circu; it. Consider the following steps to find Hamiltonian cycle by Step 1: Start with any arbitrary vertex Say V1, choose the v: to v; to form an initial path of one edge. Construct this path different vertices as described in step 2. Step 2; Let v,, be the latest vertex that was added to the vertex v,._ 1 Closest tov. from all vertices that ar. and add this vertex to the path. Select those Vettices which @ circuit in this stage. Step 3: Repeat step (2) till all the vertices of G are inched in path. Step 4: Lastly form @ circuit by adding the edge comnecting to the last added vertex, i" a * The circuit obtained using required Hamiltonian circuit. Note : If we start with an arbitrary vertex in TSP then we may or 2 not minimum Hamiltonian circuit, But if we siart with a vertex wie! incident edge has the minimum weight in minimum Hamiltonian Circuit as compared wit For more details see Q.44, the nearest Neighbour method will be 2 Jaa3-47 Graph Theory examples ¢ gas Use nearest neighbour method to find the Hamiltonian circuit garting from 8 in the following graph. Find its weight, BSP[SPPU : Dec.-15] Fig. Q.44.1 Ans.: Step 1: Let a be the starting r “3 vertex, Vertex a is adjacent to b, c, d, x ¢ But minimum path is {a, d} which is the initial path. Step 2: There are three vertices adjacent to d. but closest one is C : The path is {a, d, c}. a Step 3: There are 4 vertices adjacent ‘0 c. but closest is e * The path is {a, d, c, e}. Fig. 0.44.1 (c) A Guide for Engineering Students38 7 : e. ices adjacent t0 are 4 vertices 4: There josest is b The path i (2&6 6B) Step 5: Here are all vertices are corered so to complete Hamiltonian an circuit there should be a path from b to a. -. Hamiltonian circuit is {a, d, c, e, b, a} Weight of the Hamiltonian circuit = 40. Fig. Q.44.1 (e) 4: Planer Graph 245 Explain planar graphs. pa>igppy ; May-06, Dec.-08, 09, 10, 13] Ans. : Definition : A graph is said to be planar graph if it can be drat on a plane such that no edges intersect or cross in a point other than the! end vertices. A graph G is said to be non-planar if it is not possible to draw graph be without crossing. 1. Regions : A plane representation of a graph divides the plane inte Parts or regions. They are also known as faces or windows or meshes. A region or face is characterised by the set edges forming it’s boundary: A region is said to be finite if it’s area is finite, A region is said to infinite or unbounded if it’s area is infinite. Every planer graph has ® infinite region. : . A Guide for Engineering Stte™Mathematics 3-49 Graph Theory e er the graph given below : Fig. 0.45.1 qhe graph G has 6 regions, 7 vertices and 11 edges. Region R; is an ‘pfnte region known as exterior region. We have R> = {1, 2,3} = Region bounded by e}, e2, €3 R3 = {3,€4) 5,6} Ry = {€6,¢7}, Re = {e109}. Iris observed that n= 7,e = 11,r=6 ntr-2=7+6-2=ll=e Now let us define Euler's formula. 046 State and prove Euler's formula. Aus: Statement : For any connected planar graph G, with v number of vertices, e number of edges and r number of regions v-e+r=2 ove+r-2=e Proof : Let G be a connected planar graph with v vertices, e edges and T regions. We shall prove the theorem by induction on e. Sep 1: For e = 0, we get v 1, Thus Vretr=1-O0+1=2 Hence result is true for ¢ = 0 Bt de Let e > 1, Assume that the result is true for all connected Sraphs with less than ¢ edges. Let G be a graph with v vertices, “ts and r regions, Stay % ae Case 1: If G has a pendent vertex say x then G — {x} is a led graph with v — 1 vertices, e - 1 edges and r regions. A Guide for Engineering StudentsO™ Graph = 50 Ph Theo Discrete Mathematics 2 Sy Discrete Mathematies 00 So by induction hypothesis W-N)-@-Y)+rs 2 veetrs2 Case 2: If G has no pendent vertex the G is a ae zee With circuit. Let ¢, be the edge of a circuit in G. Then G- ae enn graph with v vertices, e - 1 edges and r — | regions {I VE dye from a circuit, then it reduces region by 1}. By induction hypothesis v-(e-1)+(r-1) = 2 = v-e+r=2 ‘Thus by the principle of mathematical induction the result is true for alle, 2.47 If G (V, E) is a simple connected planar graph with v vertice and e edges then e <3v 6, Ans. : Proof : Give that, G is a simple planar graph, so each region of G is bounded by three or more edges, If G has r number of re gions then the total number of edges in G is e234. Also cach edge of G is included in exactly two regions of G. therefore 2e 234 = %s, Substitute these values in Euler's theorem, we get Vre+r=2 2e ~e+ > v-e 7 22 3v-e 26 es 3v ~ 6 Hence the proof, Corollary 2: Prove that, K. graph on 5 vertices) is not ph Proof: The complete graph on $ vertices Ks is given below : 5 (the complete lanar. Ks has 5 vertices and 10 edges. ie, v= 5 and € = 10, nf A Guide for Engineering Stdpiserete Mathematics 3-51 Graph Theory’ now 3v-6 = 15-6 =9 py corollary 1, & <$3v-6 10. <9 which is impossible, Therefore Ks is not planar graph. Ks is the smallest planar graph with respect to number of vertices, Consider the graph K3 5 | Here v = 6,e=9, : 3v-6 = 18-6=12>9=e ie e S$ 3V-6 But K;,; is not a planar graph. «The graph K3,3 is the smallest non planar graph with respect to number of edges. The graph Ks is called the Kuratowski's first graph and K3,3 is called te Kuratowski's second graph. 'n 1930, Kuratowski gave a necessary and sufficient condition for a graph ‘0 be planar, Kuratowski's Theorem : A-graph G is a planar if G does not contain “ty subgraph that is isomorphic to within vertices of degree two to either Ks orks 5 Two graphs are said to be isomorphic to within vertices of degree two if ‘hey are isomorphic or they can be reduced to isomorphic graphs by "Prated insertion of vertices of degree 2 or by merging the edges which V€ exactly one common vertex of degree 2. For example the following graph are isomorphic to within vertices of ¢ 2. (Homeomorphic) A Guide for Engineering Students SAGraph Ty, te Discrete Mathematics ny moe oN Insenionofv G Dalotion of x from ¢ Insertion schvatiene erating aha mS, Fig. 0.47.3 Q.48 Draw a planar representation of graphs given below if possipi. SS>[SPPU : May-06, Deg, 09) Fig. Q.48.1 Ans. : The Planar een) ws 6 and Gai is as s follows : Fig. Q.48.1 (a) Q.49 Identify whether the Braphs are planar or not Justify ? IS |SPPU ; Dect (i) Fig. Q.49.4 = A Guide for Engineerit8fic a piserete Mathematles 3-53 Grape Tadary ans ) Fig. Q.49.1 (a) Given graph is planar graph, ii) Given graph is isomorphic to K3,3 x = .. Given graph is not planar. Dex] Fig. Q.49.1 (c) *-Given graph is planar. 250 Show that in a connected planar graph with 6 vertices, 12 edges each of region is bounded by 3 edges. p@>[SPPU : Dec.-10, 13] Ans, : According to Eulers theorem for planar graphs. V-e+tr=2 Here v= 6,e=12 S-trs251=8 / ; , es know that, each edge contributed twice in a regions we have 12 Res,3-54 _Daserete Mathematics Grape Pre, 24 edges are distributed among § regions. So 12N2=2 = 3 edges for each region, So each region is bounded by 3 edges. Q.5t Prove that X5,3 is not planar graph. has 6 vertices and 9 ed each region has at le; a ) ges. Suppose is Planar, th ast 4 edges because it ig dite contains no tiangles. Each edge lies on boundary of two regions : Therefore, 2e > (the numberof edges in the jt Tegion) i=l 7 2e > ar 2e 24 -y) =. © s2-4 Bu e=S9andy=6 eB 9S 12-458 which is impossible. is not planar graph. Graph Colouring Q.52 Explain coloring of graphs. iz Ans. : The coloring of all Vertices of adjacent vertices have different colors is ¢ coloring or simply a coloring of graphs, A graph G is said to be properly colored Staph if each vertex of GS colored according to a Proper coloring. “8 1) Consider the following graphs with Proper coloring 2 connected ore alled a Proper col Gy Fig. Q.52.4 A Guide for Engineerin. a‘Mathematics 3258 piscrete “Graph Theory 1 chromatic Number of Graph : The chromatic number of a graph G is denoted by X(G) and defined as the minimum number of colors required to color the vertices of G so that the adjacent vertices get different colors. ‘A graph G is said to be K-colorable if all vertices of G can be properly colored using at most K different colors, Obviously, a K-colorable graph js K+ colorable. If G is k-colorable then X (G) 2 vertices, Theorem 1: 4 graph G is a tree iff there exists a unique path between every distinct pair of vertices of G. . is a connected and without Circuits, We know that a circuit forms two or more paths. But G has NO Circuit 59 N every pair of vertices, Conversely assume that there is a unique path between every pair of Wertices in G. This implies that G is a connected and G has no cycles. ie, G is 2 connected acyclic graph. Therefore G is a tree, Theorem 2: 4 graph G on n vertices is a tree iff G is connected and has exactly nj edges. Proof : Suppose graph G is a tee. Therefore’ is a connected and acycli geph. . It is sufficient to Prove that G has n — 1 edges only. We can prove this by induction principle on n, For n = |, the result is obvious Let n > 1, consider G~e’ for any ¢’ € E(G) As G is a tree, e’ is an isthmus of G, ~G-eisa disconnected graph with tw Now G, and G subgraph of G, ‘© components say G, and G2: a Gy ‘2 are connected and a cyclic graphs as G, and 6 su A Guide for Engineerin® 4yr ‘Mathematics 3-59 Graph Theory tet Gy bas vertices and mj, edges‘and G2 has n> vertices and mz edges induction principle . By m = 1-1 and m) =n2-1 erefore the number of edges in G is given by e = (e+e) +1 = ny-l+ng 141 = nytny-1 e=n-1 Co n= ny+n2) Heace G has n — 1 edges only. coaversely assume that G is connected graph and has exactly n - 1 edges. Cisim : Prove that G is a tree, Iris suffices to prove that G is noncyclic graph. Seppose G contains a cycle C. Let P denote the number of vertices in C . The number of edges in C = P 4sG is connected graph, the remaining n — P vertices must be connected vertices in C. To connect each vertex of G which is not in C, we required n - P edges as each edge of G can connect only one vertex to te vertices in C. Hence the total number of edges in G is given by ¢ = (n—P) +P =n= e =n which is contradiction + Ghas n~1 edges only. Thus G is a tree. Theorem 3: Let G be a graph with n vertices and m edges. If any of fe following is true then all are true. ) Gisa tee, ") Gis connected and m =n — 1 4) G is acyclic graph and m =n - 1 1) Every edge of G is an isthmus and G is connected. There is exactly one path between every pair of vertices in G. ‘A Guide forGraph tes 3-60 ee Thy Discrete Mathemat eorem non trivial tree contains at least two vertices of q, Th 4 ivial tree least A i ices) 1, (ie. pendent vertics ; Pr i: Let G be a tree with n vertices, rool : : diet; vxeV (G) By handshaking lemma airs ; Saixy = 2 x Number of edges in G = 2(n-1) so G is connected. xe V(G) Suppose there is no vertex of degree 1. Then 2ns Sd(x) = 2n-2 xe V(G) he. 2n $2n-2 = nSn~1 which is impossible. Thus there is at least one vertex of degree 1. By a similar argument there is one mote vertex of degree 1. Hence every tree has at least two pendent vertices, Q55 Draw all non isomorphic trees on 4 and 5 vertices, Ans. : 2) Non isomorphic trees on “oN” “~~ 4 vertices Moneta : 5) Non isomorphic trees on 5 vertices At Q56 Under what conditions trees are the complete bipartite graphs. Ans. : Suppose T is a tree which is the complete bipartite graph. Let Te Kin * T has m +n vertices and (m+n~- But ky, has mn number ‘Therefore 1) edges, T of edges, m= m+n- 1 m-m-n+] = 9 (m-1)@~1) = 9 i? A Guide for Engineering"ithematics 3-61 Mal Graph Theory m= lorn=1 2 ee m = 1 and anyn or n= 1 and anym tice Kin and K,,,; are the only complete bipartite graphs. These are jnown as star graphs. qhus Tis a star graph, 57 a) Is it possible to draw a tree with 10 vertices which has vertices either of degree 1 or 3 2 Ifpossible draw tree. Is it possible to draw same type of tree with 13 vertices ? ) For which values of n (number of vertices), such type of tree exist 7 Ans. : a) Given that tree T has 10 vertices so it must have 9 edges, Let x and y be the number of vertices of degree 1 and 3 in T respectively. 2 x+y = 10 ws (QS7.1) By handshaking lemma DY d(v) = 2Number of edges in T) Vev) x+3y =2x9=18 x+3y = 18 w (Q57.29 Solving equations (Q.57.1) and (Q.57.2) we get y=4 and’ x=6 Therefore there are 6 vertices of degree 1 and 4 vertices of degree 3. Such type of graph is given below. Now consider a tree with 13 vertices and 12 edges. BY using similar theory, we get (Q57.3) xty = 13 -(Q51. we (Q57.4 ee (9574) ;Discrete Mathematics soe Solving (Q573) and (Q57.4), we get ly = l= | which is impossible. Therefore it is not possible to draw such type of tree. b) Let T be a tee with n vertices. Let x and y be the number of vertices of degree 1 end 3 5. | respectively. T has n— 1 edges. Therefore + ; x n ~~ Qs75 By handshaking lemma x+3y = %n-1) / _ Bt3y = 2n-2 ~ (Q574 Equati ion (Q.57.6) — Equation (Q.57.5) = 2y = 2n=2_p 2y =n-2 = = B-2_n y aoa! = er x=n-y Bvgtlasel 1 These should be non negative integers. = dition on even number of vertices. 26 13,76, 7, ... Thy are such type of trees. Case 2: If n js ai i odd then 5 is not an integer there x and y are not © integer. Hence the tequired tree does not exist on odd number of vertices. A Guide for Engineeringpiscrete. Mathematics 3-63 Graph Theory 0.58 Explain centre of a tree, ‘Ans.: We know that the distance between two vertices in a connected graph ive. is a length of the shortest path between that vertices, Before defining centre of a tree first find the distance between vertices of a tree which is defined as follows : Eccentricity of a Vertex Let G be a connected graph G and ve V(G). The eccentricity of a vertex y is denoted by E(v) or e(v) and defined as the distance from v to the vertex farthest from v in G. ie. E(v) or e(v) = max d(v, v;)/Wv; € (G)} Centre of a Graph A vertex in a graph G with minimum eccentricity is called a centre of G and it's eccentricity is called as radius of G. It is denoted by 1(G). Q.59 Prove that every tree has either one or two centres. Ans, : . , Proof: Let T be a tree. Then e(v) for a vertex v is at a vertex farthest from v. As T is a tree, it is attained at a pendent vertex. We know that every non trivial tree has at least two pendent vertices, Now, delete all pendent vertices from G, we get new graph G’ which is also.a tree. Moreover as one edge at each pendent vertex is removed, the Sccentricity of any vertex will be reduced by one. Therefore centres of G Will still remain centres of G’, We continue the process with G’ until we ative at Kj or Ky. K3 has two centres and K, has one centre. Hence the proof, 60 Define cut vertex of a tree. ANS: We know that the vertex v whose removal from a connected Saph G, disconnects the graph is called as cut vertex of G. any tree all vertices except pendent vertices are cut vertices.Q.61 Define rooted tree and binary tree. ee Pee.gy i directed yelic, directed graph is called a tree, Ans. : A connected acyclic, directe a be a directed tree it f°! other words, a directed graph is sai ee Will become a tree when the directions of the edges are ignored. Q.62 Define rooted tree, : | Ans. : A directed tree is called a rooted tree if there 1S exactly oy Vertex whose incoming degree is zero and the incoming degrees oy other vertices are one, The vertex with incoming degree 0 is called the root of the Footed tree The vertex whose outgoing degree is zero is called leaf oF terminal noje A vertex whose Outgoing and incoming degrees are non zero is called branch note or an internal node, Consider the following example, Q.63 Define level and height of a tree, Aus. : A vertex v in a rooted tree is said to be at level n if there is ¢ path of length n from the Toot to the vertex v, The height of the tree is the maximum of the levels of its vertices. Example In the given graph G .... ais a Toot Vertices b, c, d are at level 1 Vertices e, f, & h and i are at level 2 Vertices j,k, 1 and m are at level 3 Vertices n, 0 and P are at level 4 Vertices q and r are at level 5 The maximum level is 5, Therefore the height of tree is 5. a 4 Guide for Engineers‘Mathematics 3-65 Graph Theory for rooted tree : In a rooted tree < for Rul f the level of a vertes i level of vertex v V< Father} is greater than ’ then u is below y. f vertex ui 15 Below vertex v anid there is If vertex edge from v to u then uy is an edg\ id) Said to be of v (or child of vy ang y is said to ai the father of u (or parent of uy i.e. Two vertices u and v are said to be . brothers if they are the sons of the same Vertex. rothers ) A leaf is a vertex without children, m 4 ) Ep=tavi.va.vs ~~+level 4 evel 2 ~Hevel 3 From is father of Pande lie at level 1. b and c are sons of root a i.e. a and c lie at level 1. , Band c and b and c are brothers. ™) 0 has three sons d, e, f “P is a father of d, e, £ tact ™) ehas two sons i and j has only one son k | SS ering Students ‘A Guide for Engineering Sio 3-88 bro i ¥) i bas no soa j bas two sons J and m. So K and J are ome son nn has no brother as nis a leaf. ars a a as eae ee BR imnant soon 64 Define Subtrees. Ass : Let T be a rooted mee and xe V(TL ondary NA NAT decedins is cle Be subi ey rooted at x. The subaee corresponding to the ode i i somspoading fo any ther ode & ealled a proer ates PE Q&S Define M-ary trees. aoe gt EB sth even SMsvior node Bas at most m soos SENSE ES 5 Sold to be regular mary tree o: fll ary i branch node has enzctly m sons, . Semess Consider the following examples. Beary tree 2-arytree 2 sary tree 66 A regular m-ary tree with p interior nodes has mp + 1 nodes at all. TS ISPPU : Decl Ans. : Proof : Let T be a regular m-ary tree with n vertices. Out of n verti there are p interior vertices or branch nodes, Therefore there are t = n — P number of sons or leaves in T. But given graph is regular and p interior nodes. So the regular m-ary tree will have mp sons. — ae sat A Guide for Engineeringeas jg n0t son. is re ave tree has total (mp + 1) number of vertices mpt+1 gat 4 ppl pinary tree. E[SPPU Due, May, 1 ye ary tee is known as binary tree if every branch node has oot? sons. _. : pate words, @ tree 1m which there is exactly one vertex of degree 2 pies of the remaining vertices of degree or three, is called a binary wt iy tee js called as regular binary tree or full binary tree if cay branch node has exactly 2 sons or zero sori. following examples oe) ii) iy) usd bug Binary tree Full binary tree \Binary tree Not binary tree ae tres, instead of referring the first or second subtree of a branch fa use to the left subtree or tight subtree of the node. dy a binary tree traversal with examples. Say gate means visiting or processing all the no ni et traversal isthe visiting of each node of, @ hy Some sequence. ae te "80 types of traversing binary me Hp ath tig traversal ; In this method, a Proceeds from the root or the descendent of the first child Guide for Engineering Students A des of a tree. A tree only onceDiscrete There are three types A) Pro order traversal + In is traversal first, root node is eft subtree and then the right subtree as shown below: B) Post order traversal : If processes first the left subtree then the right non subtree and then at the last root of the , tree as shown below. i | ; C) In order traversal: It processes i 7} the left subtree first then the root and iG CY 215) at the last right subtree. L-ubires 3 | . = subtree The prefix "in" means root is processed in between the subtrees. It is shown as below. Consider the following binary tree. Reet Root Leh subtree ii ) Post order traversal A Guide for Enginee™sport Mathematics 3-69 ‘iy order traversal i Graph Theory. Left subtree Root 2) Breadth first traversal: In this traversal, the Processing proceeds horizontally from the root of all of its children, then to its childre's children and so on until all the nodes have been processed, That is first write node of zero level, then all nodes of level 1 then level 2 and so on, . The breadth first process for the above binary is shown below. : ABECDE! Root Level 4 Level 2 Q69 Explain conversion of general tree to binary tree. Ans.: Let us explain the method to convert general tree to binary tree with the help of the following example. Consider the following steps oes ee 2) To convert it into binary tree, © fitst identify the branch from the Parent to its first or left most child. QO MQ | se branches from each parent i i become left pointers in the binary tree. © OOOODiscrete Mathematics Step 2) Connect sibling, starting with the left most or first child, using 8 branch for each sibling to its right sibling. They are the right pointers in Step 3) Now remove all unneeded ‘ranches from the Parent to children. Therefore remove ASEB=3p, AF, F5H, Fol We get the following Tequired binary tree. itsyr sshematles soa eee Si Fiery i Graph Theory The steps involve to convert the given tree j ¢ into bina ry tree are as ce Binary tree ay Co, Ostruy us it te the labeled tree of the following ig algebraic expression (19 + (xx) [@[SPPU : May-10] typ! The ex oy Pression tree for the given algebraic expression is given Guide for Engineering Students AQ.72 Represent > binary tree. An the expression ((a+5)T 4)*(6-(5+9)) using ; ES ISPPU : Mayq 's. : The binary tree for the given expression is as follows : Q.73 Write and evaluate the expression tree shown below. ; } IS {SPPU : Deo 7 sd ‘A Guide for Engineering S™«ae Mathematics ahs en 5 Graph Theory ‘Ans. : The algebraic expression of the given binary tree is (28(5)+ 10/5)) - (7)* (12/4) — (2) It's value is 2* (5+ 2)—(7*(3-1))= 14-7= 14 Q74 Find the preorder, postorder and inorder traversal of the following tree. 1S [SPPU : May-10] Ans, : i) Preorder traversal “A Guide for Engineering Students3-4 rap, Discrete Mathematics 2, mbdbfktpwv . ii) Postorder traversal ie. bfdkhpvwtm iii) Inorder traversal ie. bdfhkmptvw Out of which 2" are leaves, Hence internal vertices = (2h! _1)_ 9h = 2-1-1 = 2h_y Q.75 Find the maximum of Possible. height of a binary. tree with 3 vertices and draw graph. : Ans, : Wehave n= 13 The maximum possible height of the é binary tree is i n-1 a. The Tequired graph as shown in Fig. Q.75.1, Fig. @.78-1 Q.76 Defi me prefix code and binary search trees. may” i IS [SPPU 2, 14, 16 as nt A Guide for Enelsscrete Mathematics 327: piscrete s Graph Theory ‘Ans. ? | : 1) Aset of sequences is said to be a Prefix cod le if no sequences in the set is a prefix of another sequence in the set, For example the set {000, 001, 01, i0, 1 sequence of symbols is present at the beginning o} these sequence are distinct, is a prefix code as no f another sequences. All The set {1, 00, 000, 0001} is not a prefix code becau use the sequence 00 is the prefix of the sequences 000 and 0001, : A question comes in everyone's mind "How to construct prefix code to a full binary trees?" We can solve this question by adding some flavours and binary codes to 2 full binary trees. For a given full binary tree, we label the two branches incident from each internal node with 0 and 1. For the left branch we assign 0 and for the tight branch we assign 1 of every rooted tree or subtree. Consider the following example of full binary tree. In given figure root a has 2 sons. Left som is ab and right is af, so assign 0 to ab and 1 to af. Now b has two sons. Assign 0 to left son be and | to Tight son bk. Similarly assign 0 or 1 to every edge of a tree. Lefside (root) Rightside Now assign to each leaf, a sequence of O's and I's which is the sequence Of labels of the edges in the path from the root to the leaf. For example, d is a leaf and the path a to d is 4~b-—c~d and their respective labels are 0 - 0 °F d is 00. 0 so the prefix code For leaf e, path is a — b - ¢ —€ and labels 0-0-1 * The prefix code of e is 001 Thus the prefix code of above tree is { Reese “1 Guide for Engineering Students 000, 001, Ol, 10, 110, 111}.