EUROPEAN JOURNAL OF PURE AND APPLIED MATHEMATICS Vol. 12, No. 2, 2019, 499-505 ISSN 1307-5543 – www.ejpam.com Published by New York Business Global On Topologies Induced by Graphs Under Some Unary and Binary Operations Caen Grace S. Nianga1,∗, Sergio R. Canoy Jr.1 1 Department of Mathematics and Statistics, College of Science and Mathematics, Center of Graph Theory, Algebra, and Analysis-PRISM, Mindanao State University-Iligan Institute of Technology, 9200 Iligan City, Philippines Abstract. Let G = (V (G), E(G)) be any simple undirected graph. The open hop neighborhood of v ∈ V (G) is the set N2 G(v) = {u ∈ V (G) : dG(u, v) = 2}. Then G induces a topology τG on V (G) with base consisting of sets of the form F 2 G[A] = V (G)\N2 G[A], where N2 G[A] = A ∪ {v ∈ V (G) : N2 G(v) ∩ A 6= ∅} and A ranges over all subsets of V (G). In this paper, we describe the topologies induced by the complement of a graph, the join, the corona, the composition and the Cartesian product of graphs. 2010 Mathematics Subject Classifications: 05C76 Key Words and Phrases: Join, Corona, Lexicographic product, Cartesian product, Open hop neighborhood 1. Introduction Let G = (V (G), V (H)) be any simple undirected graph. The distance d(u, v) be- tween two vertices u and v in G is the length of a shortest path joining u and v. Let v ∈ V (G). The neighborhood of v is the set N(v) consisting of all u ∈ V (G) which are adjacent with v and the closed neighborhood is N [v] = N(v) ∪ {v}. For any A ⊆ V (G), N(A) = {x : xa ∈ E(G) for some a ∈ A} is called the neighborhood of A and N [A] = N(A) ∪ A is called the closed neighborhood of A. Moreover, for each v ∈ V (G), the open hop neighborhood of v is the set N2 G(v) = {u ∈ V (G) : dG(u, v) = 2} and the closed hop neighborhood of v is the set N2 G[v] = {v} ∪ N2 G(v). Also, for any A ⊆ V (G), N2 G(A) = {v ∈ V (G) : N2 G(v) ∩ A 6= ∅} is called the open hop neighborhood of A and the set N2 G[A] = A ∪N2 G(A) is the called closed hop neighborhood of A. Denote by F 2 G[A] the complement of N2 G[A], i.e., F 2 G[A] = V (G)\N2 G[A]. In 1983, Diesto and Gervacio in [5] proved that given a simple graphG = (V (G), E(G)), G induces a topology on V (G), denoted by τG, with base consisting of sets of the form ∗Corresponding author. DOI: https://doi.org/10.29020/nybg.ejpam.v12i2.3421 Email addresses: caengrace1997@gmail.com (C. Nianga), sergio.canoy@g.msuiit.edu.ph (S. Canoy Jr.) http://www.ejpam.com 499 c© 2019 EJPAM All rights reserved. C. Nianga, S. Canoy / Eur. J. Pure Appl. Math, 12 (2) (2019), 499-505 500 FG(A) = V (G)\NG(A), where NG(A) = A ∪ {x : xa ∈ E for some a ∈ A} and A ranges over all subsets of V (G). Their construction was further investigated in [2], [3] and [6]. In particular, Canoy and Lemence in [2] described the topologies induced by the complement of a graph, the join of graphs, composition and Cartesian product of graphs. In [1], Canoy and Gimeno presented another way of constructing a topology τG from a connected graph G by considering the family Ω(G) = {F 2 G[A] : A ⊆ V (G)} where F 2 G[A] = {x ∈ V (G) : x /∈ A and dG(x, a) 6= 2 for all a ∈ A}. They showed that this family is a base for some topology τG on V (G). This construction is also studied by Nianga et al, in [4] for any graph G. It is also shown that the family BG = {F 2 G[A] : A ⊆ V (G)} and SG = {F 2 G[v] : v ∈ V (G)} are, respectively, base and subbase for the topology τG on V (G). Concepts on Graph Theory and Topology are taken from [7] and [8], respectively. 2. Results Definition 1. The complement of graph G, denoted by G is the graph with V (G) = V (G) and uv ∈ E(G) if and only if uv /∈ E(G), where u, v ∈ V (G) = V (G). Theorem 1. Let G be any graph and G its complement. Then for each v ∈ V (G), F 2 G [v] =  FG[v] ∪  ⋂ u∈FG[v] NG(u)  , if FG[v] 6= ∅ NG(v), if FG[v] = ∅. (1) Proof. Let G be any graph and G its complement. Let v ∈ V (G) and set A = ∩u∈FG[v]NG(u). Suppose FG[v] = ∅. Then NG(v) = V (G)\{v}. Hence, v is an isolated vertex in G. Thus, F 2 G [v] = NG(v). Suppose FG[v] 6= ∅. Let u ∈ FG[v]. Then u 6= v and u /∈ NG(v). Hence, u 6= v and u ∈ NG(v). Thus, u ∈ F 2 G [v]. Next, let w ∈ A. Then w ∈ NG(u) for all u ∈ FG[v]. Since u /∈ NG(v), it follows that w 6= v. Also, w /∈ NG(u) for all u ∈ NG(v). It implies that dG(w, v) 6= 2. Hence, w ∈ F 2 G [v]. Consequently, FG[v] ∪ [∩u∈FG[v]NG(u)] ⊆ F 2 G [v]. Next, let x ∈ F 2 G [v]. Then x 6= v and x /∈ N2 G (v). If x ∈ FG[v], then we are done. Suppose x /∈ FG[v]. Then x ∈ NG(v). Suppose further that there exists u ∈ FG[v] such that x /∈ NG(u). Thus, u ∈ NG(v) and x ∈ NG(u). Also, since x ∈ NG(v), x /∈ NG(v). Thus, dG(x, v) = 2, that is, x ∈ N2 G (v), a contradiction. Therefore, x ∈ NG(u) for all u ∈ FG[v]. This shows that x ∈ A. Accordingly, F 2 G [v] ⊆ FG[v] ∪ [∩u∈FG[v]NG(u)]. This establishes the desired equality. Theorem 2. Let G be any graph and G its complement. If v is an isolated vertex of G (or of G), then {v} ∈ τG ∩ τG. Proof. Suppose v is an isolated vertex of G (or of G). Then {v} = F 2 G[V (G)\{v}] = F 2 G[V (G)\{v}] and so, {v} ∈ BG and {v} ∈ BG. Thus, {v} ∈ τG and {v} ∈ τG. Therefore, {v} ∈ τG ∩ τG. C. Nianga, S. Canoy / Eur. J. Pure Appl. Math, 12 (2) (2019), 499-505 501 Remark 1. The converse of theorem 17 is not true. Consider G = P5 = [a, b, c, d, e]. Then {e} = F 2 G[a, b] and {e} = FG[a, c]. However, e is not an isolated vertex of G nor of G. Definition 2. The join G1 + G2 of graphs G1 and G2 is the graph G with V (G) = V (G1) ∪ V (G2) and E(G) = E(G1) ∪ E(G2) ∪ {uv : u ∈ V (G1) and v ∈ V(G2)}. Theorem 3. Let G = (V (G), E(G)) and H = (V (H), E(H)) be graphs and let ∅ 6= A ⊆ V (G) and ∅ 6= B ⊆ V (H). Then (i) F 2 G+H [A] = V (H) ∪ [∩a∈ANG(a)] ; (ii) F 2 G+H [B] = V (G) ∪ [∩b∈BNH(b)] and (iii) F 2 G[∅] = V (G) ∪ V (H). Proof. Let G = (V (G), E(G)) and H = (V (H), E(H)) be graphs. Let ∅ 6= A ⊆ V (G) and ∅ 6= B ⊆ V (H). (i) Note that N2 G+H [A] = A ∪ {v ∈ V (G+H) : dG+H(v, a) = 2 for some a ∈ A}. Since V (H) ⊆ NG+H(A), N2 G+H [A] = A ∪ {v ∈ V (G) : dG+H(v, a) = 2 for some a ∈ A} = A ∪ {v ∈ V (G) : dG(v, a) 6= 1 for some a ∈ A}. Hence, F 2 G+H [A] = V (H) ∪ [∩a∈ANG(a)] . (ii) Similarly, F 2 G+H [B] = V (G) ∪ [∩b∈BNH(b)] . (iii) Clearly, F 2 G+H [∅] = V (G) ∪ V (H). Remark 2. Let G be any graph and let A1, A2 ⊆ V (G). Then N2 G[A1 ∪A2] = N2 G[A1] ∪N2 G[A2]. Theorem 4. Let G = (V (G), E(G)) and H = (V (H), E(H)) be graphs. Then for any A ⊆ V (G+H) such that A ∩ V (G) = AG 6= ∅ and A ∩ V (H) = AH 6= ∅, F 2 G+H [A] = F 2 G+H [AG] ∩ F 2 G+H [AH ]. C. Nianga, S. Canoy / Eur. J. Pure Appl. Math, 12 (2) (2019), 499-505 502 Proof. Let A ⊆ V (G + H). Suppose A ∩ V (G) = AG 6= ∅ and A ∩ V (H) = AH 6= ∅. Then x ∈ F 2 G+H [A] if and only if x /∈ N2 G+H [A]. By Remark 2, x ∈ F 2 G+H [A] if and only if x ∈ F 2 G+H [AG] ∩ F 2 G+H [AH ]. The next theorem follows from Theorem 3 (i) and (ii). Corollary 1. Let G = (V (G), E(G)) and H = (V (H), E(H)) be graphs. Then for any v ∈ V (G) ∪ V (H), F 2 G+H [v] = { V (H) ∪NG(v), if v ∈ V (G) V (G) ∪NG(v), if v ∈ V (H). (2) Definition 3. The corona G ◦H of graphs G and H is the graph obtained by taking one copy of G and |V (G)| copies H and then forming the sum 〈v〉 + Hv = v + Hv for each v ∈ V (G), where Hv is a copy of H corresponding to the vertex v. Theorem 5. Let G = (V (G), E(G)) and H = (V (H), E(H)) be graphs. Then for any a ∈ V (G ◦H), F 2 G◦H [a] =  F 2 G[a] ∪  ⋃ v∈V (G)\NG(a) V (Hv)  , if a ∈ V (G) NHw(a) ∪ [V (G)\NG(w)] ∪  ⋃ v∈V (G)\{w} V (Hv)  , if a ∈ V (Hw) (3) Proof. Let x ∈ F 2 G◦H [a]. Then x 6= a and x /∈ N2 G◦H(a). Consider the following cases: Case 1. Suppose a ∈ V (G). If x ∈ V (G), then x /∈ N2 G(a) since x /∈ N2 G◦H(a). Hence, x ∈ F 2 G[a]. Suppose x /∈ V (G). Let u ∈ V (G) such that x ∈ V (Hu). If u = a, then x ∈ V (Hu) and u ∈ V (G)\NG(a). Suppose u 6= a. Since x /∈ N2 G(a) and dG◦H(a, y) = 2 for all y ∈ V (Hz) with z ∈ NG(a), it follows that u ∈ V (G)\NG(a). Thus, F 2 G◦H [a] ⊆ F 2 G[a] ∪ [ ∪v∈V (G)\NG(a)V (Hv) ] = X. Now, let w ∈ X. If w ∈ F 2 G[a], then w /∈ N2 G[a]. Hence, w /∈ N2 G◦H [a]. This implies that w ∈ F 2 G◦H [a]. Suppose w ∈ ∪v∈V (G)\NG(a)V (Hv). Then there exists v ∈ V (G)\NG(a) such that w ∈ V (Hv). It follows that w 6= a and dG◦H(w, a) 6= 2. Thus, w ∈ F 2 G◦H [a]. Therefore, F 2 G[a] ∪ [ ∪v∈V (G)\NG(a)V (Hv) ] ⊆ F 2 G◦H [a]. Case 2. Suppose a ∈ V (Hw) for some w ∈ V (G). If x = w, then x ∈ V (G)\NG(w). Suppose x 6= w. If x ∈ V (G), then dG(x,w) 6= 1 because dG◦H(x, a) 6= 2. Hence, x ∈ V (G)\NG(w). Suppose x ∈ V (Hq) for some q ∈ V (G). If q = w, then x ∈ V (Hw). Since x 6= a and a ∈ V (Hw), x ∈ NHw(a) (otherwise, dG◦H(a, x) = 2). Suppose q 6= w. Then x ∈ V (Hq) and q ∈ V (G)\{w}. Thus, z ∈ NHw(a) ∪ [V (G)\NG(w)] ∪ [ ∪v∈V (G)\{w}V (Hv) ] = Y. C. Nianga, S. Canoy / Eur. J. Pure Appl. Math, 12 (2) (2019), 499-505 503 Suppose now that p ∈ Y. If p ∈ NHw(a), then dG◦H(p, a) = dHw(p, a) = 1. Hence, p ∈ F 2 G◦H [a]. If p ∈ V (G)\NG(w), then dG◦H(p, w) = dG(p, w) 6= 1. Hence, p 6= a and dG◦H(a, p) 6= 2. This implies that p ∈ F 2 G◦H [a]. Finally, if p ∈ ∪v∈V (G)\{w}V (Hv), then there exists r ∈ V (G)\{w} such that p ∈ V (Hr). Since dG◦H(a, p) = dG◦H(a,w) + dG◦H(r, w) + dG◦H(r, p) = 2 + dG◦H(r, w) ≥ 3, it follows that p ∈ F 2 G◦H [a]. Therefore, NHw(a) ∪ [V (G)\NG(w)] ∪ [ ∪v∈V (G)\{w}V (Hv) ] ⊆ F 2 G◦H [a]. Accordingly, the desired equality follows. Definition 4. The lexicographic product (composition) of graphs G and H, denoted by G[H], is the graph with V (G[H]) = V (G)×V (H) and (u, v)(u′, v′) ∈ E(G[H]) if and only if either uu′ ∈ E(G) or u = u′ and vv′ ∈ E(H). Theorem 6. Let G = (V (G), E(G)) and H = (V (H), E(H)) be any two graphs and let (v, a) ∈ V (G[H]). Then F 2 G[H][(v, a)] = ( F 2 G[v]× V (H) ) ∪ ( {v} × F 2 H [a] ) . Proof. Note that (x, q) ∈ F 2 G[H][(v, a)] if and only if (x, q) 6= (v, a) and dG[H]((x, q), (v, a)) 6= 2. Consider the following cases: Case 1. Suppose x = v. Then q 6= a. Since dG[H]((v, q), (v, a)) = dH(a, q) 6= 2, q ∈ F 2 H [a], q ∈ F 2 H [a]. Hence, (x, q) ∈ {v} × F 2 H [a]. Case 2. Suppose x 6= v. Then dG(x, v) = dG[H]((x, q), (v, a)) 6= 2. Hence, x ∈ F 2 G[v] and (x, q) ∈ F 2 G[v]× V (H). Therefore, F 2 G[H][(v, a)] ⊆ ( F 2 G[v]× V (H) ) ∪ ( {v} × F 2 H [a] ) . Next, let (w, p) ∈ F 2 G[v] × V (H). Then w ∈ F 2 G[v], that is, w 6= v and dG(w, v) 6= 2. It follows that (w, p) 6= (v, a) and dG[H]((w, p), (v, a)) = dG(w, v) 6= 2. This shows that (w, p) ∈ F 2 G[H][(v, a)]. Hence, F 2 G[v] × V (H) ⊆ FG[H][(v, a)]. Finally, let (z, t) ∈ {v} × F 2 H [a]. Then z = v and t ∈ F 2 H [a]. Hence, t 6= a and dH(a, t) 6= 2. Consequently, (z, t) 6= (v, a) and dG[H]((z, t), (v, a)) = dH(a, t) 6= 2, showing that (z, t) ∈ F 2 G[H][a]. Thus, {v} × F 2 H [a] ⊆ F 2 G[H][(v, a)]. This establishes the desired equality. C. Nianga, S. Canoy / Eur. J. Pure Appl. Math, 12 (2) (2019), 499-505 504 Definition 5. The Cartesian Product of two graphs G1 and G2 denoted by G1�G2 is a graph with V (G1�G2) = V (G1) × V (G2) and two vertices a = (u1, u2) and b = (v1, v2) are adjacent in G1�G2 if and only if either u1 = v1 and u2v2 ∈ E(G2) or u2 = v2 and u1v1 ∈ E(G1). Theorem 7. Let K = G�H = (V (K), E(K)), where G = (V (G), E(G)) and H = (V (H), E(H)). Then for each (v, a) ∈ V (K), F 2 K [(v, a)] = [ F 2 G[v]× {a} ] ∪ [ {v} × F 2 H [a] ] ∪ [FG[v]× V (H)\{a}] ∪ [NG(v)× FG[a]] . Proof. Let (v, a) ∈ V (K) = V (G�H) and (x, q) ∈ F 2 K [(v, a)]. Then (v, a) 6= (x, q) and dK((v, a), (x, q)) 6= 2. Now, consider the following cases: Case 1. Assume that x = v. Then q 6= a and dH(q, a) = dK((x, q), (x, a)) 6= 2 and so, q ∈ F 2 H [a]. Hence, (x, q) ∈ {v} × F 2 H [a]. Case 2. Assume that x 6= v. Subcase 1. Let q = a. Then dG(x, v) = dK((x, q), (v, q)) 6= 2 and thus, x ∈ F 2 G[v]. It follows that (x, q) ∈ F 2 G[v]× {a}. Subcase 2. Let q 6= a. Suppose that x ∈ NG(v). If q ∈ NH(a), then dK((x, q), (v, a)) = dG(x, v) + dH(q, a) = 2, a contradiction. Thus, q ∈ V (H)\NH [a]. Hence, (x, q) ∈ NG(v) × FG[a]. Suppose x /∈ NG(v). Then x ∈ FG[v]. Hence, (x, q) ∈ FG[v]× V (H)\{a}. Therefore, F 2 K [(v, a)] ⊆ [ F 2 G[v]× {a} ] ∪ [ {v} × F 2 H [a] ] ∪ [FG[v]× V (H)\{a}] ∪ [NG(v)× FG[a]] . Next, let (v, p) ∈ {v} × F 2 H [a]. Then p 6= a and dH(a, p) 6= 2. Hence, (v, p) 6= (v, a) and dK((v, p), (v, a)) = dH(a, p) 6= 2, that is, (v, p) ∈ F 2 K [(v, a)]. If (x, a) ∈ F 2 G[v] × {a}, then x 6= v and dG(x, v) 6= 2. Hence, (x, a) 6= (v, a) and dK((v, a), (x, a)) = dG(x, v) 6= 2, that is, (x, a) ∈ F 2 K [(v, a)]. Now, (y, b) ∈ NG(v)× FH [a] implies dG(y, v) = 1 and dH(b, a) ≥ 2. It follows that (y, b) 6= (v, a) and dK((y, b), (v, a)) = dG(y, v) + dH(b, a) ≥ 3. Hence, (y, b) ∈ F 2 K [(v, a)]. Finally, (z, t) ∈ [FG[v]× V (H)\{a}] implies dG(z, v) ≥ 2 and dH(t, a) ≥ 1. This means that (z, t) 6= (v, a) and dK((z, t), (v, a)) = dG(z, v) + dH(t, a) ≥ 3. Thus, (z, t) ∈ F 2 K [(v, a)]. Therefore,[ F 2 G[v]× {a} ] ∪ [ {v} × F 2 H [a] ] ∪ [FG[v]× V (H)\{a}] ∪ [NG(v)× FG[a]] ⊆ F 2 K [(v, a)]. This establishes the desired equality. REFERENCES 505 Acknowledgements This research is funded by the Philippine Department of Science and Technology- Accelerated Science and Technology Human Resource Development Program (DOST- ASTHRDP) and Mindanao State University-Iligan Institute of Technology. References [1] S. Canoy and J. Gimeno. Which Connected Graphs Induce the Indiscrete and the Discrete Topologies? Journal of Research in Science and Engineering, 1:17–19, 2004. [2] S. Canoy and R. Lemence. Another Look at the Topologies Induced by Graphs. Ma- timyas Matematika, 21:1–7, 1998. [3] S. Canoy and R. Lemence. Topologies Induced by Some Special Graphs. Journal of Mathematics, 2:45–50, 1999. [4] S. Canoy and C. G. Nianga. On A Finite Topological Space Induced by Hop Neigh- borhoods of a Graph. Advances and Applications in Discrete Mathematics, submitted. [5] S. Diesto and S. Gervacio. Finite Topological Graphs. Journal of Research and Devel- opment, MSU-IIT, 1:76–81, 1983. [6] R. Guerrero and S. Gervacio. Characterization of Graphs which Induce the Discrete and Indiscrete Topological Spaces. Matimyas Matematika, Special Issue, 1:11–15, 1989. [7] F. Harary. Graph Theory. Addison-Wesley Publishing Company, USA, 1969. [8] S. Lipschutz. General Topology, Schaum’s Outline Series. McGraw Hill International Publishing Co., 1987.