
We introduce the general Albertson irregularity index of a connected graph G and define it as Ap(G)=(∑uv∈E(G)|d(u)−d(v)|p)1p, where p is a positive real number and d(v) is the degree of the vertex v in G. The new index is not only generalization of the well-known Albertson irregularity index and σ-index, but also it is the Minkowski norm of the degree of vertex. We present lower and upper bounds on the general Albertson irregularity index. In addition, we study the extremal value on the general Albertson irregularity index for trees of given order. Finally, we give the calculation formula of the general Albertson index of generalized Bethe trees and Kragujevac trees.
Citation: Zhen Lin, Ting Zhou, Xiaojing Wang, Lianying Miao. The general Albertson irregularity index of graphs[J]. AIMS Mathematics, 2022, 7(1): 25-38. doi: 10.3934/math.2022002
[1] | Sadik Delen, Ismail Naci Cangul . Effect of edge and vertex addition on Albertson and Bell indices. AIMS Mathematics, 2021, 6(1): 925-937. doi: 10.3934/math.2021055 |
[2] | Hicham Saber, Zahid Raza, Abdulaziz M. Alanazi, Adel A. Attiya, Akbar Ali . On trees with a given number of segments and their maximum general $ Z $-type index. AIMS Mathematics, 2025, 10(1): 195-207. doi: 10.3934/math.2025010 |
[3] | Muhammad Kamran Jamil, Muhammad Imran, Aisha Javed, Roslan Hasni . On the first general Zagreb eccentricity index. AIMS Mathematics, 2021, 6(1): 532-542. doi: 10.3934/math.2021032 |
[4] | Tariq A. Alraqad, Igor Ž. Milovanović, Hicham Saber, Akbar Ali, Jaya P. Mazorodze, Adel A. Attiya . Minimum atom-bond sum-connectivity index of trees with a fixed order and/or number of pendent vertices. AIMS Mathematics, 2024, 9(2): 3707-3721. doi: 10.3934/math.2024182 |
[5] | Zhibin Du, Ayu Ameliatul Shahilah Ahmad Jamri, Roslan Hasni, Doost Ali Mojdeh . Maximal first Zagreb index of trees with given Roman domination number. AIMS Mathematics, 2022, 7(7): 11801-11812. doi: 10.3934/math.2022658 |
[6] | Zenan Du, Lihua You, Hechao Liu, Yufei Huang . The Sombor index and coindex of two-trees. AIMS Mathematics, 2023, 8(8): 18982-18994. doi: 10.3934/math.2023967 |
[7] | Shabana Anwar, Muhammad Kamran Jamil, Amal S. Alali, Mehwish Zegham, Aisha Javed . Extremal values of the first reformulated Zagreb index for molecular trees with application to octane isomers. AIMS Mathematics, 2024, 9(1): 289-301. doi: 10.3934/math.2024017 |
[8] | Zhenhua Su, Zikai Tang, Hanyuan Deng . Higher-order Randić index and isomorphism of double starlike trees. AIMS Mathematics, 2023, 8(12): 31186-31197. doi: 10.3934/math.20231596 |
[9] | Chang Liu, Jianping Li . Sharp bounds on the zeroth-order general Randić index of trees in terms of domination number. AIMS Mathematics, 2022, 7(2): 2529-2542. doi: 10.3934/math.2022142 |
[10] | Wei Gao, Zahid Iqbal, Shehnaz Akhter, Muhammad Ishaq, Adnan Aslam . On irregularity descriptors of derived graphs. AIMS Mathematics, 2020, 5(5): 4085-4107. doi: 10.3934/math.2020262 |
We introduce the general Albertson irregularity index of a connected graph G and define it as Ap(G)=(∑uv∈E(G)|d(u)−d(v)|p)1p, where p is a positive real number and d(v) is the degree of the vertex v in G. The new index is not only generalization of the well-known Albertson irregularity index and σ-index, but also it is the Minkowski norm of the degree of vertex. We present lower and upper bounds on the general Albertson irregularity index. In addition, we study the extremal value on the general Albertson irregularity index for trees of given order. Finally, we give the calculation formula of the general Albertson index of generalized Bethe trees and Kragujevac trees.
Let G be a simple undirected connected graph with the vertex set V(G) and the edge set E(G). For v∈V(G), N(v) denotes the set of all neighbors of v, and d(v)=|N(v)| denotes the degree of vertex v in G. The minimum and the maximum degree of G are denoted by δ(G) and Δ(G), or simply δ and Δ, respectively. A pendant vertex of G is a vertex of degree one. A graph G is called (Δ,δ)-semiregular if {d(u),d(v)}={Δ,δ} holds for all edges uv∈E(G). Denote by Pn and K1,n−1 the path and the star with n vertices, respectively.
In 1997, the Albertson irregularity index of a connected graph G, introduced by Albertson [1], is defined as
Alb(G)=∑uv∈E(G)|d(u)−d(v)|. |
This index has been of interest to mathematicians, chemists and scientists from related fields due to the fact that the Albertson irregularity index plays a major role in irregularity measures of graphs [3,4,7,8,17], predicting the biological activities and properties of chemical compounds in the QSAR/QSPR modeling [12,24] and the quantitative characterization of network heterogeneity [9]. By the natural extension of the Albertson irregularity index, Gutman et al., [13] recently proposed the σ-index as follows:
σ(G)=∑uv∈E(G)(d(u)−d(v))2=F−2M2, |
where F and M2 are well-known the forgotten topological index and the second Zagreb index of a graph G, respectively. Recently, the σ-index of a connected graph G is studied, such as the characterization of extremal graphs [5] and mathematical relations between the σ-index and other graph irregularity indices [21].
The generalization of topological index is a trend of mathematical chemistry in recent years. Many classical topological indices are generalized, such as the general Randić index [6], the first general Zagreb index [18], the general sum-connectivity index [31], the general eccentric connectivity index [28], etc. Motivated by this fact, we propose the general Albertson irregularity index of a graph G as follows:
Ap(G)=(∑uv∈E(G)|d(u)−d(v)|p)1p, |
where p is a positive real number. Evidently, A1(G)=Alb(G) and A22(G)=σ(G). The other motivation is that the topological index formed from distance function of the degree of vertex has attracted extensive attention of scholars. In 2021, Gutman [10] proposed the Sombor index of a graph G and defined it as SO(G)=∑uv∈E(G)√d2(u)+d2(v), which is the Euclidean norm of d(u) and d(v). According to Gutman [11], it is imaginable to use other distance function to study properties of graphs. Based on this, it is not difficult to find that Ap(G) is the Minkowski norm of d(u) and d(v), which is unification of absolute distance, Euclidean distance and Chebyshev distance. Hence Ap(G)=Δ−δ as p becomes infinite. In particular, Ap(G) is the lp-norm of d(u) and d(v) for p≥1.
We will first recall some useful notions and lemmas used further in Section 2. In Section 3, upper and lower bounds on the general Albertson irregularity index of graphs are given, and the extremal graphs are characterized. In Section 4, the first two trees with minimum general Albertson irregularity index are determined in all trees of fixed order. In Section 5, the general Albertson index of the well-known generalized Bethe trees and Kragujevac trees is obtained.
Let u∨G be the graph by adding all edges between the vertex u and V(G), see for example in Figure 1. The first general Zagreb index of a graph G is defined as Zp(G)=∑v∈V(G)dp(v) for any real number p. The distance between two vertices u,v∈V(G), denoted by d(u,v), is defined as the length of a shortest path between u and v. The eccentricity of v, ε(v), is the distance between v and any vertex which is furthest from v in G. The line graph L(G) is the graph whose vertex set are the edges in G, where two vertices are adjacent if the corresponding edges in G have a common vertex. Let Tn be the set of trees with n vertices. A spider is a tree with at most one vertex of degree more than two.
Lemma 2.1. (Power mean inequality) Let x1,x2,…,xn be positive real numbers and p, q real numbers such that p>q. Then,
(1nn∑i=1xpi)1p≥(1nn∑i=1xqi)1q, |
with equality if and only if x1=x2=⋯=xn.
Lemma 2.2. (Hölder inequality) Let (a1,a2,…,an) and (b1,b2,…,bn) be two n-tuples of real numbers and let p, q be two positive real numbers such that 1p+1q=1. Then
|n∑i=1aibi|≤(n∑i=1|ai|p)1p(n∑i=1|bi|q)1q, |
with equality if and only if |ai|p=λ|bi|q for some real constant λ, 1≤i≤n.
Lemma 2.3. ([26]) If p≥1 is an integer and 0≤x1,x2,…,xn≤n−1, then
(n∑i=1xpi)1p≤(n−1)1−1pn∑i=1x1pi. |
Lemma 2.4. (Minkowski inequality) Let (a1,a2,…,an) and (b1,b2,…,bn) be two n-tuples of real numbers. If p≥1. Then
(n∑i=1|ai+bi|p)1p≤(n∑i=1|ai|p)1p+(n∑i=1|bi|p)1p, | (2.1) |
with equality if and only if ai=λbi for some real constant λ, 1≤i≤n. For 0<p<1, the inequality (2.1) gets reversed.
Lemma 2.5. ([20]) Let x=(x1,x2,…,xk,…) be a non-zero vector. Then for p≥2,
||x||p≤||x||2, |
with equality if and only if all but one of the xi are equal to 0.
Lemma 2.6. Let G be a connected graph. Suppose there exists a vertex u∈V(G) with d(u)≥3, v1,v2,…,vl and w1,w2,…,wt are two path components in G−u, where N(u)={v1,w1,u1,u2,…,ud(u)−2}. Let G′=G−uw1+vlw1. Then
(i) If p>0 and d(u)>d(ui), then Ap(G)>Ap(G′).
(ii) If p≥1 and d(u)≥d(ui), then Ap(G)>Ap(G′).
(iii) If p>0 and Δ=d(u)=3, then Ap(G)>Ap(G′).
Proof. Let s=d(u)−2. Since d(u)>d(ui), d(u)≥3, d(v1)≤2 and d(w1)≤2, we have |d(u)−d(ui)|p−|d(u)−d(ui)−1|p>0 and |d(u)−d(v1)|p−|d(u)−d(v1)−1|p+1>0. Then
App(G)−App(G′)=∑uv∈E(G)|d(u)−d(v)|p−∑u′v′∈E(G′)|d(u′)−d(v′)|p=s∑i=1(|d(u)−d(ui)|p−|d(u)−d(ui)−1|p)+|d(u)−d(w1)|p+|d(u)−d(v1)|p−|d(u)−d(v1)−1|p+1>0. |
Thus we have Ap(G)>Ap(G′).
If p≥1 and d(u)=d(ui), then we have
App(G)−App(G′)=s∑i=1(|d(u)−d(ui)|p−|d(u)−d(ui)−1|p)+|d(u)−d(w1)|p+|d(u)−d(v1)|p−|d(u)−d(v1)−1|p+1=−(d(u)−2)+|d(u)−d(w1)|p+|d(u)−d(v1)|p−|d(u)−d(v1)−1|p+1>0. |
Thus we have Ap(G)>Ap(G′).
If p>0 and Δ=d(u)=3, then we have
App(G)−App(G′)=|d(u)−d(u1)|p−|d(u)−d(u1)−1|p+|d(u)−d(w1)|p+|d(u)−d(v1)|p−|d(u)−d(v1)−1|p+1≥−1+|3−d(w1)|p+|3−d(v1)|p−|3−d(v1)−1|p+1=|3−d(w1)|p+|3−d(v1)|p−|2−d(v1)|p>0. |
Thus we have Ap(G)>Ap(G′).
Combining the above arguments, we have the proof.
Theorem 3.1. Let G be a connected graph with m edges. If p>q, then
Ap(G)≥m1p−1qAq(G) |
with equality if and only if G is a regular graph (when G is non-bipartite) or G is a (Δ,δ)-semiregular bipartite graph (when G is bipartite).
Proof. By Lemma 2.1, we have
(1m∑uv∈E(G)|d(u)−d(v)|p)1p≥(1m∑uv∈E(G)|d(u)−d(v)|q)1q, |
that is,
1m1pAp(G)≥1m1qAq(G), |
that is,
Ap(G)≥m1p−1qAq |
with equality if and only if |d(u)−d(v)| is a constant for every edge uv in G, that is, G is a regular graph (when G is non-bipartite) or G is a (Δ,δ)-semiregular bipartite graph (when G is bipartite).
Corollary 3.2. Let G be a connected graph with m edges. Then
Alb(G)≤√m(F−2M2), |
with equality if and only if G is a regular graph (when G is non-bipartite) or G is a (Δ,δ)-semiregular bipartite graph (when G is bipartite).
Theorem 3.3. Let G be a connected graph. If 1p+1q=1, then
Ap(G)Aq(G)≥F−2M2, |
with equality if and only if p=2, or G is a regular graph (when G is non-bipartite) or G is a (Δ,δ)-semiregular bipartite graph (when G is bipartite).
Proof. For ai=bi=d(u)−d(v) and apply Lemma 2.2. Then
∑uv∈E(G)(d(u)−d(v))2≤(∑uv∈E(G)|d(u)−d(v)|p)1p(∑uv∈E(G)|d(u)−d(v)|q)1q, |
that is,
F−2M2≤Ap(G)Aq(G), |
with equality if and only if p=2, or G is a regular graph (when G is non-bipartite) or G is a (Δ,δ)-semiregular bipartite graph (when G is bipartite).
Theorem 3.4. Let G be a connected graph with m edges. If p≥1 is an integer, then
Ap(G)≤(m−1)1−1p(A1p(G))1p. |
Proof. Let xi=|d(u)−d(v)| in Lemma 2.3. Then
(∑uv∈E(G)|d(u)−d(v)|p)1p≤(m−1)1−1p∑uv∈E(G)|d(u)−d(v)|1p, |
that is,
Ap(G)≤(m−1)1−1p(A1p(G))1p. |
Theorem 3.5. Let G be a connected graph with m edges. If p≥1, then
Ap(G)≥(m+pAlb(G))1p−m1p. |
If 0<p<1, then
Ap(G)≤(m+pAlb(G))1p−m1p. |
Proof. Let ai=1 and bi=|d(u)−d(v)| in Lemma 2.4. Then
(∑uv∈E(G)(1+|d(u)−d(v)|)p)1p≤Ap(G)+m1p |
for p≥1. By Bernoulli inequality, we have
Ap(G)+m1p≥(∑uv∈E(G)(1+p|d(u)−d(v)|))1p=(m+pAlb(G))1p, |
that is,
Ap(G)≥(m+pAlb(G))1p−m1p. |
For 0<p<1, by Lemma 2.4 and Bernoulli inequality, we have the proof.
Theorem 3.6. Let G be a connected graph. If p≥2, then
Ap(G)≤√F−2M2 | (3.1) |
with equality if and only if d(u)−d(v)≠0 for unique edge uv and 0 for the other edges in G.
Proof. By Lemma 2.5, we have
Ap(G)≤A2(G)=(∑uv∈E(G)|d(u)−d(v)|2)12=√σ(G)=√F−2M2 |
with equality if and only if d(u)−d(v)≠0 for only one edge uv and 0 for the other edges in G.
Remark 3.7. There exist many graphs such that the equality in (3.1) holds, the following graphs are examples (See Figure 2).
Theorem 3.8. Let G be a connected graph with n vertices. Then
Ap(G)≤[Zp(L(G))]1p |
with equality if and only if G≅K1, n−1.
Proof. By definition of Ap(G), we have
Ap(G)=(∑uv∈E(G)|d(u)−d(v)|p)1p=(∑uv∈E(G)|(d(u)−1)−(d(v)−1)|p)1p≤[∑uv∈E(G)(d(u)+d(v)−2)p]1p=[∑v∈V(L(G))(d(v))p]1p=[Zp(L(G))]1p |
with equality if and only if (d(u)−1)(d(v)−1)≤0, that is, d(v)=1 for every edge uv in G, that is, G is a star K1, n−1.
Corollary 3.9. Let G be a connected graph with n vertices and m edges. Then
Alb(G)≤Z1(L(G))=Z2(G)−2mandσ(G)≤Z2(L(G)) |
with equality if and only if G≅K1, n−1.
Theorem 3.10. Let u be a vertex and G be a connected graph. Then
Ap(u∨G)=[Zp(¯G)+App(G)]1p, |
where ¯G is the complement of G.
Proof. By definition of u∨G, we have
Ap(u∨G)=(∑wv∈E(u∨G)|d(w)−d(v)|p)1p=(∑v∈V(G)|n−d(v)−1|p+∑wv∈E(G)|d(w)−d(v)|p)1p=[∑v∈V(¯G)dp(v)+App(G)]1p=[Zp(¯G)+App(G)]1p. |
Corollary 3.11. Let G be a connected graph with n vertices and m edges. Then
Alb(u∨G)−Alb(G)=n(n−1)−2m,σ(u∨G)−σ(G)+Z2(G)=n(n−1)2−4m(n−1). |
Theorem 3.12. Let u be a pendant vertex of a connected graph G with n≥3 vertices. If G+Pt (t≥1) is the graph by adding a new (pendant) path to u, then
(i) Ap(G+Pt)>Ap(G) for 0<p<1.
(ii) Ap(G+Pt)=Ap(G) for p=1.
(iii) Ap(G+Pt)<Ap(G) for p>1.
Proof. Let v be the unique neighbour of u in G. Since ap+bp>(a+b)p for a>0, b>0 and 0<p<1, we have
App(G+Pt)=∑rs∈E(G+e)|d(r)−d(s)|p=(d(v)−2)p+(2−1)p+∑rs∈E(G),r,s≠u|d(r)−d(s)|p>(d(v)−1)p+∑rs∈E(G),r,s≠u|d(r)−d(s)|p=App(G), |
for 0<p<1. Thus Ap(G+Pt)>Ap(G). By a similar reasoning as above, we have the proof of (ii) and (iii).
Corollary 3.13. Let u be a pendant vertex of a connected graph G with n≥3 vertices. If G+Pt (t≥1) is the graph by adding a new (pendant) path to u, then
σ(G+Pt)<σ(G). |
Theorem 4.1. Let Tn∈Tn. Then
21p≤Ap(Tn)≤(n−2)(n−1)1p. |
The lower bound is attained if and only if Tn≅Pn. The upper bound is attained if and only if Tn≅K1,n−1.
Proof. If Δ≥3, then Tn has at least three pendant vertices. Thus Ap(Tn)>31p>21p=Ap(Pn). In addition, Ap(Tn)≤(Δ−1)(n−1)1p≤(n−2)(n−1)1p=Ap(K1,n−1).
Theorem 4.2. Let n≥10, Tn∈Tn−{Pn}, T1 and T3 shown as in Figure 3.
(i) If p>1, then Ap(Tn)≥61p with equality if and only if Tn≅T1.
(ii) If p=1, then Ap(Tn)≥6 with equality if and only if Tn is a spider with Δ=3.
(iii) If 0<p<1, then Ap(Tn)≥(2p+1+2)1p with equality if and only if Tn≅T3.
Proof. Let d1≥d2≥d3≥⋯≥dn be the degree sequence of Tn, and let k be the number of non-pendant edges uv with d(u)≠d(v). Then
App(Tn)=∑uv∈E(G)|d(u)−d(v)|p≥d1+(d2−2)+(d3−2)+k. |
If d1≥7, then App(Tn)>7.
If d1≥6 and d2≥3, then App(Tn)>6+(3−2)=7.
If d1≥6 and d2=2, then App(Tn)>6+1=7.
If d1=d2=5, then App(Tn)>5+(5−2)=8.
If d1=5 and d2=4, then App(Tn)>5+(4−2)=7.
If d1=5 and d2=3, then App(Tn)>5+(3−2)+1=7.
If d1=5 and d2=2, then App(Tn)={4p+1+3p+1,3⋅4p+2⋅3p+2,2⋅4p+3p+1+3,4p+4⋅3p+4,5⋅3p+5. Thus App(Tn)>6.
If d1=d2=4, then App(Tn)≥4+(4−2)+1>7.
If d1=4 and d2=d3=d4=3, then App(Tn)>4+(3−2)+(3−2)+(3−2)=7.
If d1=4 and d2=d3=3, then App(Tn)>4+(3−2)+(3−2)+1=7.
If d1=4 and d2=3, then App(Tn)>4+(3−2)+1=6.
If d1=4 and d2=2, then App(Tn)={3p+1+2p+1,2⋅3p+2p+1+2,3p+3⋅2p+3,4⋅2p+4.
If d1=3, we can applying Lemma 2.6 repeatedly to the vertices with degree three. Thus the minimum value of Tn has four cases, shown as in Figure 3. By direct computing, we have
App(T1)=2p+1+2,App(T2)=App(T′2)=2p+4,App(T3)=6. |
By comparing the above cases, we have that T1, T2, T′2 and T3 are the candidates with minimum general Albertson index among Tn−{Pn}. Further, we have App(T1)>App(T2)=App(T′2)>App(T3) for p>1, App(T1)=App(T2)=App(T′2)=App(T3) for p=1, and App(T1)<App(T2)=App(T′2)<App(T3) for 0<p<1.
Theorem 4.3. Let Tn be a tree with n vertices. If p≥1, then
Ap(Tn)≥(Δε1−p(vΔ))1p(Δ−1), |
where vΔ is a vertex of the maximum degree.
Proof. Let vΔv1v2…vl1 be the path from the vertex vΔ to pendant vertex vl1. Then d(vΔ,vl1)=l1. Let f(x)=xp. Since f(x) is a increasing and convex function for x>0 and p≥1, we have
1l1(f(|d(vΔ)−d(v1)|)+f(|d(v1)−d(v2)|)+⋯+f(|vl1−1−d(vl1)|))≥f(|d(vΔ)−d(v1)|+|d(v1)−d(v2)|+⋯+|vl1−1−d(vl1)|l1)≥f((d(vΔ)−d(v1)+d(v1)−d(v2)+⋯+vl1−1−d(vl1))l1)=f(Δ−1l1), |
that is,
|d(vΔ)−d(v1)|p+⋯+|vl1−1−d(vl1)|p≥l1−p1(Δ−1)p. |
Since Tn has at least Δ pendant vertices, we have
Ap(Tn)≥(l1−p1+l1−p2+⋯+l1−pΔ)1p(Δ−1), |
where l1,l2,…,lΔ is the distance from maximum degree vertex vΔ to pendant vertex vli, 1≤i≤Δ. Note that ε(vΔ)=maxv∈V(G)d(vΔ,v)≥li for 1≤i≤Δ. Thus we have Ap(Tn)≥(Δε1−p(vΔ))1p(Δ−1).
Corollary 4.4. Let Tn be a tree with n vertices. Then
Alb(Tn)≥Δ(Δ−1) |
with equality if and only if G is a spider.
In this section, we give the calculation formula of the general Albertson index of generalized Bethe trees and Kragujevac trees which are a wide range of applications in the field of mathematics [22,25], cheminformatics [15,27,29], statistical mechanics [19], etc.
A generalized Bethe tree [23] is a rooted tree in which vertices of the same level (height) have the same degree. We usually use Bk to denote the generalized Bethe tree with k levels with the root at the level 1. More specifically, Bk,d denotes a Bethe tree [16] of k levels with the root degree d, and the vertices between the level 2 and k−1 all have degree d+1. A regular dendrimer tree [14] Tk,d is a special case of Bk, where the degrees of all internal vertices are d.
Theorem 5.1. Let Bk be the generalized Bethe tree where the degree of each level is d1≥d2≥⋯≥dk−1,dk=1. Then
Ap(Bk)=d1p1(|d1−d2|p+k∑i=2|di−di−1|pi∏j=2(dj−1))1p. |
Proof. By definition of the generalized Bethe tree, we have
App(Bk)=∑uv∈E(G)|d(u)−d(v)|p=d1[|d1−d2|p+|d2−d3|p(d2−1)+|d3−d4|p(d3−1)(d2−1)+⋯+|dk−1−dk|p(dk−1−1)⋯(d2−1)]=d1(|d1−d2|p+k∑i=2|di−di−1|pi∏j=2(dj−1)). |
Thus we have the proof.
Corollary 5.2. Let Bk,d and Tk,d be the Bethe tree and a regular dendrimer tree, respectively. Then
Ap(Bk,d)=(d+dp+k−1)1pandAp(Tk,d)=[d(d−1)p+k−2]1p. |
Let P3 be the 3-vertex tree, rooted at one of its terminal vertices, see Figure 4. For k=2,3,…, construct the rooted tree Rk by identifying the roots of k copies of P3. The vertex obtained by identifying the roots of P3-trees is the root of Rk. Let d≥2 be an integer and γ1,γ2,…,γd be rooted trees, i.e., γ1,γ2,…,γd∈{R2,R3,…}. A Kragujevac tree KT [15] is a tree possessing a vertex of degree d, adjacent to the roots of γ1,γ2,…,γd. This vertex is said to be the central vertex of KT, whereas d is the degree of KT. The subgraphs γ1,γ2,…,γd are the branches of KT. Recall that some (or all) branches of KT may be mutually isomorphic.
Theorem 5.3. Let KT be a Kragujevac tree with n vertices and γi≅Rki, i=1,2,…,d. Then
Ap(KT)=[n−d−12+d∑i=1(ki(ki−1)p+|ki−d+1|p)]1p. |
Proof. Since 1+d∑i=1(2ki+1)=n, by definition of the Kragujevac tree, we have
App(KT)=∑uv∈E(G)|d(u)−d(v)|p=d∑i=1[ki+ki(ki+1−2)p+|d−(ki+1)p|]=n−d−12+d∑i=1(ki(ki−1)p+|ki−d+1|p). |
Thus we have the proof.
Corollary 5.4. Let KT be a Kragujevac tree with n vertices and γi≅Rk, i=1,2,…,d. Then
Ap(KT)=[n−d−12+dk(k−1)p+d|k−d+1|p]1p. |
In this paper, we propose the general Albertson irregularity index which extends classical Albertson irregularity index and σ-index. The tight bounds of the general Albertson irregularity index are established. Additionally, the general Albertson irregularity index of trees are studied. In 2014, the total irregularity of a graph G, introduced by Abdo, Brandt and Dimitrov [2], is defined as irrt(G)=∑{u,v}⊆V(G)|d(u)−d(v)|. For measuring the non-self-centrality of a graph, the non-self-centrality number of G was introduced in [30] as N(G)=∑{u,v}⊆V(G)|ε(u)−ε(v)|. Based on these, we can propose the general total irregularity and the general non-self-centrality number of a graph G as follows:
irrp(G)=(∑{u,v}⊆V(G)|d(u)−d(v)|p)1pandNp(G)=(∑{u,v}⊆V(G)|ε(u)−ε(v)|p)1p, |
where the summation goes over all the unordered pairs of vertices in G. The research interaction among Ap(G), irrp(G) and Np(G) will be carried out in the near future.
The authors are grateful to the anonymous referee for careful reading and valuable comments which result in an improvement of the original manuscript. This work was supported by the Qinghai science and technology plan project (No. 2021-ZJ-703) and the National Natural Science Foundation of China (No. 11771443).
The authors declare no conflict of interest.
[1] | M. O. Albertson, The irregularity of a graph, Ars Combin., 46 (1997), 219–225. |
[2] | H. Abdo, S. Brandt, D. Dimitrov, The total irregularity of a graph, DMTCS, 16 (2014), 201–206. |
[3] |
H. Abdoa, N. Cohenb, D. Dimitrov, Graphs with maximal irregularity, Filomat, 28 (2014), 1315–1322. doi: 10.2298/FIL1407315A. doi: 10.2298/FIL1407315A
![]() |
[4] |
H. Abdo, D. Dimitrov, The irregularity of graphs under graph operations, Discuss. Math. Graph T., 34 (2014), 263–278. doi: 10.7151/dmgt.1733. doi: 10.7151/dmgt.1733
![]() |
[5] |
H. Abdo, D. Dimitrov, I. Gutman, Graphs with maximal σ irregularity, Discrete Appl. Math., 250 (2018), 57–64. doi: 10.1016/j.dam.2018.05.013. doi: 10.1016/j.dam.2018.05.013
![]() |
[6] | B. Bollobás, P. Erdős, Graphs of extremal weights, Ars Combin., 50 (1998), 225–233. |
[7] |
X. D. Chen, Y. P. Hou, F. G. Lin, Some new spectral bounds for graph irregularity, Appl. Math. Comput., 320 (2018), 331–340. doi: 10.1016/j.amc.2017.09.038. doi: 10.1016/j.amc.2017.09.038
![]() |
[8] | D. Dimitrov, T. Réti, Graphs with equal irregularity indices, Acta Polytech. Hung., 11 (2014), 41–57. |
[9] |
E. Estrada, Quantifying network heterogeneity, Phys. Rev. E Stat. Nonlin. Soft Matter Phys., 82 (2010), 066102. doi: 10.1103/PhysRevE.82.066102. doi: 10.1103/PhysRevE.82.066102
![]() |
[10] | I. Gutman, Geometric approach to degree-based topological indices: Sombor indices, MATCH Commun. Math. Comput. Chem., 86 (2021), 11–16. |
[11] |
I. Gutman, Some basic properties of Sombor indices, Open J. Discret. Appl. Math., 4 (2021), 1–3. doi: 10.30538/psrp-odam2021.0047. doi: 10.30538/psrp-odam2021.0047
![]() |
[12] |
I. Gutman, P. Hansen, H. Mélot, Variable neighborhood search for extremal graphs. 10. comparison of irregularity indices for chemical trees, J. Chem. Inf. Model., 45 (2005), 222–230. doi: 10.1021/ci0342775. doi: 10.1021/ci0342775
![]() |
[13] | I. Gutman, M. Togan, A. Yurttas, A. S. Cevik, I. N. Cangul, Inverse problem for sigma index, MATCH Commun. Math. Comput. Chem., 79 (2018), 491–508. |
[14] | I. Gutman, Y.N. Yeh, S.L. Lee, J.C. Chen, Wiener numbers of dendrimers, MATCH Commun. Math. Comput. Chem., 30 (1994), 103–115. |
[15] | S. A. Hosseini, M. B. Ahmadi, I. Gutman, Kragujevac trees with minimal atom-bond connectivity index, MATCH Commun. Math. Comput. Chem., 71 (2014), 5–20. |
[16] |
O. J. Heilmann, E. H. Lieb, Theory of monomer-dimer systems, Commun. Math. Phys., 25 (1972), 190–232. doi: 10.1007/BF01877590. doi: 10.1007/BF01877590
![]() |
[17] |
M. A. Henninga, D. Rautenbach, On the irregularity of bipartite graphs, Discrete Math., 307 (2007), 1467–1472. doi: 10.1016/j.disc.2006.09.038. doi: 10.1016/j.disc.2006.09.038
![]() |
[18] | X. L. Li, J. Zheng, A unifled approach to the extremal trees for difierent indices, MATCH Commun. Math. Comput. Chem., 54 (2005), 195–208. |
[19] |
M. Ostilli, Cayley trees and Bethe lattices: A concise analysis for mathematicians and physicists, Physica A, 391 (2012), 3417–3423. doi: 10.1016/j.physa.2012.01.038. doi: 10.1016/j.physa.2012.01.038
![]() |
[20] |
I. Rivin, Counting cycles and finite dimensional Lp norms, Adv. Appl. Math., 29 (2002), 647–662. doi: 10.1016/S0196-8858(02)00037-4. doi: 10.1016/S0196-8858(02)00037-4
![]() |
[21] |
T. Réti, On some properties of graph irregularity indices with a particular regard to the σ-index, Appl. Math. Comput., 344–345 (2019), 107–115. doi: 10.1016/j.amc.2018.10.010. doi: 10.1016/j.amc.2018.10.010
![]() |
[22] |
O. Rojo, R. D. J. Alarcón, Line graph of combinations of generalized Bethe trees: Eigenvalues and energy, Linear Algebra Appl., 435 (2011), 2402–2419. doi: 10.1016/j.laa.2010.10.008. doi: 10.1016/j.laa.2010.10.008
![]() |
[23] |
O. Rojo, M. Robbiano, An explicit formula for eigenvalues of Bethe trees and upper bounds on the largest eigenvalue of any tree, Linear Algebra Appl., 427 (2007), 138–150. doi: /10.1016/j.laa.2007.06.024. doi: 10.1016/j.laa.2007.06.024
![]() |
[24] | T. Réti, R. Sharafdini, H. Haghbin, Á. Drégelyi-Kiss, Graph irregularity indices used as molecular descriptors in QSPR studies, MATCH Commun. Math. Comput. Chem., 79 (2018), 509–524. |
[25] |
M. Robbianoa, V. Trevisan, Applications of recurrence relations for the characteristic polynomials of Bethe trees, Comput. Math. Appl., 59 (2010), 3039–3044. doi: 10.1016/j.camwa.2010.02.023. doi: 10.1016/j.camwa.2010.02.023
![]() |
[26] |
L. A. Székely, L. H. Clark, R. C. Entringer, An inequality for degree sequences, Discrete Math., 103 (1992), 293–300. doi: 10.1016/0012-365X(92)90321-6. doi: 10.1016/0012-365X(92)90321-6
![]() |
[27] |
M. K. Siddiqui, M. Imran, M. A. Iqbal, Molecular descriptors of discrete dynamical system in fractal and Cayley tree type dendrimers, J. Appl. Math. Comput., 61 (2019), 57–72. doi: 10.1007/s12190-019-01238-1. doi: 10.1007/s12190-019-01238-1
![]() |
[28] |
T. Vetrík, M. Masre, General eccentric connectivity index of trees and unicyclic graphs, Discrete Appl. Math., 284 (2020), 301–315. doi: 10.1016/j.dam.2020.03.051. doi: 10.1016/j.dam.2020.03.051
![]() |
[29] | Y. Wu, F. Y. Wei, B. L. Liu, Z. Jia, The generalized (terminal) Wiener polarity index of generalized Bethe trees and coalescence of rooted trees, MATCH Commun. Math. Comput. Chem., 70 (2013), 603–620. |
[30] |
K. X. Xu, K. C. Das, A. D. Maden, On a novel eccentricity-based invariant of a graph, Acta Math. Sin., 32 (2016), 1477–1493. doi: 10.1007/s10114-016-5518-z. doi: 10.1007/s10114-016-5518-z
![]() |
[31] |
B. Zhou, N. Trinajstić, On general sum-connectivity index, J. Math. Chem., 47 (2010), 210–218. doi: 10.1007/s10910-009-9542-4. doi: 10.1007/s10910-009-9542-4
![]() |
1. | Zainab Alsheekhhussain, Tamás Réti, Akbar Ali, M. T. Rahim, Weighted Graph Irregularity Indices Defined on the Vertex Set of a Graph, 2022, 2022, 2314-4785, 1, 10.1155/2022/7834080 | |
2. | Akbar Ali, Abeer M. Albalahi, Abdulaziz M. Alanazi, Akhlaq Ahmad Bhatti, Amjad E. Hamza, On the maximum sigma index of k-cyclic graphs, 2023, 325, 0166218X, 58, 10.1016/j.dam.2022.10.009 |