A generalization of a Tur´an’s theorem about maximum clique on graphs

One of the most important Tur ´ an’s theorems establishes an inequality between the maximum clique and the number of edges of a graph. Since 1941, this result has received much attention and many of the different proofs involve induction and a probability distribution. In this paper we detail finite procedures that gives a proof for the Tur ´ an’s Theorem. Among other things, we give a generalization of this result. Also we apply this results to a Nikiforov’s inequality between the spectral radius and the maximum clique of a graph.


Introduction
Many works in graph theory deals with upper bounds for the number of edges in a graph. Some examples are [5], [9] and [6]. In this scope, the famous Turán's Theorem is one of the most important results in graph theory about cliques and the number of edges in a graph and is stated as follows . (1) This result gives the maximum number of edges that a simple graph of n vertices can have if it doesn't contain a clique of a certain size [10]. Also this result provides what is known as extremal problems in graph theory [3]. We recommend [1] for some proofs.
In [1] there are four proofs of Theorem 1.1. The third proof present a sketch of procedures to prove the theorem and was based on ideas in [2,7] and [11]. In [7] the proofs are based on finite induction and in [7,11] there are generalizations of Turán's theorem. Theorem 1.1 was utilized by Nikiforov in [8] to relate the spectral radius and the maximum clique of a graph. Theorem 1.2. (Nikiforov) Let G be a simple graph with n vertices having the spectral radius λ and cl(G) the cardinality of a maximum clique. Then .
In this paper we detail the procedures described in [1] to obtain better inequalities involving not only the clique number but also the amount of some cliques in the graph. As consequence we apply these inequalities to improve Nikiforov's Theorem 1.2.
The rest of the paper is organized as follows: In section 3 we present the detailed procedures developed. These are described in the proofs of Theorems 3.1 and 3.2. In section 4 we develop our improved inequalities.

Notations
In this paper we will denote by G = (V (G), E(G), ψ G ) a finite unoriented graph with V (G) = V the set of vertices, E(G) = E the set of edges and ψ G the incidence function. If there is no confusion, we will simply make mention to the graph G. The vertices will be denoted by v (or v i ) and the edges by e = v i v j . N (v) will be the set containing all neighboring vertices of the vertex v. If v is included in the set, we have a star N (v) = N (v) ∪ {v}. A particular k-clique in G sometimes will be denoted Cl k (G). The cardinality of a maximum clique will by cl(G). Let A, B be non-empty subsets of V . A and B are disconnected if there is no a ∈ A and b ∈ B such that ab ∈ E. We will denote by |X| the cardinality of set X, ⟨a, b⟩ the canonical scalar product of a and b in R n and ||x|| the canonical norm of vector x.
We define the set Let G be a finite graph. All proofs of the following results were based on maximizing the function f G : D → R such that www.ejgta.org A generalization of a Turán theorem about maximum clique | Douglas F. G. Santiago et al.
Each w i represents the weight of the vertex v i and w = (w 1 , . . . , w n ) is the (discrete) probability distribution over V (G). If ψ G (e k ) = v i v j we say that w i w j is the weight of the edge e k . In this way, f G (w) gives the sum of all the weights of the edges of G. For the rest of the article, we will only consider graphs with at least one edge. The weight of a non-empty set A ⊂ V (G) is the sum of all weights of the vertices v ∈ A. Furthermore, denote by s r the sum of the weights of N (v r ) for r ∈ {1, . . . , n}.

Procedures
In this section let w = (w 1 , . . . , w n ) be a probability distribution over V (G). We present two Theorems indicating finite procedures on the weights of a distribution w that forms the base for the proof of our main results.
Proof. Note that f G (w ′ ) is the same as f G (w) plus w j multiplied by all weights related to the adjacencies of v i (since w ′ i = w i + w j ), and plus −w j multiplied by all weights related to the adjacencies of v j (since w ′ j = 0), then: As s i ≥ s j then w i (s i − s j ) ≥ 0 and we conclude that Theorem 3.1. Let G be a simple graph with n vertices. For all probability distribution w on V (G), there is a k-clique, say Cl k (G), and a probability distribution Proof. The proof is based on a finite procedure. If G is complete, we will consider that there is nothing to do and the theorem is proved. Suppose then that G is a non-complete graph. Take Reorder the vertices of G such that s 1 ≥ s 2 ≥ · · · ≥ s n . Since G is not complete, let v i 1 be the first vertex in the chosen order that does not have all the vertices of G connected to it. For each vertex disconnected with v i 1 create a new distribution w ′ as explained in Lemma 3.1. Note that in all steps the relation s i 1 ≥ s j is still valid for vertices v j disconnected to v i 1 . At the end all the vertex disconnected to v i 1 has zero weight. If all vertices of N (v i 1 ) are connected to each other, then N (v i 1 ) is a |N (v i 1 )|-clique satisfying the thesis of this theorem.
Otherwise, among all the vertices connected to v i 1 find one, say v i 2 , such that it is not connected to all vertices of N (v i 1 ) whose value of s i 2 is maximum between those vertices in N (v i 1 ). Repeat the procedure now with vertex v i 2 . Make this until we have a probability distribution w such that the only weights If w = (w 1 , . . . , w n ) is a probability distribution on V (G), we will say that w is also a probability distribution on a k-clique Cl Remember that a homogeneous probability distribution has all weights equal. The next lemma shows that for all probability distributions w on a k-clique Cl k (G), the homogeneous probability distribution w on the k-clique Cl k (G) satisfies f G (w) ≥ f G (w).
Lemma 3.2. Let G be a simple graph with n vertices and consider w a probability distribution over a k-clique Cl k (G).
and then we conclude If A 0 = Cl k (G) there is nothing to do, else, consider the vertex v r+1 with weight w r+1 . Take ε = w r+1 −w 1 r+1 > 0. Consider a probability distribution w 1 such that w 1 r+1 = w r+1 − ε, w 1 r = w r + ε and w 1 k = w k for all k ̸ = r ̸ = r + 1. By Lemma 3.2 f G (w 1 ) ≥ f G (w). Consider a probability distribution w 2 such that w 2 r+1 = w 1 r+1 − ε = w r+1 − 2ε, w 2 r−1 = w r−1 + ε, w 2 r = w 1 r and w 2 k = w 1 for k ∈ {1, . . . , r − 2}. Again by Lemma 3.2, we have f G (w 2 ) ≥ f G (w). At the end of this process we have a set A 1 = {v 1 , . . . , v r , v r+1 } ⊃ A 0 where all vertices have weights equal to w 1 + ε. If A 1 = Cl k (G) there is nothing to do, otherwise repeat the process until we have a homogeneous probability distribution w on the clique Cl k (G). According Lemma 3.2, we have f G (w) ≥ f G (w).

Results
Equations (1) and (2) can estimate, respectively, the number of edges and the spectral radius of a simple graph based on the maximum clique. We will show how the number of edges can also depend on the number of cliques of the graph (maximum or not). Consequently, based on the proof of Theorem 1.2, we improve the results as long as we previously know some disconnected cliques in the graph.

Generalization of Turán's Theorem
Theorem 4.1. Let G be a simple graph with n vertices. Assume the following conditions: where k is the cardinality of a k-clique Cl k (G) from G containing v and γ = 1 n r i=1 k i ; Proof. Take a homogeneous probability distribution w on V (G). We have f G (w) = |E| 1 n 2 . Clearly the vertices v such that the weight of N (v) are maximum are those with maximum degree. Let A be the set of all vertices of G that are not vertices of any of the cliques in the CL collection. Apply then the procedure described in Theorem 3.1 in vertex v with the vertices of set A until there is a probability distribution over A that is null in A ∖ Cl k (G), where Cl k (G) is a k-clique from G containing v. Apply then the procedure described in Theorem 3.2 on Cl k (G) until all the vertices weights in Cl k (G) be equal. After these procedures, all vertex weights in CL are equal to 1 n and all vertex weights in Cl k (G) are equal to 1−γ k . By Lemmas 3.1 and 3.2 we have then a new distribution w ′ on G, such that |E| 1 represents the contribution from Cl k (G) and represents the contribution from CL. Solving the inequality the result follows.
In Equation (8), note that r i=1 k i n , i.e, the amount of the edges considering all the cliques in CL.
If there are no cliques on CL the previous theorem is reduced to Theorem 1.1.
We can obtain a shorter inequality from Theorem 4.1, taking account that 1 − 1 for all clique Cl k (G). Then we have Corollary 4.1. With the same hypotheses as the previous theorem, we have It is clear that γ and k i n ∈ [0, 1], ∀i = 1, . . . , r, moreover γ + r i=1 k i n = 1, then ς < 1 and inequality (9) is better than inequality (1).

Generalization of Nikiforov's Theorem
Theorem 4.2. With the same hypotheses as Theorem 4.1 we have that the spectral radius λ satisfies Proof. If |E(G)| = 0 then λ = 0 and the result is obvious. Suppose from now on that |E(G)| ≥ 1.
Fix an enumeration of the vertices of G. Let A be its respective adjacency matrix and y =    a unit eigenvector related to λ that is, ||y|| = 1. We know that: i v k j the k-th edge of G associated with the number y k i · y k j such that y k i , y k j ∈ {y 1 , . . . , y n } which are the coordinates of the y vector. Then in the inner product (11) we have: The number 2 comes from symmetry of A. We can then rewrite Equation (12) We also have: as long as: By Cauchy-Schwartz inequality, we have: Replacing Equation (13), (14) and (15) in Equation (16), we get: Because y is unitary, w = (y 2 1 , . . . , y 2 n ) is a probability distribution on V (G). Through combinatorial analysis it is easy to see that the number of edges E k in the complete graph induced by a k-clique is equal to: Then a k-clique with homogeneous probability distribution w ′ satisfies Replacing Equation (18) in Equation (19), we get: According procedures explained in Theorem3.1 and Theorem 3.2, for any probability distribution w, there is a k-clique such that: We have k ≤ cl(G), so we conclude that: Replacing Equation (22) in Equation (17), we get: Here is an example. Let G be the graph with 12 vertices and 14 edges of Figure 1. Its spectral radius λ = 2.6729197, the cardinality of the maximum clique is cl(G) = 3 and the maximum degree in the graph is equal to 4. In blue, we have a 2-clique Cl 2 (G), a 3-clique Cl 3 (G) and in red a star N (v 7 ) all disconnected. With this sets we have ς = 31 72 and according Equation (9) in Corollary 4.1, |E| ≤ 20.6666667 (Equation (1) gives |E| ≤ 48 ). According Equation (10) in Theorem 4.2, λ ≤ 5.249339 (Equation (2) gives λ ≤ 8).
It follows another example. Let H be the graph with 30 vertices and 33 edges of Figure 2. Its spectral radius λ = 2.9883861, the cardinality of the maximum clique is cl(H) = 2 and the maximum degree in the graph is equal to 6. In blue, we have four 2-cliques and in red a star N (v 16 ) all disconnected. With this sets we have ς = 5 9 and according Equation (9) in Corollary 4.1, |E| ≤ 125 (Equation (1) gives |E| ≤ 225 ). According Equation (10) in Theorem 4.2, λ ≤ 11.180340 (Equation (2) gives λ ≤ 15).