Searcharxiv⌕ Search

arXiv subjects

Ercan Altınışık

Publications and source records attributed to Ercan Altınışık.

5 recordsLinked to original sources

Some results and a conjecture on certain subclasses of graphs according to the relations among certain energies, degrees and conjugate degrees of graphs

Let $G$ be a simple graph of order $n$ with degree sequence $(d)=(d_1,d_2,\ldots,d_n)$ and conjugate degree sequence $(d^*)=(d_1^*,d_2^*,\ldots,d_n^*)$. In \cite{AkbariGhorbaniKoolenObudi2010,DasMojallalGutman2017} it was proven that $\mathcal{E}(G)\leq \sum_{i=1}^{n} \sqrt{d_i}$ and $\sum_{i=1}^{n} \sqrt{d_i^*} \leq LEL(G) \leq IE(G) \leq \sum_{i=1}^{n} \sqrt{d_i}$, where $\mathcal{E}(G)$, $LEL(G)$ and $IE(G)$ are the energy, the Laplacian-energy-like invariant and the incidence energy of $G$, respectively, and in \cite{DasMojallalGutman2017} it was concluded that the class of all connected simple graphs of order $n$ can be dividend into four subclasses according to the position of $\mathcal{E}(G)$ in the order relations above. Then, they proposed a problem about characterizing all graphs in each subclass. In this paper, we attack this problem. First, we count the number of graphs of order $n$ in each of four subclasses for every $1\leq n \leq 8$ using a Sage code. Second, we present a conjecture on the ratio of the number of graphs in each subclass to the number of all graphs of order $n$ as $n$ approaches the infinity. Finally, as a first partial solution to the problem, we determine subclasses to which a path, a complete graph and a cycle graph of order $n\geq 1$ belong.

math.CO↗

Non-divisibility of LCM Matrices by GCD Matrices on GCD-closed Sets

In this paper, we consider the divisibility problem of LCM matrices by GCD matrices in the ring $M_n(\mathbb{Z})$ proposed by Hong in 2002 and in particular a conjecture concerning the divisibility problem raised by Zhao in 2014. We present some certain gcd-closed sets on which the LCM matrix is not divisible by the GCD matrix in the ring $M_n(\mathbb{Z})$. This could be the first theoretical evidence that Zhao's conjecture might be true. Furthermore, we give the necessary and sufficient conditions on the gcd-closed set $S$ with $|S|\leq 8$ such that the GCD matrix divides the LCM matrix in the ring $M_n(\mathbb{Z})$ and hence we partially solve Hong's problem. Finally, we conclude with a new conjecture that can be thought as a generalization of Zhao's conjecture.

math.NT↗

On the Ilmonen-Haukkanen-Merikoski Conjecture

Let $K_n$ be the set of all $n\times n$ lower triangular (0,1)-matrices with each diagonal element equal to $1$, $L_n = \{ YY^T: Y\in K_n\}$ and let \begin{equation*} c_n = \min_{Z\in L_n} \left\lbrace μ_n^{(1)}(Z):μ_n^{(1)} (Z) \text{ is the smallest eigenvalue of } Z \right\rbrace . \end{equation*} The Ilmonen-Haukkanen-Merikoski conjecture (the IHM conjecture) states that $c_n$ is equal to the smallest eigenvalue of $Y_0Y_0^T$, where \begin{equation*} (Y_0)_{ij}=\left\lbrace \ \begin{array}{cl} 0 & \text{if } \ i j. \end{array} \right. \end{equation*} In this paper we present a proof of this conjecture. In our proof we use an inequality for spectral radii of nonnegative matrices.

math.CO↗

Determinants of Circulant Matrices with Some Certain Sequences

Let $\{a_k\}$ be a sequence of real numbers defined by an $m$th order linear homogenous recurrence relation. In this paper we obtain a determinant formula for the circulant matrix $A=circ(a_1, a_2, \cdots, a_n)$, providing a generalization of determinantal results in papers of Bozkurt \cite{Bozkurt}, Bozkurt and Tam \cite{BozkurtTam}, and Shen, et al. \cite{ShenCenHao}.

math.CA↗