arXiv subjects
Publications and source records attributed to Abderrahim Boussaïri.
An $n\times n$ real matrix $Q$ is quasi-orthogonal if $Q^{\top}Q=qI_{n}$ for some positive real number $q$. If $M$ is a principal sub-matrix of a quasi-orthogonal matrix $Q$, we say that $Q$ is a quasi-orthogonal extension of $M$. In a recent work, the authors have investigated this notion for the class of real skew-symmetric matrices. Using a different approach, this paper addresses the case of symmetric matrices.
A real matrix $Q$ is quasi-orthogonal if $Q^{\top}Q=qI$, for some positive real number $q$. We prove that any $n\times n$ skew-symmetric matrix $S$ is a principal sub-matrix of a skew-symmetric quasi-orthogonal matrix $Q$, called a quasi-orthogonal extension of $S$. Moreover, we determine the least integer $d$ such that $S$ has a quasi-orthogonal extension of order $n+d$. This integer is called the quasi-orthogonality index of $S$. Lastly, we give a spectral characterization of skew-adjacency matrices of tournaments with quasi-orthogonality index at most three.
A tournament is $k$-spectrally monomorphic if all the $k\times k$ principal submatrices of its adjacency matrix have the same characteristic polynomial. Transitive $n$-tournaments are trivially $k$-spectrally monomorphic. We show that there are no other for $k\in \{3,\ldots,n-3\} $. Furthermore, we prove that for $n\geq 5$, a non-transitive $n$-tournament is $(n-2)$-spectrally monomorphic if and only if it is doubly regular. Finally, we give some results on $(n-1)$-spectrally monomorphic regular tournaments.
A tournament is unimodular if the determinant of its skew-adjacency matrix is $1$. In this paper, we give some properties and constructions of unimodular tournaments. A unimodular tournament $T$ with skew-adjacency matrix $S$ is invertible if $S^{-1}$ is the skew-adjacency matrix of a tournament. A spectral characterization of invertible tournaments is given. Lastly, we show that every $n$-tournament can be embedded in a unimodular tournament by adding at most $n - \lfloor\log_2(n)\rfloor$ vertices.
This paper solves the following problem about Hermitian matrices related to the theory of $2$-structures:\emph{ }Let $n$ be a positive integer and $k$ be an integer with $k\in \{3,\ldots,n-3\}$. Characterize the Hermitian matrices $A$ such that the characteristic polynomials of the $k\times k$ submatrices of $A$ are all equal. Such matrices are called $k$-spectrally monomorphic. A crucial step to obtain this characterization is proving that if a matrix $A$ is $k$-spectrally monomorphic then it is $l$-spectrally monomorphic for $l$ in $\{1,\ldots,min\{k, n-k\}\}$.
An $n$-tournament $T$ with vertex set $V$ is simple if there is no subset $M$ of $V$ such that $2\leq \left \vert M\right \vert \leq n-1$ and for every $x\in V\setminus M$, either $M\rightarrow x$ or $x \rightarrow M$. The simplicity index of an $n$-tournament $T$ is the minimum number $s(T)$ of arcs whose reversal yields a non-simple tournament. Müller and Pelant (1974) proved that $s(T)\leq\frac{n-1}{2}$, and that equality holds if and only if $T$ is doubly regular. As doubly regular tournaments exist only if $n\equiv 3\pmod{4}$, $s(T)<\frac{n-1}{2}$ for $n\not\equiv3\pmod{4}$. In this paper, we study the class of $n$-tournaments with maximal simplicity index for $n\not\equiv3\pmod{4}$.
A generalized tournament matrix $M$ is a nonnegative matrix that satisfies $M+M^{t}=J-I$, where $J$ is the all ones matrix and $I$ is the identity matrix. In this paper, a characterization of generalized tournament matrices with the same principal minors of orders $2$, $3$, and $4$ is given. In particular, it is proven that the principal minors of orders $2$, $3$, and $4$ determine the rest of the principal minors.
The idiosyncratic polynomial of a graph $G$ with adjacency matrix $A$ is the characteristic polynomial of the matrix $ A + y(J-A-I)$, where $I$ is the identity matrix and $J$ is the all-ones matrix. It follows from a theorem of Hagos (2000) combined with an earlier result of Johnson and Newman (1980) that the idiosyncratic polynomial of a graph is reconstructible from the multiset of the idiosyncratic polynomial of its vertex-deleted subgraphs. For a digraph $G$ with adjacency matrix $A$, we define its idiosyncratic polynomial as the characteristic polynomial of the matrix $ A + y(J-A-I)+zA^{T}$. By forbidding two fixed digraphs on three vertices as induced subdigraphs, we prove that the idiosyncratic polynomial of a digraph is reconstructible from the multiset of the idiosyncratic polynomial of its induced subdigraphs on three vertices. As an immediate consequence, the idiosyncratic polynomial of a tournament is reconstructible from the collection of its $3$-cycles. Another consequence is that all the transitive orientations of a comparability graph have the same idiosyncratic polynomial.
A diamond is a $4$-tournament which consists of a vertex dominating or dominated by a $3$-cycle. Assuming the existence of skew-conference matrices, we give a complete characterization of $n$-tournaments with the maximum number of diamonds when $n\equiv0\pmod{4}$ and $n\equiv3\pmod{4}$. For $n\equiv2\pmod{4}$, we obtain an upper bound on the number of diamonds in an $n$-tournament and we give a matricial characterization of tournaments achieving this bound.
Let $H$ be a 3-uniform hypergraph. A tournament $T$ defined on $V(T)=V(H)$ is a realization of $H$ if the edges of $H$ are exactly the 3-element subsets of $V(T)$ that induce 3-cycles. We characterize the 3-uniform hypergraphs that admit realizations by using a suitable modular decomposition.
In this paper, we address the following problem due to Frankl and Füredi (1984). What is the maximum number of hyperedges in an $r$-uniform hypergraph with $n$ vertices, such that every set of $r+1$ vertices contains $0$ or exactly $2$ hyperedges? They solved this problem for $r=3$. For $r=4$, a partial solution is given by Gunderson and Semeraro (2017) when $n=q+1$ for some prime power number $q\equiv3\pmod{4} $. Assuming the existence of skew-symmetric conference matrices for every order divisible by $4$, we give a solution for $n\equiv0\pmod{4} $ and for $n\equiv3\pmod{4}$.
The notions of weakly bipartitive and bipartitive families were introduced by Montgolfier (2003) as a general tool for studying some decomposition of graphs and other combinatorial structures. In this paper, we give a matrix description of these notions.
Our motivation comes from the work of Engel and Schneider (1980). Their main theorem implies that two symmetric matrices have equal corresponding principal minors of all orders if and only if they are diagonally similar. This study was continued by Hartfiel and Loewy (1984). They found sufficient conditions under which two $n\times n$ matrices\ $A$ and $B$ have equal corresponding principal minors of all orders if and only if $B$ or its transpose $B^{t}$ is diagonally similar to $A$. In this paper, we give a new way to construct a pair of skew-symmetric having equal corresponding principal minors of all orders.
For a real matrix $M$, we denote by $sp(M)$ the spectrum of $M$ and by $\left \vert M\right \vert $ its absolute value, that is the matrix obtained from $M$ by replacing each entry of $M$ by its absolute value. Let $A$ be a nonnegative real matrix, we call a \emph{signing} of $A$ every real matrix $B$ such that $\left \vert B\right \vert =A$. In this paper, we study the set of all signings of $A$ such that $sp(B)=αsp(A)$ where $α$ is a complex unit number. Our work generalizes some results obtained in [1, 5, 8].
A 2-structure on a set $S$ is given by an equivalence relation on the set of ordered pairs of distinct elements of $S$. A subset $C$ of $S$, any two elements of which appear the same from the perspective of each element of the complement of $C$, is called a clan. The number of elements that must be added in order to obtain a 2-structure the only clans of which are trivial is called the primitive bound of the 2-structure. The primitive bound is determined for arbitrary 2-structures of any cardinality. This generalizes the classical results of Erdős et al. and Moon for tournaments, as well as the result of Brignall et al. for finite graphs, and the precise results of Boussa\"ıri and Ille for finite graphs, providing new proofs which avoid extensive use of induction in the finite case.
Given a graph $G$, a subset $M$ of $V(G)$ is a module of $G$ if for each $v\in V(G)\setminus M$, $v$ is adjacent to all the elements of $M$ or to none of them. For instance, $V(G)$, $\emptyset$ and $\{v\}$ ($v\in V(G)$) are modules of $G$ called trivial. Given a graph $G$, $ω_M(G)$ (respectively $α_M(G)$) denotes the largest integer $m$ such that there is a module $M$ of $G$ which is a clique (respectively a stable) set in $G$ with $|M|=m$. A graph $G$ is prime if $|V(G)|\geq 4$ and if all its modules are trivial. The prime bound of $G$ is the smallest integer $p(G)$ such that there is a prime graph $H$ with $V(H)\supseteq V(G)$, $H[V(G)]=G$ and $|V(H)\setminus V(G)|=p(G)$. We establish the following. For every graph $G$ such that $\max(α_M(G),ω_M(G))\geq 2$ and $\log_2(\max(α_M(G),ω_M(G)))$ is not an integer, $p(G)=\lceil\log_2(\max(α_M(G),ω_M(G)))\rceil$. Then, we prove that for every graph $G$ such that $\max(α_M(G),ω_M(G))=2^k$ where $k\geq 1$, $p(G)=k$ or $k+1$. Moreover $p(G)=k+1$ if and only if $G$ or its complement admits $2^k$ isolated vertices. Lastly, we show that $p(G)=1$ for every non prime graph $G$ such that $|V(G)|\geq 4$ and $α_M(G)=ω_M(G)=1$.
Given a graph G, a subset M of V (G) is a module of G if for each v \in V (G) \diagdownM, v is adjacent to all the elements of M or to none of them. For instance, V(G), \varnothing and {v} (v \in V(G)) are modules of G called trivial. Given a graph G, m(G) denotes the largest integer m such that there is a module M of G which is a clique or a stable set in G with |M|=m. A graph G is prime if |V(G)|\geq4 and if all its modules are trivial. The prime bound of G is the smallest integer p(G) such that there is a prime graph H with V(H)\supseteqV(G), H[V(G)] = G and |V(H)\diagdownV(G)|=p(G). We establish the following. For every graph G such that m(G)\geq2 and log_2(m(G)) is not an integer, p(G)=\lceil log_2(m(G)) \rceil. Then, we prove that for every graph G such that m(G)=2^k where k\geq1, p(G)=k or k + 1. Moreover p(G)=k+1 if and only if G or its complement admits 2^k isolated vertices. Lastly, we show that p(G) = 1 for every non-prime graph G such that |V(G)|\geq4 and m(G)=1.