Available online at www.sciencedirect.

com
ScienceDirect
AKCE International Journal of Graphs and Combinatorics xxx (xxxx) xxx
www.elsevier.com/locate/akcej

A new matrix representation of multidigraphs

Sasmita Barika ,∗, Gopinath Sahoob
a School of Basic Sciences, IIT Bhubaneswar, Bhubaneswar, 752050, India
b Indian Statistical Institute, Delhi, 110016, India

Received 2 April 2019; received in revised form 2 July 2019; accepted 25 July 2019
Available online xxxx

Abstract
In this article, we introduce a new matrix associated with a multidigraph, named as the complex adjacency matrix. We study
the spectral properties of bipartite multidigraphs corresponding to the complex adjacency matrix. It is well known that a simple
undirected graph is bipartite if and only if the spectrum of its adjacency matrix is symmetric about the origin (with multiplicity).
We show that the result is not true in general for multidigraphs and supply a class of non-bipartite multidigraphs which have
this property. We describe the complete spectrum of a multi-directed tree in terms of the spectrum of the corresponding modular
tree. As a consequence, we get a class of Hermitian matrices for which the spectrum of a matrix in the class and the spectrum
of the modulus (entrywise) of the matrix are the same.
⃝c 2019 Kalasalingam University. Production and Hosting by Elsevier B.V. This is an open access article under the CC BY-NC-ND
license (http://creativecommons.org/licenses/by-nc-nd/4.0/).
Keywords: Multidigraph; Complex adjacency matrix; Complex adjacency spectrum; SO-property

1. Introduction
Throughout this article, we consider multidigraphs without having self-loops. By a multidigraph, we mean a
digraph (directed graph), where multiple directed edges between pair of vertices are allowed (see [1]). Two vertices
in a multidigraph are said to be adjacent if there is at least one directed edge between them. Treating each undirected
edge as equivalent to two oppositely oriented directed edges with the same end vertices, the class of all undirected
graphs may be viewed as a subclass of the class of multidigraphs.
The adjacency matrix is a popular matrix representation of a graph and the relationship between the eigenvalues
of adjacency matrix with the graph structure has been studied by many researchers in the past, see for example [1,2].
For a multidigraph G on n vertices, the adjacency matrix of G is defined [1] as the n×n matrix A(G) = [ai j ], whose
i jth entry ai j is equal to the number of directed edges originating from the vertex i and terminating at the vertex j.
From the definition, it is clear that the adjacency matrix of a multidigraph is not symmetric, in general. So it may
have complex eigenvalues. As a result, the comparison of eigenvalues for different multidigraphs is not possible
because the interlacing results [3] cannot be applied. Furthermore, it is known that not only the eigenvalues but
also the eigenvectors of different matrix representations of an undirected graph carry information about the structure
of a graph, see for example [4–6] and the references therein. Moreover, a graph is completely determined by its

Peer review under responsibility of Kalasalingam University.

∗ Corresponding author.
E-mail addresses: [email protected] (S. Barik), [email protected] (G. Sahoo).

https://doi.org/10.1016/j.akcej.2019.07.002
0972-8600/⃝ c 2019 Kalasalingam University. Production and Hosting by Elsevier B.V. This is an open access article under the CC BY-NC-ND
license (http://creativecommons.org/licenses/by-nc-nd/4.0/).

Please cite this article as: S. Barik and G. Sahoo, A new matrix representation of multidigraphs, AKCE International Journal of Graphs and Combinatorics (2019),
https://doi.org/10.1016/j.akcej.2019.07.002.
2 S. Barik and G. Sahoo / AKCE International Journal of Graphs and Combinatorics xxx (xxxx) xxx

Fig. 1. A multidigraph G on 7 vertices.

adjacency eigenvalues and the corresponding eigenvectors. This is evident from the fact that a graph G is uniquely
determined by A(G). If G is an undirected simple graph, then A(G) is symmetric. Thus, if x1 , x2 , . . . , xn are n
linearly independent eigenvectors of A(G) corresponding to the eigenvalues λ1 , λ2 , . . . , λn , respectively, consider
the n ×n matrix V with columns as xi , then A(G) = VΛV T , where Λ = diag(λ1 , . . . , λn ). But, the adjacency matrix
of a multidigraph most often fails to possess a complete set of linearly independent eigenvectors. There may be
some eigenvalues of the matrix for which geometric multiplicity is less than its algebraic multiplicity. For example,
consider the following multidigraph.

Example 1. Consider the multidigraph G in Fig. 1. The known adjacency matrix of G is given by
⎡ ⎤
0 1 0 2 0 0 0
⎢0 0 2 0 0 0 0⎥
⎢ ⎥
⎢3 0 0 2 2 1 0⎥
⎢0 0 0 0 0 0 0⎥ .
⎢ ⎥
A(G) = ⎢ ⎥
⎢0 0 0 0 0 0 1⎥
⎢ ⎥
⎣0 0 1 0 0 0 0⎦
0 0 0 0 0 0 0
Notice that it is an asymmetric real matrix of order 7 and its characteristic polynomial is given by
√ √
φ(A(G), x) = x 7 − x 5 − 6x 4 = x 4 (x − 2)(x + 1 − 2i)(x + 1 + 2i).
√ √
So the eigenvalues of A(G) are 2, −1 − 2i, −1 + 2i and 0 (algebraic multiplicity of 0 is 4 and that of other
eigenvalues is 1 each). The corresponding eigenvectors are given by
⎛ √ ⎞ ⎛ ⎞
−1 + 2i −1
√
⎟ ⎜ 1 − 2i ⎟
3
⎜
⎛ ⎞ ⎜ ⎟ ⎜ ⎟ ⎛ ⎞ ⎛ ⎞ ⎛ ⎞
1 ⎜ ⎜
1√ ⎟ ⎜
⎟ ⎜
1√ ⎟
⎟ 1 0 0
⎜2⎟ ⎜ ⎟ ⎜ ⎟ ⎜ 0 ⎟ ⎜−2⎟ ⎜0⎟
⎜ ⎟ ⎜ −1 − 2i ⎟ ⎜ −1 + 2i ⎟ ⎜ ⎟ ⎜ ⎟ ⎜ ⎟
⎜2⎟ ⎜ ⎟ ⎜ ⎟ ⎜0⎟ ⎜0⎟ ⎜0⎟
⎜0⎟ , ⎜
⎜ ⎟ ⎜ 2 ⎟,⎜
⎟ ⎜ 2 ⎟ , ⎜ 0 ⎟ , ⎜ 1 ⎟ and ⎜ 0 ⎟ ,
⎟ ⎜ ⎟ ⎜ ⎟ ⎜ ⎟
⎜ ⎟ ⎜
⎜0⎟ ⎜ 0 ⎟
⎟ ⎜
⎜ 0 ⎟ ⎜ ⎟
⎟ ⎜0⎟ ⎜0⎟
⎜ ⎟ ⎜ ⎟
⎜1⎟
⎜ ⎟ ⎜ ⎟ ⎜ ⎟ ⎜ ⎟ ⎜ ⎟ ⎜ ⎟
⎝1⎠ ⎜ 0 ⎟ ⎜ 0 ⎟ −3
⎟ ⎝ ⎠ ⎝−2⎠ ⎝−2⎠
1 1
⎜ ⎟ ⎜
0 ⎜ ⎜
⎟ ⎜
⎟ ⎜
⎟
⎟ 0 0 0
⎝ 2 ⎠ ⎝ 2 ⎠
0 0
respectively. The last three linearly independent eigenvectors correspond to the 0 eigenvalue. Note here that the
geometric multiplicity of 0 is one less than its algebraic multiplicity.

Please cite this article as: S. Barik and G. Sahoo, A new matrix representation of multidigraphs, AKCE International Journal of Graphs and Combinatorics (2019),
https://doi.org/10.1016/j.akcej.2019.07.002.
 S. Barik and G. Sahoo / AKCE International Journal of Graphs and Combinatorics xxx (xxxx) xxx 3

To avoid the above described difficulty, we associate a new Hermitian matrix to a multidigraph, which reduces
to the usual adjacency matrix in case of an √
undirected simple graph, when each undirected edge is treated as a pair
of oppositely oriented edges. Let i denote −1, the imaginary unit.

Definition 2. Let G = (V, E) be a multidigraph with V = {1, 2, . . . , n}. Let bi j and f i j denote the number of
directed edges from j to i and from i to j, respectively. Then the complex adjacency matrix AC (G) of G is an
n × n matrix AC (G) = [ai j ] whose rows and columns are indexed by V and the i jth entry is given by
( f i j + bi j ) ( f i j − bi j )
ai j = + × i.
2 2
The eigenvalues and eigenvectors of AC (G) are called the AC -eigenvalues and AC -eigenvectors of G, respectively.
The spectrum of AC (G) is called the complex adjacency spectrum (or in short AC -spectrum ) of G and is denoted
by σAC(G). Note that AC (G) is Hermitian, so all its eigenvalues are real and we have a complete set of linearly
independent eigenvectors.

Example 3. Consider the multidigraph G in Fig. 1. The complex adjacency matrix of G is given by
⎡ ⎤
1 i 3 3i
⎢ 0 +
2 2 2
−
2
1+i 0 0 0 ⎥
⎢ ⎥
⎢1 i ⎥
⎢ −
⎢2 2 0 1+i 0 0 0 0 ⎥⎥
⎢ ⎥
⎢ 3 3i ⎥
⎢ +
⎢2 1−i 0 1+i 1+i 1 0 ⎥
2 ⎥
⎥.
⎢ ⎥
AC (G) = ⎢
⎢ 1−i
⎢ 0 1−i 0 0 0 0 ⎥⎥
⎢
⎢ 0 1 i⎥
0 1−i 0 0 0 + ⎥
2 2⎥
⎢ ⎥
⎢
⎢ ⎥
⎢ 0 0 1 0 0 0 0 ⎥
⎢ ⎥
⎣ 1 i ⎦
0 0 0 0 − 0 0
2 2
It is a complex Hermitian matrix of order 7 and its characteristic polynomial is given by
29 37 3 15
φ(AC (G), x) = x 7 − x 5 − 3x 4 + x 3 + x 2 − x.
2 2 2 4
The eigenvalues of AC (G) (correct to 4 places of decimals) are −3.4903, −1.1707, −0.5319, 0, 0.4670, 0.9885 and
3.7375, and the corresponding eigenvectors are
⎛ ⎞ ⎛ ⎞ ⎛ ⎞ ⎛ ⎞
−0.4316 − 0.2239i −0.5163 + 0.0457i −0.1172 + 0.0784i −0.0000 + 0.0000i
⎜ 0.2859 − 0.2217i ⎟ ⎜ 0.0203 − 0.0593i ⎟ ⎜−0.2352 + 0.0879i⎟ ⎜−0.3082 + 0.4144i⎟
⎜ ⎟ ⎜ ⎟ ⎜ ⎟ ⎜ ⎟
⎜ 0.0000 + 0.6701i ⎟ ⎜ 0.0000 − 0.2116i ⎟ ⎜ 0.0000 − 0.1445i ⎟ ⎜ 0.0000 + 0.0000i ⎟
⎜−0.0042 − 0.2515i⎟ , ⎜ 0.5827 − 0.2993i ⎟ , ⎜ 0.3447 − 0.0960i ⎟ , ⎜ 0.1541 − 0.2072i ⎟ ,
⎜ ⎟ ⎜ ⎟ ⎜ ⎟ ⎜ ⎟
⎜ ⎟ ⎜ ⎟ ⎜ ⎟ ⎜ ⎟
⎜−0.2002 − 0.2002i⎟ ⎜ 0.2845 + 0.2845i ⎟ ⎜−0.3542 − 0.3542i⎟ ⎜ 0.0000 + 0.0000i ⎟
⎜ ⎟ ⎜ ⎟ ⎜ ⎟ ⎜ ⎟
⎝ 0.0000 − 0.1920i ⎠ ⎝−0.0000 + 0.1807i⎠ ⎝−0.0000 + 0.2717i⎠ ⎝−0.4674 − 0.6694i⎠
0.0574 −0.2430 0.6659 0
⎛ ⎞ ⎛ ⎞ ⎛ ⎞
0.0217 − 0.1952i −0.0585 − 0.4222i −0.2615 − 0.4318i
⎜−0.5390 + 0.1210i⎟ ⎜−0.4299 + 0.0028i⎟ ⎜ 0.0815 − 0.1970i ⎟
⎜ ⎟ ⎜ ⎟ ⎜ ⎟
⎜ 0.0000 + 0.1649i ⎟ ⎜ 0.0000 + 0.1846i ⎟ ⎜ 0.0000 − 0.6511i ⎟
⎜−0.0184 − 0.1111i⎟ , ⎜−0.2995 − 0.1811i⎟ and ⎜−0.3597 − 0.2198i⎟ , respectively.
⎜ ⎟ ⎜ ⎟ ⎜ ⎟
⎜ ⎟ ⎜ ⎟ ⎜ ⎟
⎜−0.2732 − 0.2732i⎟ ⎜ 0.3824 + 0.3824i ⎟ ⎜−0.1807 − 0.1807i⎟
⎜ ⎟ ⎜ ⎟ ⎜ ⎟
⎝ 0.0000 + 0.3532i ⎠ ⎝ 0.0000 + 0.1867i ⎠ ⎝ 0.0000 − 0.1742i ⎠
−0.5850 0.3868 −0.0483
Thus, all the eigenvalues of AC (G) are real and we have exactly 7 mutually orthogonal eigenvectors.

Please cite this article as: S. Barik and G. Sahoo, A new matrix representation of multidigraphs, AKCE International Journal of Graphs and Combinatorics (2019),
https://doi.org/10.1016/j.akcej.2019.07.002.
4 S. Barik and G. Sahoo / AKCE International Journal of Graphs and Combinatorics xxx (xxxx) xxx

Notice that, twice the real part of the i jth entry of AC (G) is equal to the total number of directed edges between
i and j and the imaginary part stands for the effective/resultant orientation of directed edges from i to j. So, if we
view an undirected graph as a multidigraph by considering an edge of the graph same as two oppositely oriented
directed edges, then the complex adjacency matrix of a graph is same as its adjacency matrix. Given a multidigraph
G, the adjacency matrix A(G) and the complex adjacency matrix AC (G) of a multidigraph G are related in the
following way:
A(G) = Re(AC (G)) + Im(AC (G)).
In a more descriptive way, if AC (G) = [ci j ] and A(G) = [ai j ], then
( ) ( )
ai j + a ji ai j − a ji
ci j = + i.
2 2
Now that we have a Hermitian matrix associated to a multidigraph, we can always compare complex adjacency
spectra of two multidigraphs on a given number of vertices. Using the interlacing results for Hermitian matrices [3],
we have the following immediate results.

Lemma 4. Let G be a multidigraph on vertices 1, . . . , n and H be a multidigraph produced from G by deleting

a directed edge e from G. If λ1 (G) ≤ λ2 (G) ≤ · · · ≤ λn (G) and λ1 (H ) ≤ λ2 (H ) ≤ · · · ≤ λn (H ) are the
AC -eigenvalues of G and H , respectively, then λ1 (G) ≤ λ2 (H ); λi−1 (H ) ≤ λi (G) ≤ λi+1 (H ), for i = 2, . . . , n − 1,
and λn−1 (H ) ≤ λn (G).

Proof. Suppose that the deleted directed edge of G is e = (i, j). Then AC (G) = AC (H ) + B, where B is the matrix
with bi j = 12 (1 + i), b ji = 12 (1 − i) and all other entries zero. Since B has exactly one positive eigenvalue and
exactly one negative eigenvalue, the result follows by applying Weyl’s theorem [3]. □

Lemma 5. Let G be a multidigraph on n + 1 vertices and H be obtained by deleting one vertex from G. If
λ1 (G) ≤ λ2 (G) ≤ · · · ≤ λn+1 (G) are the AC -eigenvalues of G written in nondecreasing order and λ1 (H ) ≤
λ2 (H ) ≤ · · · ≤ λn (H ) are the AC -eigenvalues of H written in nondecreasing order, then
λ1 (G) ≤ λ1 (H ) ≤ λ2 (G) ≤ · · · ≤ λn (G) ≤ λn (H ) ≤ λn+1 (G).

Proof. Observe that AC (H ) is a principal submatrix of AC (G). In particular,

[ ]
AC (H ) y
AC (G) = ,
y∗ 0
for some vector y ∈ Cn . Now using Cauchy’s interlacing theorem [3], the result follows immediately. □
The above lemma ensures that if λ is an AC -eigenvalue of the multidigraph G with multiplicity at least 2, then
λ is also an AC -eigenvalue of H with multiplicity at least 1.
To avoid drawing several arcs between a pair of vertices in a multidigraph, we use the following alternative.
Take the underlying undirected simple graph of the multidigraph and give an arbitrary orientation to each of its
edges. If in the new (oriented) graph, there is a directed edge from vertex i towards vertex j, then assign that edge
with weight equal to the i jth component of the complex adjacency matrix of that multidigraph.
So in this way, we draw a weighted { directed graph in place of a multidigraph, where
} the weights are complex
numbers belonging to the set W+ = a2 + b2 i : a, b ∈ Z, a ≥ |b| ≥ 0 and 2|(a − b) \{0}. If G is a multidigraph,
then G u and G w denote its underlying undirected simple graph and associated weighted directed graph, respectively.
Fig. 2 shows a multidigraph, its underlying undirected simple graph and the associated weighted directed graph.

Remark 6. Note that, in the above process, if we fix the orientation of the edges of G w in the following way, then
we can get a unique weighted mixed graph associated with the multidigraph. For any edge {i, j} in the underlying
weighted graph, give an orientation from i to j (from j to i) if the imaginary part of the i jth entry is positive
(negative) and keep the edge {i, j} as undirected if the i jth entry is real and positive.

Please cite this article as: S. Barik and G. Sahoo, A new matrix representation of multidigraphs, AKCE International Journal of Graphs and Combinatorics (2019),
https://doi.org/10.1016/j.akcej.2019.07.002.
 S. Barik and G. Sahoo / AKCE International Journal of Graphs and Combinatorics xxx (xxxx) xxx 5

Fig. 2. A multidigraph G, its underlying undirected simple graph G u and associated weighted directed graph G w .

Notice that, in this way, we can always associate a weighted digraph with a multidigraph such that the complex
adjacency matrix of a multidigraph is same as the adjacency matrix of the associated weighted digraph. In
the past decade, the spectral properties of complex weighted digraphs have been studied. In [7], Bapat, Kalita
and Pati introduced the adjacency matrix for a weighted digraph with complex weights of unit modulus. Later,
Kalita [8] gave a characterization of the unicyclic weighted digraphs G with weights from the set {±1, ±i} whose
adjacency matrix A(G) satisfies property (SR), i.e., “if λ is an eigenvalue of A(G), then λ1 is also an eigenvalue
of A(G) with the same multiplicity”. Recently, Sahoo [9] considered complex adjacency matrix of a digraph and
showed that not only its eigenvalues but also its eigenvectors carry a lot of information about the structure of the
digraph.
We emphasize here that an undirected simple graph H can be viewed as a multidigraph G. In that case, the
adjacency matrix A(H ) is same as the complex adjacency matrix AC (G). Hence, we may view the study of complex
adjacency spectrum of a multidigraph as a general activity.
w
We write i −→ j to mean that there are Re(w) + Im(w) directed edges from i to j and Re(w) − Im(w) directed
edges from j to i in the multidigraph G. In this case, if w ∈ W+ , then the vertices i and j in G are called adjacent;
otherwise they are nonadjacent. A multidigraph G is said to be bipartite multidigraph if its underlying undirected
simple graph is bipartite. Similarly, we can define a multi-directed tree, a path multidigraph, a star multidigraph,
etc. Let G be a multidigraph on vertices 1, 2, . . . , n. Then the modular graph of G, denoted by |G|, is the weighted
undirected graph obtained from G w by replacing each of its edge weights by their modulus.
In this article, we supply some results on the complex adjacency spectra of multidigraphs. In Section 2, we
compute the complex adjacency spectra of some special class of bipartite multidigraphs. It is well known that
a simple undirected connected graph is bipartite if and only if its adjacency spectrum is symmetric about the
origin (with multiplicity). We show that the result is not true, in general, for multidigraphs with respect to complex
adjacency spectrum and supply a class of non-bipartite multidigraphs which have this property. In Section 3, we
describe the complete AC -spectra of some special multi-directed trees. Further, given any multi-directed tree T , we
w
prove that if we change the direction of any edge in the associated weighted tree Tw (that is, if we replace i −→ j by
w
→ j for two adjacent vertices i and j) then the complete AC -spectrum of the corresponding new multi-directed
i −
tree remains unchanged. Furthermore, we prove that a multi-directed tree T and its modular tree |T | share the same
AC -spectrum.
Following notations are being used in the rest of the paper. The n × 1 vector with each entry 1 (respectively,
0) is denoted by 1n (respectively, 0n ). Mn denotes the set of complex matrices of order n. By In , we denote the
identity matrix of size n. For z ∈ C, the set of all complex numbers, z, arg(z) and |z| represent the conjugate,
n
the argument and the modulus of z. We choose arg(0) = 0, as a convention. By x = (xi )i=1 , we mean a column
vector of length n. A vector, whose ith and the jth components are 1 and −1, respectively and rest all components
are 0, is denoted by ϵi, j . The Euclidean norm of x is denoted by ∥x∥, while the Euclidean inner product of two
vectors x, y is denoted by ⟨x, y⟩. The transpose and conjugate transpose of a matrix A are denoted by A T and A∗ ,
respectively.

Please cite this article as: S. Barik and G. Sahoo, A new matrix representation of multidigraphs, AKCE International Journal of Graphs and Combinatorics (2019),
https://doi.org/10.1016/j.akcej.2019.07.002.
6 S. Barik and G. Sahoo / AKCE International Journal of Graphs and Combinatorics xxx (xxxx) xxx

Fig. 3. A non-bipartite multidigraph G whose AC -spectrum is symmetric about origin and G w , the corresponding associated weighted cycle
digraph.

2. AC –Spectra of bipartite multidigraphs

[ ] ( )
0 B x (1)
Note that, if A = is a square matrix and x = is an eigenvector for an eigenvalue λ, then
( (1) ) B∗ 0 x (2)
−x
x̂ = is an eigenvector of A for the eigenvalue −λ. In fact, if x1 , . . . , xk are k linearly independent
x (2)
eigenvectors of A, then x̂1 , . . . , x̂k are also linearly independent. That is, the eigenvalues of A are symmetric about
the origin.
The complex adjacency matrix of a bipartite multidigraph G has the above mentioned form and hence its AC -
eigenvalues are symmetric about the origin. We say a multidigraph G satisfies SO-property if the AC -spectrum of
G is symmetric about origin. Note that the converse of this is true for connected undirected graphs. But, in the case
of multidigraphs it is not true in general. See the following example.

Example 7. Consider the multidigraph G as shown in Fig. 3. One can check that the characteristic polynomial of
AC (G) is
√ √
( 7 + 3 2 )( 2 7 − 3 2 )
φ(AC (G), x) = x x −
2
x − .
2 2
Thus the AC -spectrum of G is symmetric about origin but G is not bipartite (since its underlying undirected simple
graph is a cycle on odd number of vertices).

Now, a natural question that arise here is “which non-bipartite multidigraphs satisfy the SO-property?”. Here we
provide a class of non-bipartite multidigraphs with SO-property.
wi
Let G be a cycle multidigraph with vertices 1, . . . , n and with the associated weighted digraph G w . If i −
→ (i +1)
wn
for i = 1, . . . , n − 1 and n −→ 1 in G w and no other vertices are adjacent, then we denote the cycle multidigraph G
by Cn (w), where w = (wi )i=1 n
. Depending on whether the number of vertices of a cycle multidigraph is odd or even,
it can be categorized as odd or even cycle multidigraph. Since the even cycle multidigraphs are bipartite, hence they
satisfy SO-property. To find out which odd cycle multidigraphs satisfy SO-property, we consider weight of a cycle
multidigraph which is defined as the product of weights of all the directed edges of the associated weighted proper
cycle digraph. (Note: A proper cycle digraph is a cycle digraph such that each of its directed edges are oriented
either clockwise or anticlockwise.) Notice that the weight of the cycle multidigraph as shown in Fig. 3 is 5i 4
. The
following theorem supplies a necessary and sufficient condition under which an odd cycle multidigraph satisfies
SO-property.

Theorem 8. Let G = Cn (w) be an odd cycle multidigraph on n vertices, where w = (wi )i=1
n
∈ Wn+ . Then the
weight of G is purely imaginary if and only if G satisfies SO-property.

Please cite this article as: S. Barik and G. Sahoo, A new matrix representation of multidigraphs, AKCE International Journal of Graphs and Combinatorics (2019),
https://doi.org/10.1016/j.akcej.2019.07.002.
 S. Barik and G. Sahoo / AKCE International Journal of Graphs and Combinatorics xxx (xxxx) xxx 7

Proof. The complex adjacency matrix of G can be expressed as

0 w1 0 · · · wn
⎡ ⎤
0
⎢w1 0 w2 · · · 0 0 ⎥
⎢ 0 w2 0 · · ·
⎢ ⎥
0 0 ⎥
AC (G) = ⎢ . .. .. . . .. .. ⎥ .
⎢ ⎥
⎢ .. . . . . . ⎥
wn−1 ⎦
⎢ ⎥
⎣0 0 0 ··· 0
wn 0 0 · · · w n−1 0
Let the characteristic polynomial of AC (G) be given by
φ(AC (G), x) = x n + an−1 x n−1 + · · · + a2 x 2 + a1 x + a0 .
Now, suppose w1 w2 . . . wn = a + bi, where a, b ∈ R. Then a0 = det(AC (G)) = w1 w2 . . . wn + w1 w 2 . . . w n =
a + bi + a − bi = 2a. Hence, a0 = 0 if and only if a = 0. Furthermore, observe that all the principal minors of
odd orders are zero. Therefore, the characteristic polynomial of AC (G) becomes
φ(AC (G), x) = x n + an−2 x n−2 + · · · + a3 x 3 + a1 x.
Hence the result follows. □
w
Let G = (V, E) be a multidigraph and G 1 = (V1 , E 1 ) be such that V1 ⊂ V , E 1 ⊂ E. For i, j ∈ V1 , if i −→ j
w
in G 1 for w ∈ W+ implies i − → j in G, then we call G 1 as a sub-multidigraph of G. Let us call a multidigraph
in which weights of all odd cycle sub-multidigraphs are purely imaginary as an im-bipartite multidigraph. Thus,
all bipartite multidigraphs are also im-bipartite. The following theorem gives a further sufficient condition for the
AC -spectrum of a multidigraph to be symmetric about the origin.

Theorem 9. An im-bipartite multidigraph satisfies SO-property.

Proof. Let G be an im-bipartite multidigraph on vertices 1, 2, . . . , n. Let p ≤ n be an odd number and B = [bi j ]
be the principal submatrix of AC (G) corresponding to {1, 2, . . . , p}. From Leibniz’s formula for the determinant of
a matrix, we have
p
∑ ∏
det(B) = sgn(ς) biςi ,
ς ∈S p i=1

where S p is the permutation group of {1, . . . , p} and ςi = ς (i). Since p is odd, thus ς ∈ S p is a product of
disjoint cycles of which at least one must be odd. Select an odd cycle O that contains the smallest possible label
from {1, 2, . . . , p}. Now consider the permutation, ς ′ obtained from ς, by replacing O with its inverse O −1 . Using
the hypothesis, the total contribution of ς and ς ′ towards det B is
∏
(weight of O + weight of O −1 ) weight of other cycles = 0,
as the weight of O is purely imaginary. Hence det(B) = 0. With a similar argument one can observe that the
determinant of any principal submatrix of order p is zero. It follows that, the coefficient of x n− p in the characteristic
polynomial of A, is zero. □
Next we consider some special types of bipartite multidigraphs and characterize their AC -spectrum. In the case of
undirected bipartite graphs, we have a special class of graphs called the complete bipartite graphs. The adjacency
spectrum of a complete bipartite graph contains exactly two nonzero eigenvalues which can be obtained easily
from the number of vertices in each part. Motivated by this, we define below some special classes of bipartite
multidigraphs and obtain their AC -spectra .
p
Let w = (wi ), m = (m i ) ∈ W+ . Consider a bipartite multidigraph G with V1 = {1, 2, . . . , p} and
wi
V2 = { p + 1, p + 2, . . . , p + q} as its disjoint vertex set partition. If i −
→ j for all i ∈ V1 and j ∈ V2 , then
we call G a type-I bipartite multidigraph and denote by K p,q (w). If w1 = · · · = w p = α, then we call G as a
semi-regular bipartite multidigraph and denote by α-K p,q .

Please cite this article as: S. Barik and G. Sahoo, A new matrix representation of multidigraphs, AKCE International Journal of Graphs and Combinatorics (2019),
https://doi.org/10.1016/j.akcej.2019.07.002.
8 S. Barik and G. Sahoo / AKCE International Journal of Graphs and Combinatorics xxx (xxxx) xxx

Fig. 4. The associated weighted digraph of a type-2 bipartite multidigraph K rp,q (w, m).

wi
→ j for i ∈ V1 , j ∈ S
Further, if V2 is again partitioned into two disjoint sets S and T with |S| = r and if i −
mi
→ j for i ∈ V1 , j ∈ T , then we call G a type-II bipartite multidigraph and denote it by K rp,q (w, m). Fig. 4
and i −
shows the associated weighted digraph of K rp,q (w, m).
Our next theorem describes the AC -spectrum of type-II bipartite multidigraph.

Theorem 10. Let G = K rp,q (w, m) be a type-II bipartite multidigraph as described above. Then the AC -spectrum
of G consists of
(i) 0 √
with multiplicity p + q − 4,
√
r ∥w∥2 +(q−r )∥m∥2 ± (r ∥w∥2 −(q−r )∥m∥2 )2 +4r (q−r )⟨w,m⟩⟨m,w⟩
(ii) ± 2
each with multiplicity 1.

Proof. Consider the p × q matrix B = w w

m , with the first r columns [as vector] w and
[ ]
··· m ···
0 B
the rest q − r columns as vector m. Then the adjacency matrix of G can be expressed as A = . Since
B∗ 0
p
w, m ∈ W+ , there are at least p − 2 linearly independent vectors say, x1 , x2 , . . . , x p−2 in W , which are orthogonal
p

to both w and m. Now, consider the following p + q − 4 linearly independent vectors

⎛ ⎞ ⎛ ⎞ ⎛ ⎞
0p 0p xl
⎝ ϵ1, j ⎠ , ⎝ 0r ⎠ , and ⎝ 0r ⎠ ,
0q−r ϵ1,k 0q−r
for j = 2, 3, . . . , r ; k = 2, 3, . . . , q − r ; l = 1, 2, . . . , p − 2. Observe that 0 is one eigenvalue of A afforded by
these p + q − 4 eigenvectors.
Looking at the structure of the matrix A, let us consider a vector v of the form
⎛ ⎞
k1 (w + m)
v = ⎝ k2 1r ⎠ , where k1 and k2 are some constants.
1q−r
Note that v is orthogonal to the p + q − 4 vectors mentioned above. Now, if v is an eigenvector of A corresponding
to the eigenvalue λ (say), then v must satisfy the equation Av = λv.
Now from the matrix equation, we get
k2 nwi + (q − n)m i = k1 (wi + m i )λ for i = 1, . . . , p; (1)
p
∑
k1 (wi + m i )wi = k2 λ; (2)
i=1

Please cite this article as: S. Barik and G. Sahoo, A new matrix representation of multidigraphs, AKCE International Journal of Graphs and Combinatorics (2019),
https://doi.org/10.1016/j.akcej.2019.07.002.
 S. Barik and G. Sahoo / AKCE International Journal of Graphs and Combinatorics xxx (xxxx) xxx 9

p
∑
k1 (wi + m i )m i = λ, (3)
i=1

where k1 , k2 and λ are the unknowns.

Multiplying Eq. (1) by wi and m i , separately, for i = 1, . . . , p, we have the following:
k2 n⟨w, w⟩ + (q − n)⟨m, w⟩ = k1 λ(⟨w, w⟩ + ⟨m, w⟩), (4)
k2 n⟨w, m⟩ + (q − n)⟨m, m⟩ = k1 λ(⟨w, m⟩ + ⟨m, m⟩). (5)
Further, Eqs. (2) and (3) can be restated as
k1 (⟨w, w⟩ + ⟨m, w⟩) = k2 λ, (6)
k1 (⟨w, m⟩ + ⟨m, m⟩) = λ. (7)
Eliminating k1 , we get
k2 λ2 = k2r ⟨w, w⟩ + (q − r )⟨m, w⟩, λ2 = k2r ⟨w, m⟩ + (q − r )⟨m, m⟩.
Now eliminating k2 from these two equations, we get an expression for λ from which we can get the eigenvalues
of A and hence the result. □
By considering r = q, as an immediate corollary we get the following result which describes the AC -spectrum
of a type-I bipartite multidigraph.
p
Corollary 11. Let w = (wi ) ∈ W+ . Then the AC -spectrum of K p,q (w) consists of 0 with multiplicity p + q − 2
√ √
and ± q∥w∥ with multiplicity 1 each. In particular, the spectrum of α-K p,q consists of ±|α| pq and 0 with
multiplicity p + q − 2.

Remark 12. Since each undirected edge between a pair of vertices can be considered as two oppositely oriented
directed edges between those vertices, hence as a special case of Corollary 11 we get the spectrum of a complete
√
bipartite undirected graph K p,q . It consists of ± pq and 0 with multiplicity p + q − 2.

3. Results on AC -spectra of multi-directed trees

This section contains results on the spectral properties of multi-directed trees.
A star multidigraph is a multidigraph whose underlying undirected simple graph is a star graph. Let G be a star
wi
multidigraph on n vertices with vertex 1 as the central vertex. Let w = (wi ) ∈ Wn−1 → i for i = 1, . . . , n −1,
+ . If 1 −
then we denote G by Sn (w). If all the components of w are equal, that is, wi = α for all i = 1, . . . , n − 1, then we
call Sn (w) as semiregular star multidigraph and denote by α-Sn . The following corollary follows from Corollary 11
by considering Sn (w) = K 1,n−1 (w) which describes the complete AC -spectrum of a star multidigraph.

Corollary 13. Let w = (wi ) ∈ Wn−1 + . Let Sn (w) be a star multidigraph with central vertex 1. Then the spectrum
of Sn (w) consists of 0 with multiplicity n − 2 and ±∥w∥
√ with multiplicity 1 each. More specifically, the spectrum
of α-Sn consists of 0 with multiplicity n − 2 and ±|α| n − 1 with multiplicity 1 each.
By a double star multidigraph, we mean a multidigraph G for which G u is a double star. Consider two star
multidigraphs Sn (w) and S N (m), where w and m are two vectors of length n − 1 and N − 1, respectively whose
components belong to the set W+ . Let 1 and 1′ are the central vertices of Sn (w) and S N (m), respectively. Let
wc
wc ∈ W+ . Then the multidigraph formed by joining the central vertices of Sn (w) and S N (m) such that 1 −→ 1′ ,
is known as a double star multidigraph and denoted by Sn,N (w, m; wc ). Instead of joining the central vertices, if
we merge a pendant vertex (say n) of Sn (w) to a pendant vertex (say N ′ ) of S N (m), then the multidigraph thus
produced is called a pendant-merge double star multidigraph and denoted by Sn,N (w, m), see Fig. 5. Note that
Sn,N (w, m; wc ) and Sn,N (w, m) have n + N and n + N − 1 number of vertices, respectively.
Our next two results describe the AC -spectrum of a double star multidigraphs.

Theorem 14. Let G = Sn,N (w, m; wc ) be a double star multidigraph. Then the AC -spectrum of G consists of

Please cite this article as: S. Barik and G. Sahoo, A new matrix representation of multidigraphs, AKCE International Journal of Graphs and Combinatorics (2019),
https://doi.org/10.1016/j.akcej.2019.07.002.
10 S. Barik and G. Sahoo / AKCE International Journal of Graphs and Combinatorics xxx (xxxx) xxx

Fig. 5. The associated weighted digraphs of a double star multidigraph and a pendant-merge double star multidigraph.

(i) 0 √
with multiplicity n + N − 4 and
√
|wc |2 +∥w∥2 +∥m∥2 ± (|wc |2 +∥w∥2 +∥m∥2 )2 −4∥w∥2 ∥m∥2
(ii) ± 2
each with multiplicity 1.

Proof. The complex adjacency matrix of Sn,N (w, m; wc ) can be expressed as

w1 · · · wn−1 wc
⎡ ⎤
0 0 ··· 0
⎢ w1 0 ··· 0 0 0 ··· 0 ⎥
⎢ .. .. . . .. .. .. .. ..
⎢ ⎥
⎢ . . . . . . . .
⎥
⎥
⎢ wn−1 0 · · ·
⎢ ⎥
0 0 0 · · · 0 ⎥.
⎥
AC (G) = ⎢
w
⎢
⎢ c 0 · · · 0 0 m 1 · · · m N −1 ⎥
⎥
⎢ 0 0 · · · 0 m 1 0 · · · 0 ⎥
⎢ . . . . . . . .
⎢ ⎥
⎣ .. .. .. .. .. .. .. ..
⎥
⎦
0 0 ··· 0 m N −1 0 · · · 0
Now, proof of part (i) is obvious by considering the nullity of AC (G). To prove part (ii), consider the vector
)T
v = k1 w 1 · · · wn−1 k2 k3 m 1 · · · k3 m N −1 ,
(

where k1 , k2 and k3 are some constants. If v happens to be an eigenvector corresponding to an eigenvalue λ of

AC (G), then it must satisfy AC (G)v = λv. From which we get
n−1
∑
wi wi + k2 wc = k1 λ,
i=1
k1 wi = wi λ, for i = 1, . . . , n − 1,
N −1
∑
k1 wc + m i m i = k2 λ,
i=1
k1 m i = k3 m i λ, for i = 1, . . . , N − 1.

Now eliminating k1 , k2 and k3 from these equations, we have

λ4 − (|wc |2 + ∥w∥2 + ∥m∥2 )λ2 + ∥w∥2 ∥m∥2 = 0.

Note that λ ̸= 0 as w and m are nonzero vectors. Hence the result follows. □

The following result describes the complete spectrum of Sn,N (w, m) in terms of n, N and the norms of the weight
vectors w, m.

Please cite this article as: S. Barik and G. Sahoo, A new matrix representation of multidigraphs, AKCE International Journal of Graphs and Combinatorics (2019),
https://doi.org/10.1016/j.akcej.2019.07.002.
 S. Barik and G. Sahoo / AKCE International Journal of Graphs and Combinatorics xxx (xxxx) xxx 11

Theorem 15. Let w = (wi ) ∈ Wn−1 + , m = (m i ) ∈ W+ . Let G = Sn,N

N −1
( (w,
) m) be a pendant-merge double star
w
multidigraph. If ŵ = (w1 , . . . , wn−2 )T , m̂ = (m 1 , . . . , m N −2 )T and t = , then the AC -spectrum of Sn,N (w, m)
m
consists of
(i) 0 √
with multiplicity n + N − 5,
√
∥t∥2 ± ∥t∥4 −4(|wn−1 |2 ∥ŵ∥2 +|m N −1 |2 ∥m̂∥2 +∥ŵ∥2 ∥m̂∥2 )
(ii) ± 2
each with multiplicity 1.

Proof. The complex adjacency matrix of G = Sn,N (w, m) can be expressed as

w1 · · · wn−2 wn−1
⎡ ⎤
0 0 0 ··· 0
⎢ w1 0 ··· 0 0 0 0 ··· 0 ⎥
⎢ .. .. . . .. .. .. . ..
⎢ ⎥
⎢ . . . . 0 . . .. .
⎥
⎥
⎢ wn−2 0 · · ·
⎢ ⎥
0 0 0 0 · · · 0 ⎥
w ⎥.
⎢ ⎥
AC (G) = ⎢
⎢ n−1 0 · · · 0 0 m N −1 0 · · · 0 ⎥
⎢ 0 0 · · · 0 m N −1 0 m 1 · · · m ⎥
N −2 ⎥
⎢
⎢ 0 0 · · · 0 0 m 1 0 · · · 0 ⎥
⎢ . . . . . . . . .
⎢ ⎥
⎣ .. .. .. .. .. .. .. .. ..
⎥
⎦
0 0 ··· 0 0 m N −2 0 · · · 0
Proof( of part (i) is immediate by observing the nullity of) AC (G). To prove part (ii), consider a vector v of the
T
form k1 w1 · · · wn−2 k2 k3 k4 m 1 · · · k4 m N −2 , where k1 , k2 , k3 and k4 are some constants. If v is
an eigenvector of AC (G) corresponding to an eigenvalue (say) λ, then from A(G w )v = λv we have the following
equations:
∥w∥2 + k2 wn = k1 λ,
k1 = λ,
k1 wn + k3 m N = k2 λ,
k4 ∥m∥2 + k2 m N = k3 λ,
k3 = k4 λ.
Eliminating k1 , k2 , k3 and k4 from the above system of equation, we get
λ4 − (∥w∥2 + ∥m∥2 )λ2 + |wn |2 ∥w∥2 + |m N |2 ∥m∥2 + ∥w∥2 ∥m∥2 = 0.
Note that λ ̸= 0 as w and m are nonzero vectors. Hence the result follows. □
A path multidigraph is a multidigraph G for which G u is a path. Let G be a path multidigraph on vertices
wi
1, 2, . . . , n. If i −
→ (i + 1) for i = 1, . . . , n − 1 in G and w = (wi ) ∈ Wn−1
+ , then we denote G by Pn (w). If
w1 = · · · = wn−1 = α, then we call the path multidigraph as the semi-regular path multidigraph and denote by
α-Pn . Since the complex adjacency matrix of a semi-regular path multidigraph is a tridiagonal Toeplitz matrix [10],
we have the following immediate lemma.

Lemma 16. Let α-Pn be a semi-regular ) multidigraph on n vertices. If α = r e in polar form, then the
( path
iθ

AC -spectrum of α-Pn consists of 2r cos n+1 for j = 1, . . . , n.

jπ

Next, we study the effect on the AC -spectrum of a multi-directed tree by reversing the orientation of any of its
edge. The following is an important observation which is true for a more general class of multidigraphs.

Theorem 17. Let G be a multidigraph on vertices 1, 2, . . . , n. Suppose that i and j be two vertices in G such that
w
→ j, for some w ∈ W+ and by removing all the arcs between vertices i and j, G becomes disconnected (that
i −
w w w
→ j is a cut arc in G w ). Let H be the multidigraph obtained from G by changing i −
is, i − → j to i −→ j. Then G
and H have the same AC -spectra.

Please cite this article as: S. Barik and G. Sahoo, A new matrix representation of multidigraphs, AKCE International Journal of Graphs and Combinatorics (2019),
https://doi.org/10.1016/j.akcej.2019.07.002.
12 S. Barik and G. Sahoo / AKCE International Journal of Graphs and Combinatorics xxx (xxxx) xxx

Fig. 6. Example of a multidigraph G, its weighted digraph G w and corresponding modular graph |G|.

Proof. Suppose that λ is an eigenvalue of AC (G) with corresponding eigenvector x = (xk )nk=1 . That is, AC (G)x =
λx. Since by deleting all the directed edges between i and j the multidigraph G becomes disconnected, hence we
get two components of G, say G 1 and G 2 . Suppose that i ∈ V (G 1 ) and j ∈ V (G 2 ). Now choose x̂ = (x̂ k )nk=1
such that x̂ i = w x ; x̂ j = x j ; x̂ p = w
w i
x , for all p ∈ V (G 1 ); and x̂ q = xq , for all q ∈ V (G 2 ). Then it can be
w p
observed that x̂ is an eigenvector corresponding to the eigenvalue λ of AC (H ). Hence the result follows. □

Since by deleting all the arcs between any two adjacent vertices of a multi-directed tree, the resulting multidigraph
is disconnected, we have the following immediate corollary.

Corollary 18. Let T be a multi-directed tree on vertices 1, . . . , n. Let i and j be two adjacent vertices in T such
w w w
→ j, for some w ∈ W+ . Let H be the multi-directed tree obtained from T by changing i −
that i − → j to i − → j.
Then T and H have the same complex adjacency spectra.

Let G be a multidigraph on vertices 1, 2, . . . , n. Then the modular graph of G, denoted by |G|, is the weighted
w
undirected simple graph obtained from G w by replacing each of its edge weights by their modulus. That is, if i −
→ j
in G for some w ∈ W+ , then the edge {i, j} has weight |w| in |G|. Observe that the adjacency matrix of |G| is
A(|G|) = |AC (G)|, where by |A| we mean the entrywise modulus of a matrix A. See Fig. 6 for more clarification.
The following theorem gives a relationship between the AC -spectra of a multi-directed tree and its modular tree.

Theorem 19. Let T be a multi-directed tree on n vertices and |T | be its modular tree. Let AC (T ) and A(|T |) be
the complex adjacency matrix and the adjacency matrix of T and |T |, respectively. Then both T and |T | share
same AC -spectrum, that is
σ AC (T ) = σ AC (|T |).
Furthermore, if x and y are eigenvectors of AC (T ) and A(|T |), respectively, corresponding to an eigenvalue λ, then
|x| = |y|.

n
Proof. Let y = (yi )i=1 be an eigenvector of A(|T |) = [ai j ]n×n corresponding to an eigenvalue λ. Now from the
matrix equation A(|T |)y = λy, we have
n
∑
ai j y j = λyi for i = 1, . . . , n.
k=1

Let AC (T ) = [ci j ]n×n . Notice that |AC (T )| = A(|T |) = AC (|T |) where by |A|, we mean entry-wise modulus of a
matrix A. Hence we have ai j = |ci j | = ci j earg(ci j ) . Thus,
n
∑
ci j y j ei arg(ci j ) = λyi .
k=1

Please cite this article as: S. Barik and G. Sahoo, A new matrix representation of multidigraphs, AKCE International Journal of Graphs and Combinatorics (2019),
https://doi.org/10.1016/j.akcej.2019.07.002.
 S. Barik and G. Sahoo / AKCE International Journal of Graphs and Combinatorics xxx (xxxx) xxx 13

n w
Choose a vector x = (xi )i=1 such that |x| = |y| and for i −
→ j in T ,
{
|y j |eiθ , if yi y j ≥ 0
xj =
|y j |ei(θ+π )
, otherwise
where θ = arg(xi ) + arg(w) and arg(x1 ) = 0. This is possible, as there is no cycle sub-multidigraphs. Hence x j , for
j = 1, 2, . . . , n, gets a unique valuation. Now, it is easy to observe that x is an eigenvector of AC (T ) corresponding
to the eigenvalue λ and hence we are done. □
From the well known Perron–Frobenius theorem [11], we know that the spectral radius of a nonnegative
irreducible matrix is simple and positive. Further, it tells that each component of the eigenvector corresponding
to the largest eigenvalue, commonly known as Perron vector, is positive. Using this idea, we have the following
immediate remark to state.

Remark 20. Since A(|T |) is a nonnegative irreducible matrix of the modular graph |T | of a multi-directed tree
T , from Theorem 19, the largest eigenvalue of AC (T ) is simple and positive. Furthermore, if x is an eigenvector
of AC (T ) corresponding to its largest eigenvalue, then since |Arg(w)| ≤ π4 for any w ∈ W+ , therefore the absolute
values of the difference between the principal arguments of any two components of x whose corresponding vertices
are adjacent can never be greater than π4 .
Given any Hermitian matrix A = [ai j ] of order n × n whose all diagonal entries are zero, we can associate with
it a graph G on n vertices such that two vertices i and j are adjacent in G if and only if ai j ̸= 0. The following
theorem is one of our main results which gives a sufficient condition under which A and |A| have the same set of
eigenvalues, where A is a Hermitian matrix. The proof of the result is very similar to that of Theorem 19. Especially,
from the proof of Theorem 19, notice that the values of the entries of AC (T ) do not play any role, rather the zero
pattern present in this matrix accounts. Hence we have the following result which is the most important result of
this section.

Theorem 21. Let A be a Hermitian matrix of order n × n with all its diagonal entries zero such that its associated
graph is a tree. Then A and |A| have the same set of eigenvalues. More generally, if D is a real diagonal matrix
of order n × n, then D + A and D + |A| also have the same set of eigenvalues.

Example 22. Let

⎡ √ ⎤ ⎡ √ ⎤
0
√ − 2 0 0 0 √0 2 √0 0 0
⎢− 2 3 1 − 3i 0 i⎥ ⎢ 2 3 10 0 1⎥
√
⎥. So |P| = ⎢ 0 ⎥.
⎢ ⎥ ⎢ ⎥
P=⎢ ⎢ 0 1 + 3i −1 0.5 0⎥ ⎢ 10 −1 0.5 0 ⎥
⎣ 0 0 0.5 0 0 ⎦ ⎣ 0 0 0.5 0 0⎦
0 −i 0 0 1.8 0 1 0 0 1.8
Notice that the matrix P satisfies the condition given in Theorem 21. From computation through matlab, we get
the spectra of P and |P| as
σ (P) = −3.0229, −0.2406, 0.1424, 1.6478, 5.2734 = σ (|P|)
( )

which agrees with the result given in Theorem 21.

4. Conclusion
Even though associating a complex Hermitian matrix to a digraph is not new, the use of such matrices for a
multidigraph is new to the literature. The real part of the complex adjacency matrix provides information about
the total number of directed edges between any two vertices of a multidigraph, while the imaginary part gives the
effective direction of any multi-directed edge.
In case of a multi-directed tree T , the complex adjacency spectrum of T is same as the adjacency spectrum of its
modular graph |T |. Besides, the eigenvectors of the complex adjacency matrix of T specify the effective orientation
of any multi-directed edge in T . Furthermore, we have got a class of Hermitian matrices for which the spectrum
of a matrix in the class and the spectrum of the modulus (entrywise) of the matrix are the same.

Please cite this article as: S. Barik and G. Sahoo, A new matrix representation of multidigraphs, AKCE International Journal of Graphs and Combinatorics (2019),
https://doi.org/10.1016/j.akcej.2019.07.002.
14 S. Barik and G. Sahoo / AKCE International Journal of Graphs and Combinatorics xxx (xxxx) xxx

We get a class of non-bipartite multidigraphs whose eigenvalues are symmetric about origin. In the process,
we attempt to find the class of all Hermitian matrices whose eigenvalues are symmetric about origin. Here we get
partial answers to this class of matrices.

Acknowledgments
The authors sincerely thank the reviewer for many valuable suggestions which improved the presentation of the
article. The corresponding author acknowledges SERB, Government of India for financial support through grant
(MTR/2017/000080). The second author acknowledges SERB, Government of India for financial support through
grant (PDF/2018/000519).

References
[1] D.M. Cvetković, M. Doob, H. Sachs, Spectra of Graphs: Theory and Application, Academic press, 1980.
[2] R.B. Bapat, Graphs and Matrices, Springer, 2010.
[3] R.A. Horn, C.R. Johnson, Matrix Analysis, Cambridge university press, 2012.
[4] Y. Koren, Drawing graphs by eigenvectors: theory and practice, Comput. Math. Appl. 49 (2005) 1867–1888.
[5] S.-L. Lee, Y.-N. Yeh, On eigenvalues and eigenvectors of graphs, J. Math. Chem. 12 (1993) 121–135.
[6] D.L. Powers, Graph partitioning by eigenvectors, Linear Algebra Appl. 101 (1988) 121–133.
[7] R.B. Bapat, D. Kalita, S. Pati, On weighted directed graphs, Linear Algebra Appl. 436 (2012) 99–111.
[8] D. Kalita, S. Pati, A reciprocal eigenvalue property for unicyclic weighted directed graphs with weights from {±1, ±i}, Linear Algebra
Appl. 449 (2014) 417–434.
[9] G. Sahoo, Complex adjacency spectra of digraphs, Linear and Multilinear Algebra, http://dx.doi.org/10.1080/03081087.2019.1591337.
[10] A. Böttcher, S.M. Grudsky, Spectral Properties of Banded Toeplitz Matrices, SIAM Publications, 2005.
[11] H. Minc, Nonnegative Matrices, John Wiley & Sons, Inc., 1988.

Please cite this article as: S. Barik and G. Sahoo, A new matrix representation of multidigraphs, AKCE International Journal of Graphs and Combinatorics (2019),
https://doi.org/10.1016/j.akcej.2019.07.002.