Following the spectral graph theory and algebraic techniques in graph theory, we continue a Coxeter spectral study of finite posets and edge-bipartite graphs (or signed graphs in the sense of Harary and Zaslavsky). A connection between properties of the Coxeter spectrum specc J ⊆ C and digraph isomorphism problem for Hasse digraphs of positive and non-negative posets J is also studied. In particular, we study in details a class of posets J with a non-negativity condition, in connection with the Coxeter spectral properties of the simply-laced Euclidean diagrams { Dn, n ≥ 4, E6, E7, E8}. We show that symbolic and numerical computer calculations in Python, C and Linux tools allow us to present a complete classification of these posets J, with at most 15 points, by means of their Coxeter spectra specc J . The main classification ideas and the algorithms used in the classification are presented in Sections 4 and 6. We end the paper by showing how our poset classification results apply to the isomorphism problem for a special class of digraphs.
I. INTRODUCTIONFollowing the spectral graph theory problems, a graph coloring techniques, and algebraic methods in graph and poset theory studied in [1], [3], [4], [11], [15], [16], we continue a Coxeter spectral study of finite posets and edge-bipartite graphs [5], [7], [16]-[22] (or signed graphs in the sense of Harary [9] and Zaslavsky [23]). We study a connection between properties of the Coxeter spectrum specc J ⊆ C and the digraph isomorphism problem for Hasse digraphs of positive and non-negative posets J, compare with [1] and [3]. Our main inspiration for the study comes from the representation theory of posets and digraphs, Lie theory, Diophantine geometry problems, and combinatorics, see [16] and [18]. For further motivation of the study the reader is referred to Section 5 and to the articles [5]-[7], [13], [14], [17]-[20], and [22].A natural way to apply spectral graph theory to directed graphs (digraphs) is to assign to any digraph D = (V D , E D ) the adjacency matrix Ad D = [a ij ] ∈ M n (Z), where n = |V D | and we set a ij = 1 if (i, j) ∈ E D and 0 otherwise. We denote by F AdD (t) := det(t · E − Ad D ) ∈ Z[t]