In order to use graphical matrices, we need to replace (sub)graphs with the invariants of choice. To exemplify this, we employ two graph invariants often used in the structure-property-activity modeling: the Randić connectivity index [79] , also known as the vertex-connectivity index [361,362] and the Hosoya index [25]. The numbers that replace the subgraphs in the graphical matrices are obtained by summing (in the case of the Randić connectivity index) and multiplying (in the case of the Hosoya index) the corresponding graph invariants. There are also other ways in which to construct numerical matrices from graphical matrices. The values of the connectivity indices and Wiener indices for acyclic subgraphs are taken from our book on computational chemical graph theory [93]. Below are given two types of numerical realization of the four graphical matrices presented in Figures 43-46
2.1. Use of the Randic Connectivity index
2.2. Use of the Hosoya Index