Graph convolutional network (GCN) is generalization of convolutional neural network (CNN) to work with arbitrarily structured graphs. A binary adjacency matrix is commonly used in training a GCN. Recently, the attention mechanism allows the network to learn a dynamic and adaptive aggregation of the neighborhood. We propose a new GCN model on the graphs where edges are characterized in multiple ...

A molecular graph M is a simple graph in which atoms and chemical bonds are the vertices and edges of M, respectively. The molecular graph M is called a fullerene graph, if M is the molecular graph of a fullerene molecule. It is well-known that such molecules exist for even integers n ≥ 24 or n = 20. The aim of this paper is to investigate the topological properties of a class of fullerene mole...

Different graphical tools (Bond Graph, Signed diercted graph, Temporal Causal Garaph, ...) have been developed to perform diagnosis of chemical processes. As hypergraphs generalize graphs in which an edge can connect any number of vertices, this paper aims to generate a hypergraph model-based fault detection and isolation. The directed hypergraph model of chemical reaction kinetics have been us...

Physical systems possess a structure and display dynamics. The article describes how a structure of one class of systems, which are encountered in chemical kinetics, can be revealed. The structure may be represented by a graph that has some similarities with the graph used to display the structure of chemical molecules. The graphs are induced by one-parameter transformation groups in nonlinear ...

Among various topological indices considered in chemical graph theory, only a few have been widely studied in mathematical and chemical literatures. However, it seems that less attention has been paid to the Wiener polarity index. It was introduced by Harold Wiener1 for acyclic molecules in 1947. The Wiener polarity index of an organic molecule the molecular graph of which is G = (V,E) is defin...

reliability wiener number is a modification of the original wiener number in which probabilities are assigned to edges yielding a natural model in which there are some (or all) bonds in the molecule that are not static. various probabilities naturally allow modelling different types of chemical bonds because chemical bonds are of different types and it is well-known that under certain condition...

Design support in chemical engineering is typically connected with the configuration of the technical devices that realize a chemical process. In fact, the configuration step is a rather finishing job within the entire design process, and several hurdles have been taken by the designer before. This paper outlines both an idea and a methodology to realize a design support right from the start of...

In the research of QSAR and QSPR correlations, topological indices such as Randi’c index, Zagreb ABC geometric-arithmetic index have been proposed to analyze bio-compatibility chemical compounds. Chemical graph theory is combination Chemistry Graph theory, this deals with topology mathematical study isomerism development which has applications in QSPR. article, we consider Octahedron networks f...

We develop a binary relaxation scheme for graph matching in chemical databases. The technique works by iteratively pruning the list of matching possibilities for individual atoms based upon contextual information. Its key features include delayed decision-making, robustness to noise, and fast and eecient neural implementation. We illustrate the utility of the technique by comparing it with prob...