Structure and modeling of the network of two-Chinese-character compound words in the Japanese language

TL;DR

This paper models the network of two-Chinese-character compound words in Japanese, showing it is small-world and scale-free, with importance based on character frequency, and reproduces real network properties.

Contribution

It introduces a numerical model for the two-character network that captures its small-world and scale-free properties based on character importance.

Findings

The network is small-world and scale-free.

The model reproduces statistical properties of the actual network.

Character frequency influences network structure.

Abstract

This paper proposes a numerical model of the network of two-Chinese-character compound words (two-character network, for short). In this network, a Chinese character is a node and a two-Chinese-character compound word links two nodes. The basic framework of the model is that an important character gets many edges. As the importance of a character, we use the frequency of each character appearing in publications. The direction of edge is given according to a random number assigned to nodes. The network generated by the model is small-world and scale-free, and reproduces statistical properties in the actual two-character network quantitatively.