Universal and non-universal text statistics: Clustering coefficient for language identification
This work addresses language identification and text analysis by revealing non-universal statistical features, though it is incremental as it builds on known network methods.
The study analyzed statistical properties of texts in seven languages and random texts, confirming universal laws like Zipf's and Herdan-Heap's laws. It found that the distribution of clustering coefficients in word co-occurrence networks can differentiate between languages and distinguish natural languages from random texts.
In this work we analyze statistical properties of 91 relatively small texts in 7 different languages (Spanish, English, French, German, Turkish, Russian, Icelandic) as well as texts with randomly inserted spaces. Despite the size (around 11260 different words), the well known universal statistical laws -- namely Zipf and Herdan-Heap's laws -- are confirmed, and are in close agreement with results obtained elsewhere. We also construct a word co-occurrence network of each text. While the degree distribution is again universal, we note that the distribution of Clustering Coefficients, which depend strongly on the local structure of networks, can be used to differentiate between languages, as well as to distinguish natural languages from random texts.