On the origin of long-range correlations in texts

TL;DR

This paper investigates the origins of long-range correlations in texts, showing how hierarchical linguistic structures lead to bursty, correlated sequences that reflect semantic relevance, with implications for understanding language complexity.

Contribution

It explains the emergence of long-range correlations in texts through hierarchical linguistic levels and demonstrates their general applicability beyond language.

Findings

Correlations manifest as bursty sequences near semantic topics

Hierarchical linguistic structures generate long-range correlations

The mechanisms are applicable to other hierarchical systems

Abstract

The complexity of human interactions with social and natural phenomena is mirrored in the way we describe our experiences through natural language. In order to retain and convey such a high dimensional information, the statistical properties of our linguistic output has to be highly correlated in time. An example are the robust observations, still largely not understood, of correlations on arbitrary long scales in literary texts. In this paper we explain how long-range correlations flow from highly structured linguistic levels down to the building blocks of a text (words, letters, etc..). By combining calculations and data analysis we show that correlations take form of a bursty sequence of events once we approach the semantically relevant topics of the text. The mechanisms we identify are fairly general and can be equally applied to other hierarchical settings.