On the nature of long-range letter correlations in texts

2008/08/31 by D. Yu. Manin, Dmitrii Y. Manin, Manin, Dmitrii Y. · 1 citation
Biochemistry, Genetics and Molecular Biology · Computer Science · Mathematics · #Computation and Language (cs.CL) #FOS: Computer and information sciences #Fractal and DNA sequence analysis #Information Theory (cs.IT) #Machine Learning in Bioinformatics #cs.CL #cs.IT #math.IT

paper · pdf · doi:10.48550/arxiv.0809.0103

14 pages, 5 figures, unpublished

arxiv created 2008/08/31 · openalex publication_date 2008/08/31 · arxiv updated 2016/11/27 · openalex created_date 2025/10/10 · openalex updated_date 2026/07/28

Abstract

The origin of long-range letter correlations in natural texts is studied using random walk analysis and Jensen-Shannon divergence. It is concluded that they result from slow variations in letter frequency distribution, which are a consequence of slow variations in lexical composition within the text. These correlations are preserved by random letter shuffling within a moving window. As such, they do reflect structural properties of the text, but in a very indirect manner.

Citations

• Analysis of symbolic sequences using the Jensen-Shannon divergence
• Segmentation and Context of Literary and Musical Sequences

Cited by

• Statistical Properties of the Rooted-Tree Encoding of ℕ

Related

• An Analysis of Letter Dynamics in the English Alphabet
• A formalism for studying long-range correlations in many-alphabets sequences
• Random Text, Zipf's Law, Critical Length,and Implications for Large Language Models
• Autocorrelations Decay in Texts and Applicability Limits of Language Models
• Fractal Illusions: An Experimental Study of Long-Range Sentence-Length Correlations in Randomly Generated Natural Language Texts