Text as Statistical Mechanics Object

TL;DR

This paper models human written text using a statistical mechanics framework, deriving potential energy from large corpora, and finds that specific heat distinguishes between closed class words and specific terms.

Contribution

It introduces a novel statistical mechanics model for text analysis, linking thermodynamic concepts to linguistic features.

Findings

Specific heat parameter separates closed class words from content words.

Numerical validation confirms the model's effectiveness.

Potential energy derivation from corpus data enhances text analysis.

Abstract

In this article we present a model of human written text based on statistical mechanics approach by deriving the potential energy for different parts of the text using large text corpus. We have checked the results numerically and found that the specific heat parameter effectively separates the closed class words from the specific terms used in the text.