dorsal/arxiv
View SchemaThe network of concepts in written texts
| Authors | Silvia M. G. Caldeira, Thierry C. Petit Lobao, R. F. S. Andrade, Alexis Neme, J. G. V. Miranda |
|---|---|
| Categories | |
| ArXiv ID | physics/0508066 |
| URL | https://arxiv.org/abs/physics/0508066 |
| DOI | 10.1140/epjb/e2006-00091-3 |
Abstract
Complex network theory is used to investigate the structure of meaningful concepts in written texts of individual authors. Networks have been constructed after a two phase filtering, where words with less meaning contents are eliminated, and all remaining words are set to their canonical form, without any number, gender or time flexion. Each sentence in the text is added to the network as a clique. A large number of written texts have been scrutinized, and its found that texts have small-world as well as scale-free structures. The growth process of these networks has also been investigated, and a universal evolution of network quantifiers have been found among the set of texts written by distinct authors. Further analyzes, based on shufling procedures taken either on the texts or on the constructed networks, provide hints on the role played by the word frequency and sentence length distributions to the network structure. Since the meaningful words are related to concepts in the author's mind, results for text networks may uncover patterns in communication and language processes that occur in the mind.
{
"annotation_id": "5624fe9f-d7b7-4920-bd19-7e76192305b0",
"date_created": "2026-03-02T18:01:00.137000Z",
"date_modified": "2026-03-02T18:01:00.137000Z",
"file_hash": "dcb4b4acadd2526c216402301f33023f90ef3a47d62437d82508becb1775bafd",
"private": false,
"record": {
"abstract": "Complex network theory is used to investigate the structure of meaningful\nconcepts in written texts of individual authors. Networks have been constructed\nafter a two phase filtering, where words with less meaning contents are\neliminated, and all remaining words are set to their canonical form, without\nany number, gender or time flexion. Each sentence in the text is added to the\nnetwork as a clique. A large number of written texts have been scrutinized, and\nits found that texts have small-world as well as scale-free structures. The\ngrowth process of these networks has also been investigated, and a universal\nevolution of network quantifiers have been found among the set of texts written\nby distinct authors. Further analyzes, based on shufling procedures taken\neither on the texts or on the constructed networks, provide hints on the role\nplayed by the word frequency and sentence length distributions to the network\nstructure. Since the meaningful words are related to concepts in the author\u0027s\nmind, results for text networks may uncover patterns in communication and\nlanguage processes that occur in the mind.",
"arxiv_id": "physics/0508066",
"authors": [
"Silvia M. G. Caldeira",
"Thierry C. Petit Lobao",
"R. F. S. Andrade",
"Alexis Neme",
"J. G. V. Miranda"
],
"categories": [
"physics.data-an"
],
"doi": "10.1140/epjb/e2006-00091-3",
"title": "The network of concepts in written texts",
"url": "https://arxiv.org/abs/physics/0508066"
},
"schema_id": "dorsal/arxiv",
"source": {
"execution_id": "cc429f06-dd1c-45cc-8f4b-d756a1fac2f3",
"id": "arXiv Dataset IDs",
"type": "Model",
"variant": "snapshot-2026-03-01",
"version": "0.1.0"
},
"user_id": 1000002
}