On the fractal patterns of language structures

Citation:

Ribeiro LC, Bernardes AT, Mello H. On the fractal patterns of language structures. PLOS ONE [Internet]. 2023;18 (5) :1-20.

Date Published:

Abstract:

Natural Language Processing (NLP) makes use of Artificial Intelligence algorithms to extract meaningful information from unstructured texts, i.e., content that lacks metadata and cannot easily be indexed or mapped onto standard database fields. It has several applications, from sentiment analysis and text summary to automatic language translation. In this work, we use NLP to figure out similar structural linguistic patterns among several different languages. We apply the word2vec algorithm that creates a vector representation for the words in a multidimensional space that maintains the meaning relationship between the words. From a large corpus we built this vectorial representation in a 100-dimensional space for English, Portuguese, German, Spanish, Russian, French, Chinese, Japanese, Korean, Italian, Arabic, Hebrew, Basque, Dutch, Swedish, Finnish, and Estonian. Then, we calculated the fractal dimensions of the structure that represents each language. The structures are multi-fractals with two different dimensions that we use, in addition to the token-dictionary size rate of the languages, to represent the languages in a three-dimensional space. Finally, analyzing the distance among languages in this space, we conclude that the closeness there is tendentially related to the distance in the Phylogenetic tree that depicts the lines of evolutionary descent of the languages from a common ancestor.

Website

Publicações Recentes do Departamento de Física

Fabris JC, Toniato JD, Velten HES. Gravitação. 1st ed. São Paulo: Editora Livraria da Física; 2021 pp. 405. Publisher's Version

Cota AB. TERMODINÂMICADOS MATERIAIS. 1st ed. Ouro Preto; 2020.

abcota_termodinamica_dos_materiais.pdf

G Vieira C, P H de Oliveira MA, Fernandes WP, Matos MJS, Paniago R, de Carvalho VE, Chacham H, A Soares E. Unveiling Pseudo-Dirac Resonances in the (7×7)R19.1° Phase of Sulfur Segregated on Cu(111) Surface. The Journal of Physical Chemistry C [Internet]. Submitted :null. Publisher's Version

Gontijo RN, Moutinho MVO, Righi A, Chiu P-W, Venezuela P, Pimenta MA. Resonant enhancement of the 2G Raman band in twisted bilayer graphene. Materials Chemistry and Physics [Internet]. 2024 :129279. Publisher's Version Abstract

Raman spectroscopy is an extremely useful tool to characterize graphene systems. The strongest Raman features are the first-order G band and the second-order 2D and 2D′ bands, which are the overtones of the double resonance D and D’ bands. However, the 2G band, which is the overtone of the G band, is not usually observed in the spectra of monolayer graphene and of crystalline graphite. In this work, we present an experimental and theoretical investigation of the resonance Raman spectra in twisted bilayer graphene (TBG) with different twisting angles and using several laser excitation energies in the NIR and visible ranges. We observed that the 2G band is enhanced when the incident photons are in resonance with the transition between the van Hove singularities in the density of states of the TBG. We show that the 2G band has three contributions (2G1, 2G2 and 2G3), that are not dispersive by changing the laser excitation energy. We also present theoretical calculations showing that the 2G1 and 2G2 bands are related to combinations of the in-phase (IP) and out-of-phase (OP) vibrations of the atoms in the different layers. The Raman excitation profiles (REPs) of the 2G peaks are upshifted in comparison with the REP of the G band. This behavior was confirmed theoretically using a graphene tight binding model. We conclude that the different resonance behavior comes from the fact that the G band is a first-order process whereas the 2G band is second-order processes giving rise to overall different resonance conditions.

Freitas WD, Mazzoni MSC, Matos MJS, Galvão BRL. Predicting molecular crystals of polynitrogen (N6) structures with cage-like geometries using ab initio evolutionary algorithm. Chemical Physics Letters [Internet]. 2024;844 :141262. Publisher's Version Abstract

In this work, we employed Density Functional Theory calculations combined with search techniques based on evolutionary algorithms to predict and characterize crystalline structures composed of nitrogen (N6) cage-like molecules. We found stable molecular crystals and a rich phenomenology associated with their behavior under pressure, including atomic rebonding and semiconductor-metal transitions. This investigation resides in the context of high-energy-density materials, since molecular species containing only nitrogen atoms tend to dissociate into N2 molecules, releasing large amounts of energy.

Omwoyo E, Belich H, Fabris JC, Velten H. Black hole lensing in Kerr-de Sitter spacetimes. The European Physical Journal Plus. 2023;138 (11) :1043.