A Multilingual Sentiment Analysis System for TikTok Comments in Spanish Using RoBERTa and LSTM

Gabriel A. León-Paredes; Luis A. Alba-Narváez; Santiago M. Torres-Cordero; Christian M. Buestan-Villa

doi:10.1007/978-3-031-93103-1_21

A Multilingual Sentiment Analysis System for TikTok Comments in Spanish Using RoBERTa and LSTM

Gabriel A. León-Paredes
, Luis A. Alba-Narváez
, Santiago M. Torres-Cordero
, Christian M. Buestan-Villa

Grupo de Investigación en Cloud Computing Smart Cities & High Performance Computing (GIHP4C)

Producción científica: Capítulo del libro/informe/acta de congreso › Contribución de conferencia › revisión exhaustiva

1 Cita (Scopus)

Resumen

Social media platforms, particularly TikTok, have become an integral part of daily life for millions globally, generating vast amounts of user-generated content. For businesses and content creators, understanding the sentiments behind these user comments is crucial for gauging public perception and refining their commercial or marketing strategies. However, the unique nature of comments on TikTok, characterized by their brevity, informal language, slang, and emojis, presents significant challenges for sentiment analysis. We developed a sentiment analysis system to address these challenges using advanced Natural Language Processing (NLP) and Deep Learning (DL) techniques. The system’s architecture, combining a frontend built with Angular and Tailwind CSS and a backend powered by FastAPI and a fine-tuned RoBERTa-based model, allows for real-time analysis of large datasets. Our model, trained with PyTorch CUDA, achieved a high accuracy of 87.2%, with a precision of 90%, a recall of 83.6%, and an F1-score of 86.7%.

Idioma original	Inglés
Título de la publicación alojada	Information Technology and Systems - ICITS 2025
Editores	Alvaro Rocha, Carlos Ferrás, Hiram Calvo
Editorial	Springer Science and Business Media Deutschland GmbH
Páginas	210-219
Número de páginas	10
ISBN (versión impresa)	9783031931024
DOI	https://doi.org/10.1007/978-3-031-93103-1_21
Estado	Publicada - 2025
Evento	International Conference on Information Technology and Systems, ICITS 2025 - Mexico City, México Duración: 22 ene. 2025 → 25 ene. 2025

Serie de la publicación

Nombre	Lecture Notes in Networks and Systems
Volumen	1449 LNNS
ISSN (versión impresa)	2367-3370
ISSN (versión digital)	2367-3389

Conferencia

Conferencia	International Conference on Information Technology and Systems, ICITS 2025
País/Territorio	México
Ciudad	Mexico City
Período	22/01/25 → 25/01/25

Nota bibliográfica

Publisher Copyright:
© The Author(s), under exclusive license to Springer Nature Switzerland AG 2025.

Areas de Conocimiento del CACES

316A Desarrollo y análisis de software y aplicaciones

Acceder al documento

10.1007/978-3-031-93103-1_21

Otros archivos y enlaces

Enlace a la publicación en Scopus

Citar esto

León-Paredes, G. A., Alba-Narváez, L. A., Torres-Cordero, S. M., & Buestan-Villa, C. M. (2025). A Multilingual Sentiment Analysis System for TikTok Comments in Spanish Using RoBERTa and LSTM. En A. Rocha, C. Ferrás, & H. Calvo (Eds.), Information Technology and Systems - ICITS 2025 (pp. 210-219). (Lecture Notes in Networks and Systems; Vol. 1449 LNNS). Springer Science and Business Media Deutschland GmbH. https://doi.org/10.1007/978-3-031-93103-1_21

León-Paredes, Gabriel A. ; Alba-Narváez, Luis A. ; Torres-Cordero, Santiago M. et al. / A Multilingual Sentiment Analysis System for TikTok Comments in Spanish Using RoBERTa and LSTM. Information Technology and Systems - ICITS 2025. editor / Alvaro Rocha ; Carlos Ferrás ; Hiram Calvo. Springer Science and Business Media Deutschland GmbH, 2025. pp. 210-219 (Lecture Notes in Networks and Systems).

@inproceedings{72baae112a594dc1a2326fbbfacae923,

title = "A Multilingual Sentiment Analysis System for TikTok Comments in Spanish Using RoBERTa and LSTM",

abstract = "Social media platforms, particularly TikTok, have become an integral part of daily life for millions globally, generating vast amounts of user-generated content. For businesses and content creators, understanding the sentiments behind these user comments is crucial for gauging public perception and refining their commercial or marketing strategies. However, the unique nature of comments on TikTok, characterized by their brevity, informal language, slang, and emojis, presents significant challenges for sentiment analysis. We developed a sentiment analysis system to address these challenges using advanced Natural Language Processing (NLP) and Deep Learning (DL) techniques. The system{\textquoteright}s architecture, combining a frontend built with Angular and Tailwind CSS and a backend powered by FastAPI and a fine-tuned RoBERTa-based model, allows for real-time analysis of large datasets. Our model, trained with PyTorch CUDA, achieved a high accuracy of 87.2\%, with a precision of 90\%, a recall of 83.6\%, and an F1-score of 86.7\%.",

keywords = "Deep Learning, Multilingual Natural Language Processing, Parallel Computing, PyTorch, Sentiment Analysis, Social Media Analysis",

author = "Le{\'o}n-Paredes, \{Gabriel A.\} and Alba-Narv{\'a}ez, \{Luis A.\} and Torres-Cordero, \{Santiago M.\} and Buestan-Villa, \{Christian M.\}",

note = "Publisher Copyright: {\textcopyright} The Author(s), under exclusive license to Springer Nature Switzerland AG 2025.; International Conference on Information Technology and Systems, ICITS 2025 ; Conference date: 22-01-2025 Through 25-01-2025",

year = "2025",

doi = "10.1007/978-3-031-93103-1\_21",

language = "English",

isbn = "9783031931024",

series = "Lecture Notes in Networks and Systems",

publisher = "Springer Science and Business Media Deutschland GmbH",

pages = "210--219",

editor = "Alvaro Rocha and Carlos Ferr{\'a}s and Hiram Calvo",

booktitle = "Information Technology and Systems - ICITS 2025",

address = "Germany",

}

León-Paredes, GA, Alba-Narváez, LA, Torres-Cordero, SM & Buestan-Villa, CM 2025, A Multilingual Sentiment Analysis System for TikTok Comments in Spanish Using RoBERTa and LSTM. En A Rocha, C Ferrás & H Calvo (eds.), Information Technology and Systems - ICITS 2025. Lecture Notes in Networks and Systems, vol. 1449 LNNS, Springer Science and Business Media Deutschland GmbH, pp. 210-219, International Conference on Information Technology and Systems, ICITS 2025, Mexico City, México, 22/01/25. https://doi.org/10.1007/978-3-031-93103-1_21

A Multilingual Sentiment Analysis System for TikTok Comments in Spanish Using RoBERTa and LSTM. / León-Paredes, Gabriel A.; Alba-Narváez, Luis A.; Torres-Cordero, Santiago M. et al.
Information Technology and Systems - ICITS 2025. ed. / Alvaro Rocha; Carlos Ferrás; Hiram Calvo. Springer Science and Business Media Deutschland GmbH, 2025. p. 210-219 (Lecture Notes in Networks and Systems; Vol. 1449 LNNS).

Producción científica: Capítulo del libro/informe/acta de congreso › Contribución de conferencia › revisión exhaustiva

TY - GEN

T1 - A Multilingual Sentiment Analysis System for TikTok Comments in Spanish Using RoBERTa and LSTM

AU - León-Paredes, Gabriel A.

AU - Alba-Narváez, Luis A.

AU - Torres-Cordero, Santiago M.

AU - Buestan-Villa, Christian M.

N1 - Publisher Copyright: © The Author(s), under exclusive license to Springer Nature Switzerland AG 2025.

PY - 2025

Y1 - 2025

N2 - Social media platforms, particularly TikTok, have become an integral part of daily life for millions globally, generating vast amounts of user-generated content. For businesses and content creators, understanding the sentiments behind these user comments is crucial for gauging public perception and refining their commercial or marketing strategies. However, the unique nature of comments on TikTok, characterized by their brevity, informal language, slang, and emojis, presents significant challenges for sentiment analysis. We developed a sentiment analysis system to address these challenges using advanced Natural Language Processing (NLP) and Deep Learning (DL) techniques. The system’s architecture, combining a frontend built with Angular and Tailwind CSS and a backend powered by FastAPI and a fine-tuned RoBERTa-based model, allows for real-time analysis of large datasets. Our model, trained with PyTorch CUDA, achieved a high accuracy of 87.2%, with a precision of 90%, a recall of 83.6%, and an F1-score of 86.7%.

AB - Social media platforms, particularly TikTok, have become an integral part of daily life for millions globally, generating vast amounts of user-generated content. For businesses and content creators, understanding the sentiments behind these user comments is crucial for gauging public perception and refining their commercial or marketing strategies. However, the unique nature of comments on TikTok, characterized by their brevity, informal language, slang, and emojis, presents significant challenges for sentiment analysis. We developed a sentiment analysis system to address these challenges using advanced Natural Language Processing (NLP) and Deep Learning (DL) techniques. The system’s architecture, combining a frontend built with Angular and Tailwind CSS and a backend powered by FastAPI and a fine-tuned RoBERTa-based model, allows for real-time analysis of large datasets. Our model, trained with PyTorch CUDA, achieved a high accuracy of 87.2%, with a precision of 90%, a recall of 83.6%, and an F1-score of 86.7%.

KW - Deep Learning

KW - Multilingual Natural Language Processing

KW - Parallel Computing

KW - PyTorch

KW - Sentiment Analysis

KW - Social Media Analysis

UR - https://www.scopus.com/pages/publications/105012926450

U2 - 10.1007/978-3-031-93103-1_21

DO - 10.1007/978-3-031-93103-1_21

M3 - Conference contribution

AN - SCOPUS:105012926450

SN - 9783031931024

T3 - Lecture Notes in Networks and Systems

SP - 210

EP - 219

BT - Information Technology and Systems - ICITS 2025

A2 - Rocha, Alvaro

A2 - Ferrás, Carlos

A2 - Calvo, Hiram

PB - Springer Science and Business Media Deutschland GmbH

T2 - International Conference on Information Technology and Systems, ICITS 2025

Y2 - 22 January 2025 through 25 January 2025

ER -

León-Paredes GA, Alba-Narváez LA, Torres-Cordero SM, Buestan-Villa CM. A Multilingual Sentiment Analysis System for TikTok Comments in Spanish Using RoBERTa and LSTM. En Rocha A, Ferrás C, Calvo H, editores, Information Technology and Systems - ICITS 2025. Springer Science and Business Media Deutschland GmbH. 2025. p. 210-219. (Lecture Notes in Networks and Systems). doi: 10.1007/978-3-031-93103-1_21