Application of sentimental analysis in PQRS for the identification of trends in bayesian models in IA4SG in Boyacá
https://orcid.org/0000-0002-9559-2146
This study identifies and applies sentiment analysis techniques based on Bayesian models and neural networks to classify and determine trends in Petitions, Complaints, Claims, and Suggestions (PQRS) reported by health institutions in Boyacá, Colombia, during 2024-2025, within the framework of Artificial Intelligence for Social Good (AI4SG). Four computational models —Naive Bayes, Bayesian Networks, Gaussian Processes, and Bayesian Deep Learning— were trained and evaluated using a mixed-methods approach (quantitative and qualitative) and a Machine Learning Operations (MLOps) workflow structured in four phases: natural language processing, model training, algorithmic treatment, and cross-validation. Results show that, for Gaussian Processes, the RBF and Matern kernels achieved the lowest mean squared error (MSE) and mean absolute error (MAE), with a coefficient of determination (R²) near 0.46; Bayesian Networks revealed a higher probability of negative emotions (sadness, anger, fear) in areas such as surveillance, outpatient consultation, and radiology, and higher trust in vaccination and outpatient care; the Latent Dirichlet Allocation (LDA) model reached its best topic coherence in imaging and nursing (0.50-0.52); the Naive Bayes classifier maintained a stable recall between 0.77 and 0.83 across cross-validation folds, and the Bayesian Deep Learning model reached an F1-score of 0.76. It is concluded that these tools optimize the massive processing of PQRS, provide hospital management with a precise, automated alternative for identifying service shortcomings, and strengthen the reliability of predictions against new data.
Abdessalem, Anis Ben, Nikolaos Dervilis, David J Wagg, and Keith Worden. (2017). Automatic Kernel Selection for Gaussian Processes Regression with Approximate Bayesian Computation and Sequential Monte Carlo. Frontiers in Built Environment 3: 52. doi: 10.3389/fbuil.2017.00052
Abdullah, Abdullah A, Masoud M Hassan, and Yaseen T Mustafa. (2022). A Review on Bayesian Deep Learning in Healthcare: Applications and Challenges. IEEe Access 10: 36538–62.
Altarturi, Hamza HM, Muntadher Saadoon, and Nor Badrul Anuar. (2023). Web Content Topic Modeling Using LDA and HTML Tags. PeerJ Computer Science 9: e1459. DOI 10.7717/peerj-cs.1459
Atienza, David, Concha Bielza, and Pedro Larrañaga. (2022). PyBNesian: An Extensible Python Package for Bayesian Networks. Neurocomputing 504: 204–9. https://doi.org/10.1016/j.neucom.2022.06.112
Barrachina, Estela González, Christian Fortanet van Assendelft de Coningh, Tatiana Hidalgo-Marı́, and Cristina González Dı́az. (2026). Medición Del Sentimiento Hacia La Marca: Estudio de Caso de La Marca Vicio En Instagram. Redmarka. Revista de Marketing Aplicado 30 (1): 126–45. https://github.com/jboscomendoza/lexicos-nrc-afinn/blob/master/lexico_nrc.csv .
Bird, Steven, Ewan Klein, and Edward Loper. (2009). Natural Language Processing with Python: Analyzing Text with the Natural Language Toolkit. " O’Reilly Media, Inc.
Blei, David M. (2012). Probabilistic Topic Models. Communications of the ACM 55 (4): 77–84. 10.1145/2133806
Chahar, Ravita, and Ashutosh Kumar Dubey. (2026). Machine Intelligence for Mental Health Diagnosis: A Systematic Review of Methods, Algorithms, and Key Challenges. Computers, Materials, & Continua 86 (1): 1. https://doi.org/10.32604/cmc.2025.066990
Contreras Contreras, Ghiordy Ferney, Byron Medina Delgado, Brayan René Acevedo Jaimes, and Dinael Guevara Ibarra. (2022). Metodologı́a de Desarrollo de técnicas de Agrupamiento de Datos Usando Aprendizaje Automático. Tecnura 26 (72): 5–6. https://doi.org/10.14483/22487638.17246
Demin, Alexander, and Denis Ponomaryov. (2022). Machine Learning with Probabilistic Law Discovery: A Concise Introduction. arXiv Preprint arXiv:2212.11901.
https://doi.org/10.48550/arXiv.2212.11901
Farkhod, Akhmedov, Akmalbek Abdusalomov, Fazliddin Makhmudov, and Young Im Cho. (2021). LDA-Based Topic Modeling Sentiment Analysis Using Topic/Document/Sentence (TDS) Model. Applied Sciences 11 (23): 11091. https://doi.org/10.3390/app112311091
Fernández, Laura Rodrı́guez, Ana Laura Fernández Mena, Marı́a Patricia Torres Magaña, Manuel Antonio Rodrı́guez Magaña, and Manuel Antonio Rodrı́guez Fernández. (2024). Inteligencia Artificial En La Educación: Modelo de Lenguaje de Gran Tamaño (LLM) Como Recurso Educativo. Revista Ipsumtec 7 (2): 157–64.
Jankova, Zuzana. (2023). Latent Dirichlet Allocation (LDA) Approximation Analysis of Financial-Related Text Messages. Economic Computation & Economic Cybernetics Studies & Research 57 (1).
Jospin, Laurent Valentin, Hamid Laga, Farid Boussaid, Wray Buntine, and Mohammed Bennamoun. (2022). Hands-on Bayesian Neural Networks—a Tutorial for Deep Learning Users. IEEE Computational Intelligence Magazine 17 (2): 29–48. https://doi.org/10.1109/MCI.2022.3155327
Jurafsky, Daniel, and James H. Martin. (2026). Speech and Language Processing: An Introduction to Natural Language Processing, Computational Linguistics, and Speech Recognition, with Language Models. 3rd ed. https://web.stanford.edu/~jurafsky/slp3/.
Murphy, Kevin P. (2023). Probabilistic Machine Learning: Advanced Topics. MIT press.
Patil, Rajvardhan, and Venkat Gudivada. (2024). A Review of Current Trends, Techniques, and Challenges in Large Language Models (Llms). Applied Sciences 14 (5): 2074. https://doi.org/10.3390/app14052074
Riutort-Mayol, Gabriel, Paul-Christian Bürkner, Michael R Andersen, Arno Solin, and Aki Vehtari. (2023). Practical Hilbert Space Approximate Bayesian Gaussian Processes for Probabilistic Programming.” Statistics and Computing 33 (1): 17.
SACRISTÁN, MILLÁN SANTAMARÍA. (n.d). Machine Learning Operations (MLOps): Contexto Actual y Tendencias Futuras. PhD thesis, Universidad de La Rioja.
Silander, Tomi, and Petri Myllymaki. (2012). A Simple Approach for Finding the Globally Optimal Bayesian Network Structure. arXiv Preprint arXiv:1206.6875. https://scifaro.com/en/abs/a-simple-approach-for-finding-the-globally-optimal-bayesian-network-structure-1206.6875
Van Dael, Jackie, Tom W Reader, Alex Gillespie, Ana Luisa Neves, Ara Darzi, and Erik K Mayer. (2020). Learning from Complaints in Healthcare: A Realist Review of Academic Literature, Policy Evidence and Front-Line Insights. BMJ Quality & Safety 29 (8): 684–95. doi: 10.1136/bmjqs-2019-009704. Epub 2020 Feb 4. PMID: 32019824; PMCID: PMC7398301.
Wang, Hao, and Dit-Yan Yeung. (2016). Towards Bayesian Deep Learning: A Framework and Some Existing Methods. IEEE Transactions on Knowledge and Data Engineering 28 (12): 3395–408. https://www.researchgate.net/publication/330053760_Constructing_Site-Specific_Multivariate_Probability_Distribution_Model_Using_Bayesian_Machine_Learning
Younas, Ahtisham, Sergi Fàbregues, Sarah Munce, and John W Creswell. (2025). Framework for Types of Metainferences in Mixed Methods Research. BMC Medical Research Methodology 25 (1): 18. doi: 10.1186/s12874-025-02475-8. PMID: 39856568; PMCID: PMC11758751.
Zapata-Ros, Miguel. (n.d). Métodos Bayesianos Para Detectar Artı́culos Cientı́ficos Creados Con IA.
Zou, Yixin, Ding-Bang Luh, and Shizhu Lu. (2022). Public Perceptions of Digital Fashion: An Analysis of Sentiment and Latent Dirichlet Allocation Topic Modeling. Frontiers in Psychology 13: 986838. doi: 10.3389/fpsyg.2022.986838