Current Organizations / Institutes

The organizations/institutes from the scholar's latest article


Ivan Castillo Zuñiga has a total of 21 co-authors. Here are the top co-authors by number of shared publications:


Latest Publications

  • Open Access

    Temporal Machine Learning Models for Classifying Suspected Dengue Cases in Mexico Using Surveillance Data from 2025

    by Jorge Soria-Cruz, Enrique Luna-Ramírez, Iván Castillo-Zúñiga, Jaime Iván López-Veyna, Ma. Angélica Estrada-Ramírez, Juan Antonio González-Morales

    Background: Dengue fever remains a major public health challenge in Mexico, exhibiting pronounced seasonal behavior and substantial geographic heterogeneity. Using recent epidemiological surveillance data may improve predictive performance and better reflect the current epidemiological context. Objective: The aim of this study was to develop and compare temporal machine learning models for the binary classification of confirmed and negative dengue cases in Mexico using 2025 national surveillance data. Methods: A total of 68,222 suspected dengue cases reported in 2025 were analyzed. The outcome variable was CASE_STATUS, encoded as 0 for negative cases and 1 for confirmed cases. The dataset was divided chronologically into training (January–September), validation (October), and testing (November–December) subsets. Nine machine learning algorithms were evaluated: Random Forest, Bayesian Network, XGBoost, CatBoost, Naïve Bayes, Logistic Regression, Multilayer Perceptron, Support Vector Machine, and LightGBM. Preprocessing included scaling, encoding, age discretization for Bayesian Network, class imbalance treatment, and model-specific feature-importance analyses. Performance was assessed using accuracy, Precision, Recall, F1-score, ROC-AUC, and PR-AUC. Results: Random Forest achieved the best overall performance, with the highest test F1-score (0.7254) and PR-AUC (0.7300) at an optimized threshold of 0.397, together with a high Recall (0.8938). Bayesian Network achieved the highest test accuracy (0.7023) and ROC-AUC (0.7756), although its overall operational balance was less favorable considering class imbalance. Geographic and institutional variables were the most influential predictors across models, whereas comorbidities generally contributed less. Conclusions: Temporal machine learning models are useful for dengue case classification in Mexico, and Random Forest was the most robust approach, balancing sensitivity and overall predictive performance. From an operational perspective, this finding is especially relevant in dengue surveillance, where failure to identify true confirmed cases may have important public health consequences.

  • Specialized Word Embedding Model for the Detection of Suicidal Ideation Using Deep Learning

    by Lizethe-Guadalupe Reyna-Morán, Francisco-Javier Luna-Rosas, Julio-Cesar Martínez-Romo, Iván Castillo-Zuñiga, Marco-Antonio Hernández-Vargas, Mario-Alberto Rodríguez-Díaz

  • Open Access

    Machine Learning Models to Predict Recoveries and Deaths from COVID-19 in Mexican Society in the Post-Pandemic Era

    by Enrique Luna-Ramírez, Jorge Soria-Cruz, Iván Castillo-Zúñiga, Jaime Iván López-Veyna

    The emergence or mutation of aggressive viruses represents a latent threat to human health that could lead to new pandemics, so it is important to constantly monitor and analyze the behavior of the diseases they can cause. In this sense, the purpose of this work was to generate models to predict the behavior of recoveries and deaths from COVID-19 in Mexico in the post-pandemic era, applying machine learning techniques to data related to this disease, published by the Mexican government. Models based on artificial neural networks, logistic regression, and classification algorithms were generated and validated, yielding high rates of correct classification, accuracy, and recall, so that they could be used to make predictions about future cases of patients infected with the SARS-CoV-2 virus.


  • BIO-INSPIRED ALGORITHMS AND TRANSFER LEARNING TO OPTIMISE THE EXTRACTION OF SUICIDE-RELATED PATTERNS ON TWITTER

    by Damian Martinez, Francisco Luna Rosas, Julio Cesar Martinez Romo, Ivan Castillo Zuñiga, Marco Antonio Hernandez Vargas, Mario Alberto Rodriguez Diaz

    Artificial Intelligence (AI) has advanced significantly in several subfields, most notably Artificial General Intelligence (AGI). AGI refers to a form of AI capable of performing any intellectual task that a human being can perform, as autonomous humanoid robots would do.

  • IMPROVING TIME IN THE EXTRACTION OF LINGUISTIC PATTERNS OF SUICIDE ON TWITTER USING SWARM INTELLIGENCE

    by Damian Martinez, Francisco Luna Rosas, Julio Cesar Martinez Romo, Marco Antonio Hernandez Vargas, Mario Alberto Rodriguez Diaz, Ivan Castillo Zuñiga

    Suicide is the fourth leading cause of death for young people between 15 and 19 years of age, causing more than 703,000 deaths each year, corresponding to one death every 40 seconds, which is why it is considered a public health problem by the World Health Organization. One way to prevent suicide according to experts is to act quickly on those who are going through this situation. Twitter, like other social media platforms, is in an unique position to help, because of the cumulative feelings that are shared there, including suicidal thoughts and intentions. Given the increasing suicide rates in the world in this research we propose a model using swarm intelligence that allows us to reduce the pre-processing time of tweets in the detection of features related to suicide. The results presented prove that our model can be an alternative to reduce the time in the extraction of suicide-related linguistic patterns. Keywords: Sentiment Analysis, Suicide, Social Networks, Machine Learning, Swarm Intelligence.

  • A Review of Graph Databases

    by Jaime I. Lopez-Veyna, Ivan Castillo-Zuñiga, Mariana Ortiz-Garcia


  • Open Access

    Detection of traits in students with suicidal tendencies on Internet applying Web Mining

    by Iván Castillo-Zúñiga, Francisco-Javier Luna-Rosas, Jaime-Iván López-Veyna

    Este artículo presenta un modelo de análisis de datos en Internet basado en Minería Web con el objetivo de encontrar conocimiento sobre grandes cantidades de datos en el ciberespacio. A fin de probar el método propuesto, se analizaron páginas web sobre el suicidio como caso de estudio con la intención de identificar y detectar rasgos en estudiantes con tendencias suicidas. El procedimiento considera un Web Scraper para localizar y descargar información de Internet, así como técnicas de Procesamiento de Lenguaje Natural para la recuperación de los vocablos. Con el propósito de explorar la información, se construyó un conjunto de datos basado en Tablas Dinámicas y Ontologías Semánticas, especificando las variables predictivas en jóvenes con inclinación suicida. Por último, para evaluar la eficiencia del modelo se utilizaron algoritmos de Aprendizaje de Máquina y Aprendizaje Profundo. Cabe destacar que se optimizaron los procedimientos para la construcción del dataset (utilizando Algoritmos Genéticos) y obtención del conocimiento empleando Cómputo Paralelo y Aceleración con Unidades de Procesamientos de Gráfico (GPU). Los resultados revelan una precisión del 96,28% sobre la detección de las características en adolescentes con tendencia suicida, alcanzando el mejor resultado a través de una Red Neuronal Recurrente con un 98% de precisión. De donde se infiere que el modelo es viable para establecer bases sobre mecanismos de actuación y prevención de comportamientos suicidas, que pueden ser implementados en instituciones educativas o distintos actores de la sociedad. This article presents an Internet data analysis model based on Web Mining with the aim to find knowledge about large amounts of data in cyberspace. To test the proposed method, suicide web pages were analyzed as a study case to identify and detect traits in students with suicidal tendencies. The procedure considers a Web Scraper to locate and download information from the Internet, as well as Natural Language Processing techniques to retrieve the words. To explore the information, a dataset based on Dynamic Tables and Semantic Ontologies was constructed, specifying the predictive variables in young people with suicidal inclination. Finally, to evaluate the efficiency of the model, Machine Learning and Deep Learning algorithms were used. It should be noticed that the procedures for the construction of the dataset (using Genetic Algorithms) and obtaining the knowledge (using Parallel Computing and Acceleration with GPU) were optimized. The results reveal an accuracy of 96.28% on the detection of characteristics in adolescents with suicidal tendencies, reaching the best result through a Recurrent Neural Network with 98% accuracy. It is inferred that the model is viable to establish bases on mechanisms of action and prevention of suicidal behaviors, which can be implemented in educational institutions or different social actors.


  • Optimizing global processing time in the detection of patterns related to suicide in social networks

    by Damian Martinez Diaz, Francisco Javier Luna Rosas, Julio Cesar Martinez Romo, Marco Antonio Hernandez Vargas, Ivan Castillo Zuniga

    There is a suicide every 40 seconds in the world and it is the third cause of death for young people between 15 and 19 years old worldwide. For every suicide, many more attempt it, which is why suicide prevention remains an universal challenge and has been recognized by the World Health Organization (WHO) as a public health priority. Experts say that one of the best ways to prevent suicide is for people who are going through this urge to take their own lives to listen to people who are close to them and social networks such as Twitter or Facebook are in a unique position to help these people connect in real time in difficult situations that people with these suicidal tendencies are going through, but also represents a potential risk to receive information that could later prove harmful, either by stressing the same information or by taking some suicidal ideas. In this research we propose a model to optimize the global time processing in the detection of patterns related to suicide in the social network Twitter. Our results show that the proposed model can be a good alternative when it comes to optimizing the response time in this type of problems.

  • INTELLIGENT SYSTEM FOR DETECTION OF CYBERCRIME VOCABULARY ON WEBSITES

    by Ivan Castillo Zuñiga, Jaime Ivan Lopez Veyna, Francisco Luna Rosas, Gustavo Tirado Estrada

    While the benefits that new technologies have brought to our lives are undeniable, they have also caused new problems that were previously unknown, one of which is the so-called Cybercrime.


  • Internet Data Analysis Methodology for Cyberterrorism Vocabulary Detection, Combining Techniques of Big Data Analytics, NLP and Semantic Web

    by Iván Castillo-Zúñiga, Francisco Javier Luna-Rosas, Laura C. Rodríguez-Martínez, Jaime Muñoz-Arteaga, Jaime Iván López-Veyna, Mario A. Rodríguez-Díaz

    This article presents a methodology for the analysis of data on the Internet, combining techniques of Big Data analytics, NLP and semantic web in order to find knowledge about large amounts of information on the web. To test the effectiveness of the proposed method, webpages about cyberterrorism were analyzed as a case study. The procedure implemented a genetic strategy in parallel, which integrates (Crawler to locate and download information from the web; to retrieve the vocabulary, using techniques of NLP (tokenization, stop word, TF, TFIDF), methods of stemming and synonyms). For the pursuit of knowledge was built a dataset through the description of a linguistic corpus with semantic ontologies, considering the characteristics of cyber-terrorism, which was analyzed with the algorithms, Random Forests (parallel), Boosting, SVM, neural network, K-nn and Bayes. The results reveal a percentage of the 95.62% accuracy in the detection of the vocabulary of cyber-terrorism, which were approved through cross validation, reaching 576% time savings with parallel processing.