• 2025.01.29: Starting the New Year with New Challenges: How to Detect Misinformation in Electronic Messages! 

Electronic messages aren't limited to just emails—they encompass a wide range of communications, from chat messages to news shared on major media platforms. But is it possible to identify linguistic patterns in these messages? And what about psychological cues—can they also reveal misinformation? In this context, machine learning can become a highly valuable tool—not as an end in itself, but as a means to achieve effective solutions.

This topic has become more urgent due to platforms like Twitter (I refuse to go along with the clownish antics of a clown by calling it "X") and Facebook, which have chosen not to moderate the content shared on their systems. Tackling misinformation is not just important; it is essential. For this reason, I am currently working on proposing a project with external funding to develop innovative strategies for detecting and combating misinformation effectively in the digital age.

• 2024.08.09: Our latest research work titled "Unmasking Phishing Attempts: A Study on Detection in Spanish Emails" has been accepted for oral presentation at the Ibero-American Congress on Pattern Recognition (CIARP) to be held in November in Chile. In this article, we present an innovative approach that uses rule-based patterns to identify these cyberattacks, highlighting the importance of developing cybersecurity solutions adapted to specific linguistic and cultural contexts. The article also emphasizes the potential of our method to be applied to other languages, thus contributing to a safer digital environment globally. I want to acknowledge the contribution of my co-authors, whose dedication and knowledge were fundamental to the success of this research. Great work done here!

• 2024.06.27: Our manuscript titled "Uncovering Phishing Attacks Using Principles of Persuasion Analysis" has recently been accepted for publication in the Journal of Network and Computer Applications, a Q1 journal in the Scimago Journal Rank. This research delves into identifying persuasion principles (PoP) within phishing emails to enhance detection methods. We explore the use of message subjectivity, focusing on "how it is said" rather than just "what is said," to uncover the manipulative tactics employed by attackers. By analyzing various data representations, including word n-grams, TF-IDF, and transformer-based pre-trained models such as DAN, LASER, RoBERTa, and TRANSF, and evaluating diverse classifiers such as Naïve Bayes, KNN, Random Forest, SVM, and Artificial Neural Network models, we aim to identify the most effective combinations for detecting PoP. Our findings reveal that no single combination is universally superior, and the detection rate for PoP can vary. Notably, Random Forest emerges as the most suitable classifier for phishing detection using PoP, achieving an AUC-ROC of 0.859842. This novel approach extends the state-of-the-art in phishing detection, traditionally focused on objective message content analysis, by incorporating the subjective and persuasive elements of phishing attacks.

I extend my sincere gratitude to my coauthors Vitali Herrera-Semenets, Juan-Luis García-Mendoza, Miguel A. Álvarez-Carmona, Jorge A. González-Ordiano, Luis Zúñiga-Morales, J. Emilio Quiróz-Ibarra, Pedro A. Santander-Molina, and Jan van den Berg for their invaluable contributions and support in this research endeavor. Their work has been instrumental in developing our approach. This research was supported by Universidad Iberoamericana (Ibero) and the Institute of Applied Research and Technology (InIAT) through the project "Detection of Phishing Attacks in Electronic Messages Using Artificial Intelligence Techniques." Additionally, the authors gratefully acknowledge CONACYT for providing computational resources through the INAOE Supercomputing Laboratory's Deep Learning Platform for Language Technologies.

• 2024.06.20: Our paper titled "Towards a Novel Approach for Knowledge Base Population using Distant Supervision" has been accepted at the 16th Mexican Conference on Pattern Recognition (MCPR 2024) (http://www.mcpr.org). MCPR is a prominent conference in the field of Pattern Recognition, providing a platform to showcase innovative research. The accepted work addresses the challenge of improving accuracy in knowledge base population using distant supervision. Our novel approach is based on Deep Embedding Clustering to identify relationships in textual data, effectively reducing the incidence of false negatives. This method significantly enhances the efficiency and accuracy of relation extraction, which is crucial in applications such as intrusion detection and data stream analysis. Experimental results have demonstrated the algorithm's viability and effectiveness, surpassing several state-of-the-art methods, even in scenarios with mislabeled data. Great job done here! Thanks to all co-authors for their commitment and efforts.

• 2024.02.08: Artificial Intelligence and cybersecurity are intrinsically linked, from static systems to proactive solutions. AI is crucial in the early detection of cyber threats such as phishing and malware, but it also poses ethical and privacy challenges. Legal regulation, such as GDPR in the EU and CCPA in the US, seeks to balance security with privacy. The future of AI in cybersecurity promises advances in detection and autonomous response, although it faces challenges such as data scarcity and adversarial attacks. Multidisciplinary collaboration is key to developing more robust and ethical systems in a constantly evolving digital environment. Recently, an article developed among me and several colleagues, named "Artificial Intelligence in Cybersecurity," has been approved for publication in the next issue of the ReinvenTec journal of the Instituto Tecnológico de Tlalnepantla.

• 2023.09.20: The paper titled "A Decision Tree Induction Algorithm for Efficient Rule Evaluation using Shannon's Expansion" has been accepted at the MICAI conference (http://www.micai.org). MICAI is a prominent conference in the field of artificial intelligence, providing a platform to showcase innovative research. This work addresses the challenge of improving the performance of decision tree structures when representing sets of rules. The paper introduces a novel algorithm based on Shannon's expansion, designed to prevent the redundant evaluation of rule filters, even if they appear in multiple rules. This innovation significantly enhances efficiency during the evaluation of induced decision trees. Experimental results have demonstrated the algorithm's viability, particularly in processing-intensive scenarios such as intrusion detection and data stream analysis. Great job done here! Thanks to all co-authors for their commitment and efforts.

• 2023.07.15: Persuasion plays a significant role in marketing and computer crimes, influencing decisions and actions. Detecting persuasion is crucial for user protection. In the paper titled "Towards Automatic Principles of Persuasion Detection Using a Machine Learning Approach," Machine Learning is utilized to analyze messages, aiming to identify the most effective data representation and classification algorithm for identifying the principles of persuasion used in phishing attacks. This work has been accepted for presentation at the VIII International Congress on Artificial Intelligence and Pattern Recognition (IWAIPR 2023),  held in Varadero, Cuba, from September 27 to 29.

• 2023.08.31: Los resultados de la evaluación del taller "Ciencia de Datos para (no) ingenieros" realizada por parte de los participantes se pueden consultar aquí.

• 2023.08.31: Ya se encuentra disponible la grabación (sin editar o post-producir) de la sesión del taller "Ciencia de Datos para (no) ingenieros". Esta grabación puede ser revisada en la página web del taller  (https://sites.google.com/view/cd4ing/inicio) o directamente aquí.

• 2023.08.30: En el marco del evento "Construyendo el fututo: La Inteligencia Artificial" se estará impartiendo el taller "Ciencia de Datos para (no) ingenieros, que tiene como objetivo brindar a los participantes la oportunidad de aplicar técnicas de Ciencias de Datos utilizando herramientas no-code como Knime. La página web del taller se encuentra en: https://sites.google.com/view/cd4ing/inicio.

• 2023.07.15: Persuasion plays a significant role in marketing and computer crimes, influencing decisions and actions. Detecting persuasion is crucial for user protection. In the paper titled "Towards Automatic Principles of Persuasion Detection Using a Machine Learning Approach," Machine Learning is utilized to analyze messages, aiming to identify the most effective data representation and classification algorithm for identifying the principles of persuasion used in phishing attacks. This work has been accepted for presentation at the VIII International Congress on Artificial Intelligence and Pattern Recognition (IWAIPR 2023), to be held in Varadero, Cuba, from September 27 to 29.

• 2022.10.13: Finally, after more than two years, our review of tourism and NLP has been accepted: "Natural Language Processing applied to Tourism Research: A Systematic Review and Future Research Directions." In the words of one of its principal authors and contributor, I thank you all for your efforts, especially Angel Diaz. Without each one of you, this article would not have been possible.  Congrats!!!

• 2022.09.30: Next, October 6th and 7th will be held the MexLEF 2022 conference, organized by the INAOE, IPN, UNAM, and CICESE. At this conference, I will present the overview of Rest-Mex at IberLEF 2022: Recommendation System, Sentiment Analysis and Covid Semaphore Prediction for Mexican Tourist Texts.

• 2022.08.16: MICAI stands for Mexican International Conference on Artificial Intelligence. This year is celebrating its 21st edition, and two papers that describe the research of two colleagues and I were accepted to be presented as Oral Presentations at MICAI. The first paper was entitled: "Red Light/Green Light: a lightweight algorithm for, possibly fraudulent, online behavior change detection," which describes an approach for detecting concept drift that could be caused for malicious activity. The second paper, "Evaluation of a New Representation for Noise Reduction in Distant Supervision," proposes a new data representation for reducing noise in Distant Supervision. Both papers will be published in Springer's Lecture Notes in Artificial Intelligence series. We did a great job here. See you at MICAI 2022!

• 2022.04.24: Our paper "A Lightweight Data Representation for Phishing URLs Detection in IoT Environments" was accepted to be published in the Information Sciences journal (6.795 IF in 2021-2022, ranking it 18 out of 162 in Computer Science). This paper proposes a lightweight data representation that can be effectively used to detect malicious URLs. Furthermore, this data representation is IoT-friendly, which is the main attraction of this research. More about this paper can be found here.

• 2021.12.15: The research project "Phishing attacks detection in electronic messages through Artificial Intelligence techniques" submitted to the Call for Applied Scientific Research and Technological Development 2021 from the Institute for Applied Research and Technology (InIAT) of Iberoamerican University of Mexico City was approved. This project will start in January 2022 and will finish in December 2024. New exciting challenges are coming!

• 2021.11.30: Our new divulgation paper entitled "Generación de una Herramienta Automática para Detectar Diferencias entre Búsquedas de Internet de Destinos Turísticos Dependiendo de la región Geográfica" was accepted for being published in on KIKAME journal.

• 2021.10.28: We will be present in the LKE 2021 conference, to be held on November 04, with the paper "An Autoencoder-based Representation for Noise Reduction in Distant Supervision of Relation Extraction". 

• 2021.10.07: Our review paper entitled: "FPGA/GPU-based Acceleration for Frequent Itemsets Mining: A Comprehensive Review" was finally published in ACM Computing Surveys. This comprehensive research paper covers almost the last 20 years of research efforts in frequent set mining using FPGA and GPUs as development platforms. This was the first and, at the same time, the last paper directly derived from my Ph.D. research.

• 2021.07.16: Congratulations to Liz! Finally, she concluded her MSc project and obtained her Master's degree in Cybersecurity with the project titled: "Design and implementation of a facility access control system through facial and emotion recognition." Great achievement.

• 2021.07.01: Our paper, "A multi-measure feature selection algorithm for efficacious intrusion detection," was published in the Knowledge-Based Systems journal. This was an extremely grateful work (and scientific) experience. It was (and it is) a pleasure to work with such talented researchers. This is the first of (hopeful) many more comings.