News

Con enorme pesar, comunicamos el fallecimiento de  nuestro compañero, el profesor Gregorio Fernández Fernández.

Gregorio Fernández nació en Jaén en 1943. Realizó sus estudios de Ingeniero de Telecomunicación (UPM, 1969), Ingénieur en Automatique (Universidad Paul Sabatier, Toulouse, 1971) y Doctor Ingeniero de Telecomunicación (UPM, 1975). Ha sido  Catedrático de Universidad en el Departamento de Ingeniería de Sistemas Telemáticos desde 1983. Ha diseñado, coordinado e impartido asignaturas de grado relacionadas con la informática y la telemática. También ha impartido cursos de postgrado sobre Informática médica, Programación lógica, Sistemas expertos, Bases de datos deductivas, Aprendizaje automático, Minería de datos y Agentes inteligentes, en diversas empresas y universidades europeas y latinoamericanas. Su actividad investigadora se desarrolla en las áreas de bioingeniería e inteligencia artificial, habiendo dirigido o participado como investigador o asesor en un gran número de proyectos de I+D y dirigido seis tesis doctorales. Es autor de siete libros, coautor de cinco, coeditor de dos y traductor de cuatro, además de ser autor o coautor de veinte capítulos de libros y artículos en revistas científicas y numerosas comunicaciones a congresos.

A lo largo de su trayectoria, ha destacado su actividad docente, con la publicación de libros en Fundamentos de Ordenadores, Lógica, habitualmente en colaboración con el profesor Fernando Sáez Vacas. Uno de los logros docentes más populares fue la invención de dos ordenadores didácticos, Simplez y Algoritmez, que ofrecían un enfoque pedagógico innovador para la comprensión de los fundamentos de los ordenadores, y ha sido libro de texto de referencia en la mayoría de escuelas de telecomunicación de España.

En el plano investigador, Gregorio ha sido pionero en la bioingeniería en que realizó su tesis doctoral, así como en la Inteligencia Artficial en España, colaborando con el profesor José Cuena, y siendo el fundador y director del Grupo de Sistemas Inteligentes de la  Universidad Politécnica de Madrid. Como investigador y docente, deja una larguísima lista de discípulos, distribuidos por universidades y empresas de España.

Decía Henry Adams, que el maestro deja una huella para la eternidad; nunca puede decir cuándo se detiene su influencia. Las enseñanzas de Gregorio siguen vivas en sus alumnos, que recuerdan cómo ha combinado el rigor académico con el sentido de humor, el buen uso del español, y la motivación para ampliar los conocimientos recibidos.

Queremos expresar en este obituario nuestra gratitud profunda a Gregorio, y  un abrazo afectuoso a su familia.

The article Contextualization of a Radical Language Detection System Through Moral Values and Emotions has been recently published in the IEEE Access journal (JCR Q2 2022, 3.9 IF). The publicacion is authored by Patricia Alonso and Oscar Araque. This work was supported by the European Union’s Horizon 2020 Research and Innovation Programme under Grant 962547.

The full paper can be found here.

 

Abstract:

The popularity of current communication technologies has boosted the spread of polarization and radical ideologies, which can be exploited by terrorist organizations. Building upon previous research, this work focuses on the task of automatic radicalization detection in texts using natural language processing and machine learning techniques. In this way, we investigate the effectiveness of integrating moral values through the Moral Foundations Theory (MFT). Moral values play a crucial role in identifying ideological inclinations and can have a significant impact on the radicalization detection task. Our approach distinguishes itself in the feature extraction stage, leveraging moral values, emotions, and similarity-based features that utilize word embeddings. Additionally, we thoroughly evaluate the proposed representations with three distinct datasets that model radicalization and use the SHAP method to gain relevant insight into the models’ reasoning.

GSI is participating in the final conference of the project PARTICIPATION in Rome. The conference showcases the innovative and participatory methods and tools that the project has developed and tested for analysing and preventing radicalization and violent extremism in various contexts and countries over the past three years. The conference has two main themes: conspiracy theories and radicalization trends, and the role of civil society and institutions in preventing violent extremism. 

The article "Detection of the Severity Level of Depression Signs in Text Combining a Feature-Based Framework with Distributional Representations ", by Sergio Muñoz and Carlos A. Iglesias has been published in the Applied Sciences journal (2.7 impact factor, JCR Q2 2022). This work is a product of the MIRATAR and AMOR projects.

The full paper can be found at this URL.

 

Abstract:

Depression is a common and debilitating mental illness affecting millions of individuals, diminishing their quality of life and overall well-being. The increasing prevalence of mental health disorders has underscored the need for innovative approaches to detect and address depression. In this context, text analysis has emerged as a promising avenue. Novel solutions for text-based depression detection commonly rely on deep neural networks or transformer-based models. Although these approaches have yielded impressive results, they often come with inherent limitations, such as substantial computational requirements or a lack of interpretability. This work aims to bridge the gap between substantial performance and practicality in the detection of depression signs within digital content. To this end, we introduce a comprehensive feature framework that integrates linguistic signals, emotional expressions, and cognitive patterns. The combination of this framework with distributional representations contributes to fostering the understanding of language patterns indicative of depression and provides a deeper grasp of contextual nuances. We exploit this combination using traditional machine learning methods in an effort to yield substantial performance without compromising interpretability and computational efficiency. The performance and generalizability of our approach have been assessed through experimentation using multiple publicly available English datasets. The results demonstrate that our method yields throughput on par with more complex and resource-intensive solutions, achieving F1-scores above 70%. This accomplishment is notable, as the proposed method simultaneously preserves the virtues of simplicity, interpretability, and reduced computational overhead. In summary, the findings of this research contribute to the field by offering an accessible and scalable solution for the detection of depression in real-world scenarios.