The paper GSI-UPM at SemEval-2019 Task 5: Semantic Similarity and Word Embeddings for Multilingual detection of Hate speech against Immigrants and Women on Twitter, by Diego Benito, Óscar Araque, and Carlos A. Iglesias has been published at the Thirteenth International Workshop on Semantic Evaluation (SemEval-2019).

The SemEval workshop focuses on the evaluation and comparison of systems that can alyse diverse semantic phenomena in text with the aim of extending the current state of the art in semantic analysis and creating high quality annotated datasets in a range of increasingly challenging problems in natural language semantics. In particular, SemEval-2019 task 5 aims at detecting hate speech featured by two specific different targets, immigrants and women, in a multilingual perspective, for Spanish and English.

The publication represents the first major achievement of the Intelligent Systems Group in the field of hate speech, reflected in an honorable fifth position in the Spanish sub-task A and in the development of the best European system in the same sub-task.

Abstract. This paper describes the GSI-UPM system for SemEval-2019 Task 5, which tackles multilingual detection of hate speech on Twitter. The main contribution of the paper is the use of a method based on word embeddings and semantic similarity combined with traditional paradigms, such as n-grams, TF-IDF and POS. This combination of several features is fine-tuned through ablation tests, demonstrating the usefulness of different features. While our approach outperforms baseline classifiers on different sub-tasks, the best of our submitted runs reached the 5th position on the Spanish sub-task A.

The SemEval-2019 workshop was held June 6-7, 2019 in Minneapolis, USA, collocated with the Annual Conference of the North American Chapter of the Association for Computational Linguistics: Human Language Technologies (NAACL-HLT 2019).

Mañana viernes 7 de junio y el lunes 10 de junio, varios miembros del GSI defienden sus trabajos fin de carrera, estáis todos invitados.


Viernes 7 de junio, B225 - TFGs

  • 9:00 Alberto Mújica Ayuso,  Design and implementation of an agent-based social simulation of binge drinking among Spanish youngsters
  • 9:20 Luis Cristóbal Mújica, Development of an Event Detector in Twitter Streams based on Mention-Anomaly Detection for the City of Madrid
  • 9:40 Álvaro de Pablo, Design and Development of a Stylometry Library for Texts in Spanish and English. Application to Terrorist and Radical Texts.
  • 10:00 Sergio Gil, Design and implementation of an API Gateway based on NodeJS and Express
  • 10:20 Luis Martín de Vidales, Design and development of a diabetes and overweight machine learning classifier using social media mining
  • 10:40 Daniel Carlander-Reuterfelt, Development of a Cognitive Bot for Data Science tutoring based on a Big Data Natural Language Analytics Platform
  • 11:00 Alberto Llópez Santiago, Bicycle traffic using OpenStreetMap routing
  • 11:20 Álvaro Garnica, Development of a Data visualization Application for an IoT platform using Javascript Frameworks


Lunes 10 de junio, B225 -  TFMs MUIT

  • 10:50 Alberto Sánchez López, Development of a Food Image Classification System based on Transfer Learning with Convolutional Neural Networks
  • 11:15 Diego Benito, Design and Development of a Hate Speech Detector in Social Networks Based on Deep Learning Technologies.
  • 11:40 Pablo Aznar, Development of a Deep Learning based Attack Detection System for Smart Grids 

Ayer 27 de mayo se celebró en la ETSI Telecomunicación de la UPM un acto de homenaje "in memoriam" al profesor Fernando Sáez Vacas. Al inicio del acto se inauguró una placa para dar nombre a la ciberteca de la Escuela, ciberteca Fernando Sáez Vacas.


El acto fue presidido por el Rector Guillermo Cisneros y organizado por la 47ª Promoción, que ha estado ligada a Fernando, dado que esta promoción inició la andadura de Satelec y Fernando les acompañó en el viaje fin de carrera, manteniendo reuniones desde entonces.

En el acto se repasó la trayectoria académica de Fernando. Dos doctorandos y colaboradores de Fernando, los profesores Gregorio Fernández y Gonzalo León comentaron diversos aspectos de su personalidad, como su capacidad y empeño en redactar textos docentes empleando un lenguaje preciso y claro, su visión, y su capacidad de transferencia tecnológica con la empresa, proponiendo tesis empresariales. También participaron dos alumnos,  Mariano  González, director general de Bull, y  Mateo Valero, director del Barcelona Supercomputing Center y reciente premio Charles Babbage 2017, que comentaron su labor como profesor y mentor. Finalmente, Ángel Mayoral, colaborador de Fernando en Bull, y Paco Román, expresidente de Vodafone España y presidente de fundación SERES, comentaron su labor profesional, ligada a la formación empresarial en Bull, y su capacidad de reflexionar sobre las implicaciones sociales de la tecnología. Durante las intervenciones, también se destacó cómo Fernando impulsó diversas asociaciones de estudiantes, como el IEEE, IAESTE o Satelec.


El artículo  A semantic data lake framework for autonomous fault management in SDN environments, por Fernando Benayas, Álvaro Carrera, Carlos A. Iglesias y Manuel García Amado has sido aceptado y publicado en la revista Transactions on Emerging Telecommunications Technologies. Esta revista se encuentra indexada en JCR: (Q3, 1.61).


Fault Management is a vital issue for any network operator since the beginning of the telecommunications era. As networks have become more and more complex, their management systems are crucial for any operator company. In this ecosystem, the Software‐Defined Networking (SDN) approach has appeared as a possible solution for different networking issues. The flexibility provided by SDN to the network management enables a great dynamism in the configuration of network devices. However, this feature introduces the cost of a potential increase in failures because every modification introduced on the control plane is a new possibility for failures to appear and cause a decrement of the quality for offered services. Because of the growing pace of the networks, the classical approach is not feasible to cope that dynamism. Increasing the number of human operators in charge of the fault management process would increase the operation cost dramatically. Thus, this paper presents an approach to apply machine learning over a big data framework for an autonomous fault management process in SDN networks. In this paper, we present a Semantic Data Lake framework for a self‐diagnosis service, which is deployed on top of an SDN management platform. Moreover, we have developed a prototype of the proposed service with different diagnosis models for SDN networks. Models and algorithms have been evaluated showing good results.



Benayas F, Carrera Á, García-Amado M, Iglesias CA. A semantic data lake framework for autonomous fault management in SDN environments. Trans Emerging Tel Tech. 2019; e3629. https: //


Green Open Access: