Search

Annotated corpora are indispensable tools to train computational models in Artificial Intelligence and Natural Language Processing. However, manual annotation is a costly, arduous, and time-consuming task, especially when the annotation is semantically complex. To address the problem, this work applies a methodology for semi-automatic annotation of datasets based on the Human-in-the-Loop paradigm. The methodology supports the building a resource, that benefits from a fine-grained annotation, to aid in the detection of Spanish violent messages sourced from social media (Twitter/X). After implementing the proposed methodology for semi-automatic violence annotation, a high quality resource was obtained (hereafter referred to as VILLANOS). The methodology consists of annotating the dataset incrementally, which delivers an increase in annotator efficiency, thereby validating the suitability of the proposal. Annotation time was reduced by 52% compared to manual annotation and performance, by training a model with the VILLANOS dataset, obtains an F 1 of 85.2%. These results demonstrate the efficiency and effectiveness of the methodology, evidencing its validity.

The proliferation of hoaxes on the Internet and toxic messages (with offensive or very negative content and high dissemination rates) constitute a current problem on the road to truthful and respectful information. Due to the enormous influence that social media have in the generation of opinion and as a channel of information for society, the efforts that various companies, organisations and institutions are making to detect and counteract the high volume of disinformation circulating on the networks are important. This project deals with the implementation of a proof of concept of a system for analysing the fairness of messages published through social media, built on the basis of various methods and algorithms from human language technologies. These methods and algorithms are the result of research that the participating groups have been working on for the last few years and are promising solutions for the determination of different levels of quality of publications in two fundamental aspects: their veracity and their toxicity. To address the proof of concept, activities aimed at the definition and integration of these technologies and their evaluation by stakeholders are proposed. This will make it possible to establish the responsiveness of these technologies to the needs of society and industry, as well as their viability to work towards higher levels of technological maturity.

Early detection of disinformation is one of the most challenging big-scale problems facing present day society. This is why the application of technologies such as Artificial Intelligence and Natural Language Processing is necessary. The vast majority of Artificial Intelligence approaches require annotated data, and generating these resources is very expensive. This proposal aims to improve the efficiency of the annotation process with a two-level semi-automatic annotation methodology. The first level extracts relevant information through summarization techniques. The second applies a Human-in-the-Loop strategy whereby the labels are pre-annotated by the machine, corrected by the human and reused by the machine to retrain the automatic annotator. After evaluating the system, the average annotation time per news item is reduced by 50%. In addition, a set of experiments on the semi-automatically annotated dataset that is generated are performed so as to demonstrate the effectiveness of the proposal. Although the dataset is annotated in terms of unreliable content, it is applied to the veracity detection task with very promising results (0.95 accuracy in reliability detection and 0.78 in veracity detection).

La llegada de Internet y de las nuevas tecnologías dio lugar al nacimiento de la era de la información, una era que ha conectado a la sociedad de forma global y le ha permitido acceder libremente a la información digital. Con esta facilidad de acceso, cualquier persona, aún sin ser experta en la materia, puede publicar y acceder a la información sin ningún coste, lo que ha ocasionado un exceso de información no contrastada que muchas veces oculta intenciones como el engaño, la manipulación o los fines económicos. De esa forma, la era de la información se ha transformado en la era de la desinformación. La incesante necesidad de estar informados ha motivado que el consumo de la información se convierta en una rutina, ya sea siguiendo las últimas noticias en portales digitales o leyendo a diario publicaciones de personas afines. Antes, la información viajaba en forma de sonido a través de la radio o en forma de tinta a través de los periódicos, pero ahora una desmedida cantidad de información se propaga a través de algoritmos. Las tecnologías han propiciado la sobreabundancia de información, así como la propagación de noticias falsas y bulos, hasta tal punto que resulta imposible contrastar y procesar manualmente tales volúmenes de desinformación en tiempo real. No obstante, lo que se considera un problema puede convertirse en una solución, pues igual que los algoritmos y el entorno digital son los causantes de la viralización de la información falsa, estos pueden ser a su vez los detectores de la desinformación. Es aquí donde el Procesamiento del Lenguaje Natural desempeña un papel clave en la relación humano-máquina, modelando el lenguaje humano a través de la comprensión y generación automática del lenguaje, y entrenando modelos a través de la retroalimentación del experto. El trabajo coordinado entre la ingeniería computacional y la lingüística es decisivo a la hora de frenar el fenómeno de la desinformación. Son necesarias las dos perspectivas para abordar la dete

El objetivo de este trabajo es la creación de una plataforma que permita la monitorización del mercado bursátil y de criptomonedas de manera actualizada. Además, esta plataforma permitirá añadir propuestas en el mercado realizadas por expertos económicos para ver su evolución y si estos suelen acertar en sus pronósticos. Esto por otro lado, con los suficientes datos, podrá ayudar a evitar a aquellos que den consejos sin realmente tener conocimiento, ya que en estos últimos años ha habido un gran aumento de “influencers” que dan consejos sobre esto, pero realmente no tienen idea alguna. Todos estos datos serán proyectados de manera visual para su fácil entendimiento. Todos estos prescriptores, con sus correspondientes propuestas y valores, serán almacenados en una base de datos en la que se almacenarán además las fuentes de donde han sido creadas las recomendaciones, los índices bursátiles y como veremos a continuación, los datos obtenidos de la herramienta que será usada para analizar el efecto de las redes sociales en los valores. Para poder descargar la herramienta utilizada se habilitará un enlace en GitHub, el cual se encontrará en la bibliografía. Por otro lado, se realizará un análisis del efecto de las redes sociales en el mercado bursátil y de criptomonedas mediante una herramienta tecnológica llamada Social Analytics para saber si realmente hay alguna correlación entre ellas y poder determinar si basarse en las redes sociales sería una estrategia determinante para invertir. Este análisis será llevado a cabo con las medidas que dispone esta herramienta, en diferentes periodos de tiempo y podrán ser representados los resultados en la plataforma que ha sido creada con la que realizaré un análisis de los resultados. En este trabajo también se profundizarán sobre algunos ejemplos que han sucedido sobre la influencia de las redes sociales, como el famoso caso de Elon Musk y la criptomoneda DogeCoin que supuso todo un cambio en la cotización del precio de esta. Fi

The CLEAR.TEXT project (TED2021-130707B-I00) researches how natural language processing technology can support the authoring of accessible content in Spanish for people with cognitive disabilities. Our main objective is to research, implement, deploy, evaluate, and ultimately provide robust technologies for natural language processing to support the authoring of accessible Spanish content for public sector organisations (at local, regional and national level) that is intelligible to people with cognitive disability, thereby widening their inclusion and empowerment in Europe. It is expected to impact positively the quality of life of people with cognitive disabilities, facilitating their access to educational, vocational, cultural, and social opportunities in public sector organisations.

Annotated corpora are indispensable tools to train computational models in Natural Language Processing. However, in the case of more complex semantic annotation processes, it is a costly, arduous, and time-consuming task, resulting in a shortage of resources to train Machine Learning and Deep Learning algorithms. In consideration, this work proposes a methodology, based on the human-in-the-loop paradigm, for semi-automatic annotation of complex tasks. This methodology is applied in the construction of a reliability dataset of Spanish news so as to combat disinformation and fake news. We obtain a high quality resource by implementing the proposed methodology for semi-automatic annotation, increasing annotator efficacy and speed, with fewer examples. The methodology consists of three incremental phases and results in the construction of the RUN dataset. The annotation quality of the resource was evaluated through time-reduction (annotation time reduction of almost 64% with respect to the fully manual annotation), annotation quality (measuring consistency of annotation and inter-annotator agreement), and performance by training a model with RUN semi-automatic dataset (Accuracy 95% F1 95%), validating the suitability of the proposal.

The development of the internet and digital technologies has inadvertently facilitated the huge disinformation problem that faces society nowadays. This phenomenon impacts ideologies, politics and public health. The 2016 US presidential elections, the Brexit referendum, the COVID-19 pandemic and the Russia-Ukraine war have been ideal scenarios for the spreading of fake news and hoaxes, due to the massive dissemination of information. Assuming that fake news mixes reliable and unreliable information, we propose RUN-AS (Reliable and Unreliable Annotation Scheme), a fine-grained annotation scheme that enables the labelling of the structural parts and essential content elements of a news item and their classification into Reliable and Unreliable. This annotation proposal aims to detect disinformation patterns in text and to classify the global reliability of news. To this end, a dataset in Spanish was built and manually annotated with RUN-AS and several experiments using this dataset were conducted to validate the annotation scheme by using Machine Learning (ML) and Deep Learning (DL) algorithms. The experiments evidence the validity of the annotation scheme proposed, obtaining the best F1m, 0.948, with the Decision Tree algorithm.

La sociedad avanza cargada de conocimientos nuevos y muy accesibles, que se publican en el mundo virtual. Es una realidad que las Tecnologías de la Información y la Comunicación (TIC) han traído muchos beneficios a nuestras vidas pero también vemos como año tras año aumenta el uso de violencia en plataformas digitales. Nuestro trabajo se enfoca en la creación de recursos que permitan la detección de mensajes violentos en la red social Twitter. Se parte de la creación de una guía de anotación de grano fino para anotar un corpus de mensajes violentos (VIL) con el fin de utilizar herramientas de aprendizaje automático que nos ayuden a detectar automáticamente el problema. Con este corpus se entrenan dos modelos de lenguaje (BETO y RoBERTa base) con los que se alcanza un valor en la métrica F1m de 97.03% y 96.51% clasificando si un tuit es o no violento., Society is moving forward full of new and very accessible knowledge, which is published in the virtual world. It is a reality that ICTs have brought many benefits to our lives but we also see how year after year the use of violence on digital platforms increases. Our work focuses on the detection of violent messages in the social network Twitter. Starting from the creation of a fine-grained annotation guide to obtain a corpus of violent messages (VIL) in order to use Machine Learning tools that help us to automatically detect the problem Two language models are trained with this corpus (BETO and RoBERTa base) with which a value of 97.03% and 96.51% is reached in the F1m metric, classifying whether or not a tweet is violent.

Disinformation is a critical problem in our society. The COVID-19 pandemic and the Russia-Ukraine war have been key events for the spreading of fake news. Assuming that fake news mixes reliable and unreliable information, we propose RUN-AS (Reliable and Unreliable Annotation Scheme), a fine-grained annotation scheme that labels the structural parts and essential content elements of a news item to enable their classification into Reliable and Unreliable. This type of annotation will be used for training systems to automatically classify the reliability of a news item. To this end, RUN dataset in Spanish was built and annotated with RUN-AS. A set of experiments were conducted to validate the annotation scheme. The experiments evidence the validity of the annotation scheme proposed, obtaining the best F1m, i.e., 0.948., La desinformación es un problema crítico en nuestra sociedad. La pandemia de covid-19 y la guerra entre Rusia y Ucrania han sido escenarios clave para la difusión de noticias falsas. Partiendo de la base de que las noticias falsas mezclan información confiable y no confiable, proponemos RUN-AS (Reliable and Unreliable Annotation Scheme), un esquema de anotación de grano fino que etiqueta las partes estructurales y los elementos de contenido esenciales de una noticia y permite clasificarlos en Confiable y No confiable. Esta anotación será usada en el entrenamiento de sistemas para la clasificación automática de la confiabilidad de una noticia. Para ello, se construyó el corpus RUN en español y se anotó con RUN-AS. Se llevó a cabo un conjunto de experimentos para validar el esquema de anotación. Los experimentos evidencian la validez del esquema de anotación propuesto, obteniendo el mejor F1m 0,948.

El modelado de skipgrams es una técnica para la generación de términos multi-palabra que conserva parte de la secuencialidad y flexibilidad del lenguaje. Sin embargo, en algunos casos el número de skipgrams generados puede ser excesivo a medida que se aumenta la distancia entre palabras. Además, esta distancia no suele ser tenida en cuenta a la hora de valorar los términos que se generan. En este trabajo proponemos una técnica para la generación y filtrado eficientes de skipgrams y un esquema de pesado que tiene en cuenta la distancia entre los términos, dando más importancia a aquellos más cercanos. Aplicaremos y evaluaremos estas propuestas en la tarea de análisis de sentimientos., Skipgram modelling is a technique for generating multi-word terms that preserves some of the sequentiality and flexibility of the language. However, in some cases the number of skipgrams generated may become excessive as the distance between words increases. Moreover, this distance is often not taken into account when evaluating the terms that are generated. In this paper we propose a technique for efficient skipgram generation and filtering, and a weighing scheme that takes into account the distance between terms, giving more importance to those closer. We will apply and evaluate these proposals in the task of sentiment analysis.

Discovering the main features of virality patterns in Twitter is the focus of this research. Five trending topics related to the COVID-19 pandemic were selected for the study, with Spanish as the target language. To carry out the discovery of virality patterns, we applied opinion mining techniques that enable us to structure the information based on the polarity of the messages and the emotions they contain. After transforming the information from an unstructured textual representation to a structured one, data mining techniques were applied, specifically association rules mining. Message patterns with the highest virality (high shares and high likes), and at the same time the most relevant characteristics of the patterns with less impact were extracted. After an exhaustive analysis of the most relevant non-redundant rules, it can be concluded that messages with a high-negative polarity and a very high emotional charge, especially emotions that have intensified with the COVID-19 pandemic, such as fear, sadness, anger and surprise are more likely to go viral in social media. By contrast, messages with little news coverage in the media, few authors, and the absence of surprise are relevant features when it comes to seeing messages with very low dissemination in social media.

El proyecto tiene como finalidad el desarrollo del sistema de información y su posterior aplicación en una plataforma web, de un sistema para el control de la venta al por mayor de productos alimenticios entre el comercio en diferentes establecimientos, como supermercados y puestos en mercados, así como a particulares.

This project pursues the dynamic modeling at a spatial-temporal level of digital entities in social media for predicting their behavior. Firstly, digital entities are modelled by identifying the characteristics of individuals through their language and footprint on the network. Then, the extraction of relationships between digital entities is one of the nuclear challenges of the project. The proposal pursues this objective on a semantic level, structuring the information into representations of knowledge suitable for logical processing. Considering the heterogeneous nature of the sources to be dealt with, filtering of information is fundamental, using metrics and quality criteria. This spatial-temporal characterization, together with screening processes, will allow us to study high-performance predictive strategies in the evolution of digital entities. This project is coordinated by the SINAI and GPLSI research groups.