Now showing items 1-20 of 42

    • A pipeline for large raw text preprocessing and model training of language models at scale 

      Armengol Estapé, Jordi (Universitat Politècnica de Catalunya, 2021-01-25)
      Master thesis
      Open Access
      Covenantee:   Universitat de Barcelona / Universitat Rovira i Virgili
      The advent of Transformer-based (i.e., based on self-attention architectures) language models has revolutionized the entire field of Natural Language Processing (NLP). Once pre-trained on large, unlabelled corpora, we can ...
    • A tool for automatic evaluation of human translation quality within a mooc environment 

      Betanzos Atienza, Miguel (Universitat Politècnica de Catalunya, 2015-10)
      Master thesis
      Open Access
      Descripción del proceso de creación de un corpus de traducciones a través de un curso ofrecido en la plataforma openEdX, y su posterior análisis a fin de entrenar un modelo de evaluación para traducciones similares que ...
    • Abstractive text summarization with attention-based mechanism 

      Sanjabi, Nima (Universitat Politècnica de Catalunya, 2018-04)
      Master thesis
      Open Access
      In this work, we explore the evolution of Sequential Neural Models, and their use as a Summarizer System. Transformer is a recently proposed model with a high potential. We experiment and compare their result in abstractive ...
    • Adversarial strategies for Reducing Gender Bias in Neural Machine Translation 

      Burgos Preciado, Julio (Universitat Politècnica de Catalunya, 2020-09)
      Master thesis
      Open Access
      In a more connected world, communication between different native speakers has became more necessary. This make that translation systems become more useful. In the last years, typical translation systems have evolved towards ...
    • Anàlisi de sentiment per a textos curts en català i castellà aprofitant dades no supervisades 

      Navarrete Jimenez, Daniel (Universitat Politècnica de Catalunya, 2021-01-24)
      Bachelor thesis
      Open Access
      There may be a lot of abusive behaviour in conversations between teenagers, which take place through social media. In this project, we develop classifiers to find out which texts present abuse such as violence, sexual ...
    • Chinese-Catalan neural machine translation with OpenNMT 

      Wang, Chaofeng (Universitat Politècnica de Catalunya, 2018-07)
      Bachelor thesis
      Restricted access - author's decision
    • Coverage model for character-based neural machine translation 

      Kazimi, Mohammad Bashir (Universitat Politècnica de Catalunya, 2017-05)
      Master thesis
      Open Access
      In recent years, Neural Machine Translation (NMT) has achieved state-of-the art performance in translating from a language; source language, to another; target language. However, many of the proposed methods use word ...
    • Deep Learning Based Textual Metadata Matcher 

      Cabrera Fernández, Francesc-Carles (Universitat Politècnica de Catalunya, 2019-10-23)
      Bachelor thesis
      Restricted access - confidentiality agreement
    • Determining Bias in Machine Translation with Deep Learning Techniques 

      Escudé Font, Joel (Universitat Politècnica de Catalunya, 2019-01)
      Master thesis
      Open Access
      The presence of biases in artificial intelligence is arising as a social challenge. In the particular application of machine translation, when you translate a sentence to a non-gender neutral language like Spanish, from a ...
    • Efficient transformers for direct speech translation 

      Alastruey Lasheras, Belén (Universitat Politècnica de Catalunya, 2021-07)
      Bachelor thesis
      Open Access
      In this thesis, we propose a new approach for Speech-to-Text translation, where thanks to an efficient Transformer we can work with a spectrogram without having to use convolutional layers before the Transformer. This ...
    • End-to-end speech translation system with attention-based mechanisms 

      Cros Vila, Laura (Universitat Politècnica de Catalunya, 2018-06)
      Bachelor thesis
      Open Access
      Speech Recognition and Text-to-Text Translation systems have been improving significantly in recent decades thanks to the improvement of both hardware and software means. However, Speech Translation is usually done as a ...
    • End-to-end Speech Translation with Self-supervised Speech Representations 

      Gallego Olsina, Gerard Ion (Universitat Politècnica de Catalunya, 2020-09-09)
      Master thesis
      Open Access
      Nowadays, there is a growing interest in the field of Speech Translation (speech-to-text). Traditionally, this task has been faced with the concatenation of Automatic Speech Recognition and Machine Translation modules. ...
    • Explorando factores de riesgo de insuficiencia cardíaca a través del aprendizaje automático 

      Pérez Soria, Beatriz (Universitat Politècnica de Catalunya, 2019-02)
      Bachelor thesis
      Open Access
      Acute myocardial infarction is one of the main causes of mortality in developed countries. There are many risk factors that can trigger a heart attack that have to do with our lifestyle today. One of the most complicated ...
    • Exploratori visual analysis of machine translaton systems 

      Lacroux, Elora (Universitat Politècnica de Catalunya, 2019-07-04)
      Master thesis
      Open Access
      This project aims to provide visualisation tools for helping researchers to understand their model. We are now able to visualise multilingual intermediate representations. We can also visualise multilingual representations ...
    • Exploring Automatic Speech Recognition with TensorFlow 

      Escur i Gelabert, Janna (Universitat Politècnica de Catalunya, 2018)
      Bachelor thesis
      Open Access
      Speech Recognition (reconocimiento de voz) es la tarea que pretende indentificar palabras habladas y convertirlas a texto. Este trabajo de fin de grado se centra en utilizar técnicas de deep learning para construir un ...
    • Exploring reinforcement learning in natural language processing 

      Muntaner González, Joan Francesc (Universitat Politècnica de Catalunya, 2021-04-30)
      Bachelor thesis
      Open Access
      El Procesament del Llenguatge Natural és el conjunt de tasques que tracten amb el llenguatge "humà". En aquest treball ens centrem en la traducció automàtica, una d'aquestes tasques, que consisteix en traduir automàticament ...
    • GeBioToolkit automatic extraction of gender-balanced multilingual corpus of wikipedia biographies 

      Li Lin, Pau (Universitat Politècnica de Catalunya, 2020-01-31)
      Master thesis
      Open Access
      We present GeBioToolkit, an automatic tool for extracting multilingual parallel corpora at sentence level, with document and gender information from Wikipedia biographies. Despite the gender inequalities present on ...
    • Gender bias in natural language processing: BioCorpus-5, a preliminary multilingual Gender-Balanced Corpus of in-domain wikipedia biographies 

      Kim Jung, Jae Hyouk (Universitat Politècnica de Catalunya, 2019-01)
      Bachelor thesis
      Open Access
      In natural language processing and the blind application of machine learning reflect social biases and stereotypes in training data. In this project, we develop a corpus for future analysis applications of this bias. The ...
    • Gender bias in neural machine translation: The case of indian languages 

      Giri, Kesudh (Universitat Politècnica de Catalunya, 2020-06)
      Bachelor thesis
      Open Access
      La traducció automàtica implica la traducció de text d'un idioma a un altre amb ajuda d'un sistema automàtic. En concret, la traducció automàtica neuronal implica l'ús d'una arquitectura basada en xarxes neuronals. Aquest ...
    • Heart Failure Factors: a database approach 

      Gállego Olsina, Gerard Ion (Universitat Politècnica de Catalunya, 2019-03)
      Bachelor thesis
      Open Access
      This project aims to find relationships between psychological stress factors and heart attacks that took place in Catalunya between 2010 and 2016. We have measured these factors through the news that were published in La ...