1 |
Reescrita sentencial baseada em traços de personalidade / Text rewrite based on psychological profilesStávracas Neto, Georges Basile 13 March 2018 (has links)
Sistemas de Geração de Língua Natural tentam produzir textos de maneira automatizada. Em sistemas desse tipo, é desejável produzir textos de forma realista - ou psicologicamente plausível - como forma de aumentar o engajamento do leitor. Uma das formas de alcançar esse objetivo é gerando textos de modo a refletir uma personalidade-alvo de interesse. Por exemplo, uma pessoa extrovertida usaria palavras mais simples e seus textos teriam mais interjeições e traços de oralidade. Esse trabalho tem o objetivo de desenvolver um modelo de reescrita sentencial para o português brasileiro com base em traços de personalidade de um locutor-alvo. Para isso, foi coletado um córpus de textos e inventários de personalidade e, com base em uma análise preliminar desses dados, foram encontrados fortes indícios de correlação entre os fatores de personalidade e as características observadas dos textos em português brasileiro. Foram gerados três modelos de lexicalização, referentes à adjetivos, substantivos e verbos. Esses modelos de lexicalização, então, foram utilizados na proposta de um modelo de reescrita sentencial para selecionar as palavras mais adequadas à personalidade-alvo. Os resultados demonstram que o uso de personalidade permite que o texto gerado seja mais próximo do desempenho humano se comparado a um sistema de baseline que faz escolhas lexicais mais frequentes / Natural Language Generation Systems attempt to produce texts in an automated fashion. In systems of this kind, it is desired to produce texts realisticaly - or at least psychologically plausible - as a way to increase reader\'s engagement. One way to achieve this goal is generating texts in such a way to reflect a target personality profile. For example, an extroverted individual would use simpler words and its texts would have more interjections and orality traces. This work proposes the development of a Brazilian Portuguese personality-based sentence rewrite model. To this end, a corpus with text samples and personality inventories has been collected, and, based on a preliminary analysis, strong correlations between personality and text features have been found. Three lexicalization models were generated, related to adjectives, nouns and verbs. These models were then used by the sentence rewrite model to select the most appropriate word for the target personality. Results show that the usage of personality allows the generated text to be closer to human performance when compared to a baseline system that makes lexical choices based on frequency
|
2 |
Reescrita sentencial baseada em traços de personalidade / Text rewrite based on psychological profilesGeorges Basile Stávracas Neto 13 March 2018 (has links)
Sistemas de Geração de Língua Natural tentam produzir textos de maneira automatizada. Em sistemas desse tipo, é desejável produzir textos de forma realista - ou psicologicamente plausível - como forma de aumentar o engajamento do leitor. Uma das formas de alcançar esse objetivo é gerando textos de modo a refletir uma personalidade-alvo de interesse. Por exemplo, uma pessoa extrovertida usaria palavras mais simples e seus textos teriam mais interjeições e traços de oralidade. Esse trabalho tem o objetivo de desenvolver um modelo de reescrita sentencial para o português brasileiro com base em traços de personalidade de um locutor-alvo. Para isso, foi coletado um córpus de textos e inventários de personalidade e, com base em uma análise preliminar desses dados, foram encontrados fortes indícios de correlação entre os fatores de personalidade e as características observadas dos textos em português brasileiro. Foram gerados três modelos de lexicalização, referentes à adjetivos, substantivos e verbos. Esses modelos de lexicalização, então, foram utilizados na proposta de um modelo de reescrita sentencial para selecionar as palavras mais adequadas à personalidade-alvo. Os resultados demonstram que o uso de personalidade permite que o texto gerado seja mais próximo do desempenho humano se comparado a um sistema de baseline que faz escolhas lexicais mais frequentes / Natural Language Generation Systems attempt to produce texts in an automated fashion. In systems of this kind, it is desired to produce texts realisticaly - or at least psychologically plausible - as a way to increase reader\'s engagement. One way to achieve this goal is generating texts in such a way to reflect a target personality profile. For example, an extroverted individual would use simpler words and its texts would have more interjections and orality traces. This work proposes the development of a Brazilian Portuguese personality-based sentence rewrite model. To this end, a corpus with text samples and personality inventories has been collected, and, based on a preliminary analysis, strong correlations between personality and text features have been found. Three lexicalization models were generated, related to adjectives, nouns and verbs. These models were then used by the sentence rewrite model to select the most appropriate word for the target personality. Results show that the usage of personality allows the generated text to be closer to human performance when compared to a baseline system that makes lexical choices based on frequency
|
3 |
Extractive Multi-document Summarization of News ArticlesGrant, Harald January 2019 (has links)
Publicly available data grows exponentially through web services and technological advancements. To comprehend large data-streams multi-document summarization (MDS) can be used. In this research, the area of multi-document summarization is investigated. Multiple systems for extractive multi-document summarization are implemented using modern techniques, in the form of the pre-trained BERT language model for word embeddings and sentence classification. This is combined with well proven techniques, in the form of the TextRank ranking algorithm, the Waterfall architecture and anti-redundancy filtering. The systems are evaluated on the DUC-2002, 2006 and 2007 datasets using the ROUGE metric. Where the results show that the BM25 sentence representation implemented in the TextRank model using the Waterfall architecture and an anti-redundancy technique outperforms the other implementations, providing competitive results with other state-of-the-art systems. A cohesive model is derived from the leading system and tried in a user study using a real-world application. The user study is conducted using a real-time news detection application with users from the news-domain. The study shows a clear favour for cohesive summaries in the case of extractive multi-document summarization. Where the cohesive summary is preferred in the majority of cases.
|
4 |
Generating and simplifying sentences / Génération et simplification des phrasesNarayan, Shashi 07 November 2014 (has links)
Selon la représentation d’entrée, cette thèse étudie ces deux types : la génération de texte à partir de représentation de sens et à partir de texte. En la première partie (Génération des phrases), nous étudions comment effectuer la réalisation de surface symbolique à l’aide d’une grammaire robuste et efficace. Cette approche s’appuie sur une grammaire FB-LTAG et prend en entrée des arbres de dépendance peu profondes. La structure d’entrée est utilisée pour filtrer l’espace de recherche initial à l’aide d’un concept de filtrage local par polarité afin de paralléliser les processus. Afin nous proposons deux algorithmes de fouille d’erreur: le premier, un algorithme qui exploite les arbres de dépendance plutôt que des données séquentielles et le second, un algorithme qui structure la sortie de la fouille d’erreur au sein d’un arbre afin de représenter les erreurs de façon plus pertinente. Nous montrons que nos réalisateurs combinés à ces algorithmes de fouille d’erreur améliorent leur couverture significativement. En la seconde partie (Simplification des phrases), nous proposons l’utilisation d’une forme de représentations sémantiques (contre à approches basées la syntaxe ou SMT) afin d’améliorer la tâche de simplification de phrase. Nous utilisons les structures de représentation du discours pour la représentation sémantique profonde. Nous proposons alors deux méthodes de simplification de phrase: une première approche supervisée hybride qui combine une sémantique profonde à de la traduction automatique, et une seconde approche non-supervisée qui s’appuie sur un corpus comparable de Wikipedia / Depending on the input representation, this dissertation investigates issues from two classes: meaning representation (MR) to text and text-to-text generation. In the first class (MR-to-text generation, "Generating Sentences"), we investigate how to make symbolic grammar based surface realisation robust and efficient. We propose an efficient approach to surface realisation using a FB-LTAG and taking as input shallow dependency trees. Our algorithm combines techniques and ideas from the head-driven and lexicalist approaches. In addition, the input structure is used to filter the initial search space using a concept called local polarity filtering; and to parallelise processes. To further improve our robustness, we propose two error mining algorithms: one, an algorithm for mining dependency trees rather than sequential data and two, an algorithm that structures the output of error mining into a tree to represent them in a more meaningful way. We show that our realisers together with these error mining algorithms improves on both efficiency and coverage by a wide margin. In the second class (text-to-text generation, "Simplifying Sentences"), we argue for using deep semantic representations (compared to syntax or SMT based approaches) to improve the sentence simplification task. We use the Discourse Representation Structures for the deep semantic representation of the input. We propose two methods: a supervised approach (with state-of-the-art results) to hybrid simplification using deep semantics and SMT, and an unsupervised approach (with competitive results to the state-of-the-art systems) to simplification using the comparable Wikipedia corpus
|
Page generated in 0.0552 seconds