Text plays a pivotal role in our daily lives, encompassing various forms such as social media posts, news articles, books, reports, and more. Consequently, Natural Language Processing (NLP) has garnered widespread attention. This technology empowers us to undertake tasks like text classification, entity recognition, and even crafting responses within a dialogue context. However, despite the expansive utility of NLP, it frequently necessitates a critical decision: whether to place trust in a model's predictions. To illustrate, consider a state-of-the-art (SOTA) model entrusted with diagnosing a disease or assessing the veracity of a rumor. An incorrect prediction in such scenarios can have dire consequences, impacting individuals' health or tarnishing their reputation. Consequently, it becomes imperative to establish a reliable method for evaluating the reliability of an NLP model's predictions, which is our focus-uncertainty estimation on NLP. Though many works have researched uncertainty estimation or NLP, the combination of these two domains is rare. This is because most NLP research emphasizes model prediction performance but tends to overlook the reliability of NLP model predictions. Additionally, current uncertainty estimation models may not be suitable for NLP due to the unique characteristics of NLP tasks, such as the need for more fine-grained information in named entity recognition. Therefore, this dissertation proposes novel uncertainty estimation methods for different NLP tasks by considering the NLP task's distinct characteristics. The NLP tasks are categorized into natural language understanding (NLU) and natural language generation (NLG, such as text summarization). Among the NLU tasks, the understanding could be on two views, global-view (e.g. text classification at document level) and local-view (e.g. natural language inference at sentence level and named entity recognition at token level). As a result, we research uncertainty estimation on three tasks: text classification, named entity recognition, and text summarization. Besides, because few-shot text classification has captured much attention recently, we also research the uncertainty estimation on few-shot text classification. For the first topic, uncertainty estimation on text classification, few uncertainty models focus on improving the performance of text classification where human resources are involved. In response to this gap, our research focuses on enhancing the accuracy of uncertainty scores by bolstering the confidence associated with winning scores. we introduce MSD, a novel model comprising three distinct components: 'mix-up,' 'self-ensembling,' and 'distinctiveness score.' The primary objective of MSD is to refine the accuracy of uncertainty scores by mitigating the issue of overconfidence in winning scores while simultaneously considering various categories of uncertainty. seamlessly integrate with different Deep Neural Networks. Extensive experiments with ablation settings are conducted on four real-world datasets, resulting in consistently competitive improvements. Our second topic focuses on uncertainty estimation on few-shot text classification (UEFTC), which has few or even only one available support sample for each class. UEFTC represents an underexplored research domain where, due to limited data samples, a UEFTC model predicts an uncertainty score to assess the likelihood of classification errors. However, traditional uncertainty estimation models in text classification are ill-suited for UEFTC since they demand extensive training data, while UEFTC operates in a few-shot scenario, typically providing just a few support samples, or even just one, per class. To tackle this challenge, we introduce Contrastive Learning from Uncertainty Relations (CLUR) as a solution tailored for UEFTC. CLUR exhibits the unique capability to be effectively trained with only one support sample per class, aided by pseudo uncertainty scores. A distinguishing feature of CLUR is its autonomous learning of these pseudo uncertainty scores, in contrast to previous approaches that relied on manual specification. Our investigation of CLUR encompasses four model structures, allowing us to evaluate the performance of three commonly employed contrastive learning components in the context of UEFTC. Our findings highlight the effectiveness of two of these components. Our third topic focuses on uncertainty estimation on sequential labeling. Sequential labeling involves the task of assigning labels to individual tokens in a sequence, exemplified by Named Entity Recognition (NER). Despite significant advancements in enhancing NER performance in prior research, the realm of uncertainty estimation for NER (UE-NER) remains relatively uncharted but is of paramount importance. This topic focuses on UE-NER, seeking to gauge uncertainty scores for NER predictions. Previous models for uncertainty estimation often overlook two distinctive attributes of NER: the interrelation among entities (where the learning of one entity's embedding depends on others) and the challenges posed by incorrect span predictions in entity extraction. To address these issues, we introduce the Sequential Labeling Posterior Network (SLPN), designed to estimate uncertainty scores for the extracted entities while considering uncertainty propagation from other tokens. Additionally, we have devised an evaluation methodology tailored to the specific nuances of wrong-span cases. Our fourth topic focuses on an overlooked question that persists regarding the evaluation reliability of uncertainty estimation in text summarization (UE-TS). Text summarization, a key task in natural language generation (NLG), holds significant importance, particularly in domains where inaccuracies can have serious consequences, such as healthcare. UE-TS has garnered attention due to the potential risks associated with erroneous summaries. However, the reliability of evaluating UE-TS methods raises concerns, stemming from the interdependence between uncertainty model metrics and the wide array of NLG metrics. To address these concerns, we introduce a comprehensive UE-TS benchmark incorporating twenty-six NLG metrics across four dimensions. This benchmark evaluates the uncertainty estimation capabilities of two large language models and one pre-trained language model across two datasets. Additionally, it assesses the effectiveness of fourteen common uncertainty estimation methods. Our study underscores the necessity of utilizing diverse, uncorrelated NLG metrics and uncertainty estimation techniques for a robust evaluation of UE-TS methods. / Doctor of Philosophy / Text is integral to our daily activities, appearing in various forms such as social media posts, news articles, books, and reports. We rely on text for communication, information dissemination, and decision-making. Given its ubiquity, the ability to process and understand text through Natural Language Processing (NLP) has become increasingly important. NLP technology enables us to perform tasks like text classification, which involves categorizing text into predefined labels, and named entity recognition (NER), which identifies specific entities such as names, dates, and locations within text. Additionally, NLP facilitates generating coherent and contextually appropriate responses in conversational agents, enhancing human-computer interaction. However, the reliability of NLP models is crucial, especially in sensitive applications like medical diagnoses, where errors can have severe consequences. This dissertation focuses on uncertainty estimation in NLP, a less explored but essential area. Uncertainty estimation helps evaluate the confidence of NLP model predictions. We propose new methods tailored to various NLP tasks, acknowledging their unique needs. NLP tasks are divided into natural language understanding (NLU) and natural language generation (NLG). Within NLU, we look at tasks from two perspectives: a global view (e.g., document-level text classification) and a local view (e.g., sentence-level inference and token-level entity recognition). Our research spans text classification, named entity recognition (NER), and text summarization, with a special focus on few-shot text classification due to its recent prominence. For text classification, we introduce the MSD model, which includes three components to enhance uncertainty score accuracy and address overconfidence issues. This model integrates seamlessly with different neural networks and shows consistent improvements in experiments. For few-shot text classification, we develop Contrastive Learning from Uncertainty Relations (CLUR), designed to work effectively with minimal support samples per class. CLUR autonomously learns pseudo uncertainty scores, demonstrating effectiveness with various contrastive learning components. In NER, we address the unique challenges of entity interrelation and span prediction errors. We propose the Sequential Labeling Posterior Network (SLPN) to estimate uncertainty scores while considering uncertainty propagation from other tokens. For text summarization, we create a benchmark with tens of metrics to evaluate uncertainty estimation methods across two datasets. This benchmark helps assess the reliability of these methods, highlighting the need for diverse, uncorrelated metrics. Overall, our work advances the understanding and implementation of uncertainty estimation in NLP, providing more reliable and accurate predictions across different tasks.

Utvecklingen av artificiell intelligens och språkmodeller har ökat drastiskt under de senaste åren vilket medfört både möjligheter såväl som risker. Med en större användning av AI-relaterade produkter och människolika chattbotar har det medfört ett intresse av att kontrollera vilken sorts data som delas med dessa verktyg. Under särskilda omständigheter kan det förekomma data som till exempel information relaterat till personer, som inte får delas. Detta projekt har av denna anledning kretsat kring att använda och jämföra olika system för automatisk namnigenkänning, med målet att förhindra sådan data från att bli delad. I projektet jämfördes tre stycken olika alternativ för att implementera system för namnigenkänning, innan det mest lämpliga alternativet valdes för implementationen. Fortsättningsvis användes de tre förtränade transformer-modellerna GPT-SW3, TinyLlama och Mistral för implementationen där dessa tre blev finjusterade på två olika dataset. Implementationsfasen involverade applicering av tekniker för att öka datastorleken, databearbetning samt modellkvantisering innan de finjusterades för namnigenkänning. En uppsättning av utvärderingsmått bestående av bland annat F1-mått användes därefter för att mäta de tränade modellernas prestanda. De tre modellerna utvärderades och jämfördes med varandra utifrån resultatet från mätningen och träningen. Modellerna uppvisade varierande resultat och prestanda där både över- och underanpassning förekom. Avslutningsvis drogs slutsatsen om att TinyLlama var den bäst presterande modellen utifrån resultatet och övriga kringliggande aspekter. / The development of artificial intelligence and large language models has increased rapidly in recent years, bringing both opportunities and risks. With a broader use of AI related products such as human-like chatbots there has been an increase in interest in controlling the data that is being shared with them. In some scenarios there is data, such as personal or proprietary information, which should not be shared. This project has therefore revolved around utilizing and comparing different Named Entity Recognition systems to prevent such data from being shared. Three different approaches to implement Named Entity Recognition systems were compared before selecting the most appropriate one to further use for the actual implementation. Furthermore, three pre-trained transformer models, GPT-SW3, TinyLlama and Mistral, were used for the implementation where these were fine-tuned on two different datasets. The implementation phase included applying data augmentation techniques, data processing and model quantization before fine-tuning the models on Named Entity Recognition. A set of metrics including precision, recall and F1-score was further used to measure the performances of the trained models. The three models were compared and evaluated against each other based on the results obtained from the measurements and the training. The models showed varying results and performances where both overfitting and underfitting occured. Finally, the TinyLlama model was concluded to be the best model based on the obtained results and other considered aspects.

La thèse a deux objectifs : le premier est de développer un analyseur qui permet d'analyser automatiquement des sources textuelles en chinois simplifié afin de segmenter les textes en mots et de les étiqueter par catégories grammaticales, ainsi que de construire les relations syntaxiques entre les mots. Le deuxième est d'extraire des informations autour des entités et des actions qui nous intéressent à partir des textes analysés. Afin d'atteindre ces deux objectifs, nous avons traité principalement les problématiques suivantes : les ambiguïtés de segmentation, la catégorisation ; le traitement des mots inconnus dans les textes chinois ; l'ambiguïté de l'analyse syntaxique ; la reconnaissance et le typage des entités nommées. Le texte d'entrée est traité phrase par phrase. L'analyseur commence par un traitement typographique au sein des phrases afin d'identifier les écritures latines et les chiffres. Ensuite, nous segmentons la phrase en mots à l'aide de dictionnaires. Grâce aux règles linguistiques, nous créons des hypothèses de noms propres, changeons les poids des catégories ou des mots selon leur contextes gauches ou/et droits. Un modèle de langue n-gramme élaboré à partir d'un corpus d'apprentissage permet de sélectionner le meilleur résultat de segmentation et de catégorisation. Une analyse en dépendance est utilisée pour marquer les relations entre les mots. Nous effectuons une première identification d'entités nommées à la fin de l'analyse syntaxique. Ceci permet d'identifier les entités nommées en unité ou en groupe nominal et également de leur attribuer un type. Ces entités nommées sont ensuite utilisées dans l'extraction. Les règles d'extraction permettent de valider ou de changer les types des entités nommées. L'extraction des connaissances est composée des deux étapes : extraire et annoter automatiquement des contenus à partir des textes analysés ; vérifier les contenus extraits et résoudre la cohérence à travers une ontologie. / We have developed an automatic analyser and an extraction module for Chinese langage processing. The analyser performs automatic Chinese word segmentation based on linguistic rules and dictionaries, part-of-speech tagging based on n-gram statistics and dependency grammar parsing. The module allows to extract information around named entities and activities. In order to achieve these goals, we have tackled the following main issues: segmentation and part-of-speech ambiguity; unknown word identification in Chinese text; attachment ambiguity in parsing. Chinese texts are analysed sentence by sentence. Given a sentence, the analyzer begins with typographic processing to identify sequences of Latin characters and numbers. Then, dictionaries are used for preliminary segmentation into words. Linguistic-based rules are used to create proper noun hypotheses and change the weight of some word categories. These rules take into account word context. An n-gram language model is created from a training corpus and selects the best word segmentation and parts-of-speech. Dependency grammar parsing is used to annotate relations between words. A first step of named entity recognition is performed after parsing. Its goal is to identify single-word named entities and noun-phrase-based named entities and to determine their semantic type. These named entities are then used in knowledge extraction. Knowledge extraction rules are used to validate named entities or to change their types. Knowledge extraction consists of two steps: automatic content extraction and tagging from analysed text; extracted contents control and ontology-based co-reference resolution.

La thèse a deux objectifs : le premier est de développer un analyseur qui permet d'analyser automatiquement des sources textuelles en chinois simplifié afin de segmenter les textes en mots et de les étiqueter par catégories grammaticales, ainsi que de construire les relations syntaxiques entre les mots. Le deuxième est d'extraire des informations autour des entités et des actions qui nous intéressent à partir des textes analysés. Afin d'atteindre ces deux objectifs, nous avons traité principalement les problématiques suivantes : les ambiguïtés de segmentation, la catégorisation ; le traitement des mots inconnus dans les textes chinois ; l'ambiguïté de l'analyse syntaxique ; la reconnaissance et le typage des entités nommées. Le texte d'entrée est traité phrase par phrase. L'analyseur commence par un traitement typographique au sein des phrases afin d'identifier les écritures latines et les chiffres. Ensuite, nous segmentons la phrase en mots à l'aide de dictionnaires. Grâce aux règles linguistiques, nous créons des hypothèses de noms propres, changeons les poids des catégories ou des mots selon leur contextes gauches ou/et droits. Un modèle de langue n-gramme élaboré à partir d'un corpus d'apprentissage permet de sélectionner le meilleur résultat de segmentation et de catégorisation. Une analyse en dépendance est utilisée pour marquer les relations entre les mots. Nous effectuons une première identification d'entités nommées à la fin de l'analyse syntaxique. Ceci permet d'identifier les entités nommées en unité ou en groupe nominal et également de leur attribuer un type. Ces entités nommées sont ensuite utilisées dans l'extraction. Les règles d'extraction permettent de valider ou de changer les types des entités nommées. L'extraction des connaissances est composée des deux étapes : extraire et annoter automatiquement des contenus à partir des textes analysés ; vérifier les contenus extraits et résoudre la cohérence à travers une ontologie. / We have developed an automatic analyser and an extraction module for Chinese langage processing. The analyser performs automatic Chinese word segmentation based on linguistic rules and dictionaries, part-of-speech tagging based on n-gram statistics and dependency grammar parsing. The module allows to extract information around named entities and activities. In order to achieve these goals, we have tackled the following main issues: segmentation and part-of-speech ambiguity; unknown word identification in Chinese text; attachment ambiguity in parsing. Chinese texts are analysed sentence by sentence. Given a sentence, the analyzer begins with typographic processing to identify sequences of Latin characters and numbers. Then, dictionaries are used for preliminary segmentation into words. Linguistic-based rules are used to create proper noun hypotheses and change the weight of some word categories. These rules take into account word context. An n-gram language model is created from a training corpus and selects the best word segmentation and parts-of-speech. Dependency grammar parsing is used to annotate relations between words. A first step of named entity recognition is performed after parsing. Its goal is to identify single-word named entities and noun-phrase-based named entities and to determine their semantic type. These named entities are then used in knowledge extraction. Knowledge extraction rules are used to validate named entities or to change their types. Knowledge extraction consists of two steps: automatic content extraction and tagging from analysed text; extracted contents control and ontology-based co-reference resolution.

La traduction automatique des documents est considérée comme l’une des tâches les plus difficiles en traitement automatique des langues et de la parole. Les particularités linguistiques de certaines langues, comme la langue arabe, rendent la tâche de traduction automatique plus difficile. Notre objectif dans cette thèse est d'améliorer les systèmes de traduction de l'arabe vers le français et vers l'anglais. Nous proposons donc une étude détaillée sur ces systèmes. Les principales recherches portent à la fois sur la construction de corpus parallèles, le prétraitement de l'arabe et sur l'adaptation des modèles de traduction et de langue.Tout d'abord, un corpus comparable journalistique a été exploré pour en extraire automatiquement un corpus parallèle. Ensuite, différentes approches d’adaptation du modèle de traduction sont exploitées, soit en utilisant le corpus parallèle extrait automatiquement soit en utilisant un corpus parallèle construit automatiquement.Nous démontrons que l'adaptation des données du système de traduction permet d'améliorer la traduction. Un texte en arabe doit être prétraité avant de le traduire et ceci à cause du caractère agglutinatif de la langue arabe. Nous présentons notre outil de segmentation de l'arabe, SAPA (Segmentor and Part-of-speech tagger for Arabic), indépendant de toute ressource externe et permettant de réduire les temps de calcul. Cet outil permet de prédire simultanément l’étiquette morpho-syntaxique ainsi que les proclitiques (conjonctions, prépositions, etc.) pour chaque mot, ensuite de séparer les proclitiques du lemme (ou mot de base). Nous décrivons également dans cette thèse notre outil de détection des entités nommées, NERAr (Named Entity Recognition for Arabic), et nous examions l'impact de l'intégration de la détection des entités nommées dans la tâche de prétraitement et la pré-traduction de ces entités nommées en utilisant des dictionnaires bilingues. Nous présentons par la suite plusieurs méthodes pour l'adaptation thématique des modèles de traduction et de langue expérimentées sur une application réelle contenant un corpus constitué d’un ensemble de phrases multicatégoriques.Ces expériences ouvrent des perspectives importantes de recherche comme par exemple la combinaison de plusieurs systèmes lors de la traduction pour l'adaptation thématique. Il serait également intéressant d'effectuer une adaptation temporelle des modèles de traduction et de langue. Finalement, les systèmes de traduction améliorés arabe-français et arabe-anglais sont intégrés dans une plateforme d'analyse multimédia et montrent une amélioration des performances par rapport aux systèmes de traduction de base. / Machine Translation is one of the most difficult tasks in natural language and speech processing. The linguistic peculiarities of some languages makes the machine translation task more difficult. In this thesis, we present a detailed study of machine translation systems from arabic to french and to english.Our principle researches carry on building parallel corpora, arabic preprocessing and adapting translation and language models. We propose a method for automatic extraction of parallel news corpora from a comparable corpora. Two approaches for translation model adaptation are explored using whether parallel corpora extracted automatically or parallel corpora constructed automatically. We demonstrate that adapting data used to build machine translation system improves translation.Arabic texts have to be preprocessed before machine translation and this because of the agglutinative character of arabic language. A prepocessing tool for arabic, SAPA (Segmentor and Part-of-speech tagger for Arabic), much faster than the state of the art tools and totally independant of any other external resource was developed. This tool predicts simultaneously morphosyntactic tags and proclitics (conjunctions, prepositions, etc.) for every word, then splits off words into lemma and proclitics.We describe also in this thesis, our named entity recognition tool for arabic, NERAr, and we focus on the impact of integrating named entity recognition in the preprocessing task. We used bilingual dictionaries to propose translations of the detected named entities. We present then many approaches to adapt thematically translation and language models using a corpora consists of a set of multicategoric sentences.These experiments open important research perspectives such as combining many systems when translating. It would be interesting also to focus on a temporal adaptation of translation and language models.Finally, improved machine translation systems from arabic to french and english are integrated in a multimedia platform analysis and shows improvements compared to basic machine translation systems.

A partir do final do século 20, o governo federal criou vários programas visando a ampliação de acesso ao saneamento básico. Embora esses programas tenham trazido o abastecimento de água potável e a coleta de resíduos sólidos para a maioria dos municípios brasileiros, o esgotamento sanitário ainda está espacialmente concentrado na região Sudeste e nas áreas mais urbanizadas. Para explicar esse padrão espacialmente concentrado, é frequentemente assumido que o tamanho das cidades realmente importa para o saneamento básico, especialmente para o esgotamento sanitário. De fato, à medida que as cidades crescem em tamanho, devemos esperar economias de escala no volume de infraestrutura de saneamento. Economias de escala na infra-estrutura implicam uma redução nos custos de saneamento básico, de forma proporcional ao tamanho da cidade, levando também a uma (esperada) relação de lei de escala (ou de potência) entre os gastos com saneamento básico e o tamanho da cidade. Usando a população, N(t), como medida do tamanho da cidade no momento t, a lei de escala para infraestrutura assume o formato Y(t) = Y0N(t)&#946 onde &#946 &#8776 0.8 < 1, Y denota o volume de infraestrutura e Y0 é uma constante. Diversas propriedades das cidades, desde a produção de patentes e renda até a extensão da rede elétrica, são funções de lei de potência do tamanho da população com expoentes de escalamento, &#946, que se enquadram em classes distintas. As quantidades que refletem a criação de riqueza e a inovação têm &#946 &#8776 1.2 > 1 (retornos crescentes), enquanto aquelas responsáveis pela infraestrutura exibem &#946 &#8776 0.8 < 1 (economias de escala). Verificamos essa relação com base em dados extraídos do Sistema Integrado de Planejamento e Orçamento (SIOP), que abrangem transferências com recursos não onerosos, previstos na Lei Orçamentária Anual (LOA), na modalidade saneamento básico. No conjunto, os valores estimados de &#946 mostram redução das transferências da União Federal para saneamento básico, de forma proporcional ao tamanho dos municípios beneficiários. Para a dotação inicial, valores programados na LOA, estimado foi de aproximadamente: 0.63 para municípios com população superior a dois mil habitantes; 0.92 para municípios acima de vinte mil habitantes; e 1.18 para municípios com mais de cinquenta mil habitantes. A segunda fonte de dados identificada foi o Diário Oficial da União (DOU), periódico do governo federal para publicação de atos oficiais. Os dados fornecidos pelo DOU referem-se aos recursos não onerosos e também aos empréstimos com recursos do Fundo de Garantia por Tempo de Serviço (FGTS). Para extração dos dados textuais foram utilizadas técnicas de Processamento de Linguagem Natural(PLN). Essas técnicas funcionam melhor quando os algoritmos são alimentados com anotações - metadados que fornecem informações adicionais sobre o texto. Por isso geramos uma base de dados, a partir de textos anotados do DOU, para treinar uma rede LSTM bidirecional aplicada à etiquetagem morfossintática e ao reconhecimento de entidades nomeadas. Os resultados preliminares obtidos dessa forma estão relatados no texto / Starting in the late 20th century, the Brazilian federal government created several programs to increase the access to water and sanitation. However, although these programs made improvements in water access, sanitation was generally overlooked. While water supply, and waste collection are available in the majority of the Brazilian municipalities, the sewage system is still spatially concentrated in the Southeast region and in the most urbanized areas. In order to explain this spatially concentrated pattern it is frequently assumed that the size of cities does really matter for sanitation services provision, specially for sewage collection. As a matter of fact, as cities grow in size, one should expect economies of scale in sanitation infrastructure volume. Economies of scale in sanitation infrastructure means a decrease in basic sanitation costs, proportional to the city size, leading also to a (expected) power law relationship between the expenditure on sanitation and city size.Using population, N(t), as the measure of city size at time t, power law scaling for infrastructure takes the form Y(t) = Y0N(t)&#946 where &#946 &#8776 0.8 < 1, Y denotes infrastructure volume and is a constant. Many diverse properties of cities from patent production and personal income to electrical cable length are shown to be power law functions of population size with scaling exponents, &#946, that fall into distinct universality classes. Quantities reflecting wealth creation and innovation have &#946 &#8776 1.2 > 1 (increasing returns), whereas those accounting for infrastructure display &#946 &#8776 0.8 < 1 (economies of scale). We verified this relationship using data from federal government databases, called Integrated Planning and Budgeting System, known as SIOP. SIOP data refers only to grants, funds given to municipalities by the federal government to run programs within defined guidelines. Preliminary results from SIOP show decrease in Federal Grants to Brazilian Municipalities, proportional to the city size. For the initial budget allocation, &#946 was found to be roughly 0.63 for municipalities above twenty thousand inhabitants; to be roughly 0.92 for municipalities above twenty thousand inhabitants; and to be roughly 1.18 for municipalities above fifty thousand inhabitants. The second data source is DOU, government journal for publishing official acts. DOU data should give us information not only about grants, but also about FGTS funds for basic sanitation loans. In order to extract data from DOU we have applied Natural Language Processing (NLP) tools. These techniques often work better when the algorithms are provided with annotations metadata that provides additional information about the text. In particular, we fed a database with annotations into a bidirectional LSTM model applied to POS Tagging and Named-entity Recognition. Preliminary results are reported in the paper

Submitted by PPG Ci?ncia da Computa??o (ppgcc@pucrs.br) on 2018-05-03T18:01:24Z No. of bitstreams: 1 DANIELA_OLIVEIRA_FERREIRA_DO_AMARAL_TES.pdf: 6343384 bytes, checksum: a1d91fe5b12fa5cfdedb20ec1baf5042 (MD5) / Approved for entry into archive by Sheila Dias (sheila.dias@pucrs.br) on 2018-05-14T19:20:24Z (GMT) No. of bitstreams: 1 DANIELA_OLIVEIRA_FERREIRA_DO_AMARAL_TES.pdf: 6343384 bytes, checksum: a1d91fe5b12fa5cfdedb20ec1baf5042 (MD5) / Made available in DSpace on 2018-05-14T19:35:09Z (GMT). No. of bitstreams: 1 DANIELA_OLIVEIRA_FERREIRA_DO_AMARAL_TES.pdf: 6343384 bytes, checksum: a1d91fe5b12fa5cfdedb20ec1baf5042 (MD5) Previous issue date: 2017-09-14 / The treatment of textual information has been increasingly relevant in many do- mains. One of the first tasks for extracting information from texts is the Named Entities Recognition (NER), which consists of identifying references to certain entities and finding out their classification. There are many NER domains, among them the most usual are medicine and biology. One of the challenging domains in the recognition of Named Entities (NE) is the Geology domain, which is an area lacking computational linguistic resources. This thesis proposes a method for the recognition of relevant NE in the field of Geology, specifically to the subarea of Brazilian Sedimentary Basin, in Portuguese texts. Generic and geological features were defined for the generation of a machine learning model. Among the automatic approaches to NE classification, the most prominent is the Conditional Ran- dom Fields (CRF) probabilistic model. CRF has been effectively used for word processing in natural language. To generate our model, we created GeoCorpus, a reference corpus for Geological NER, annotated by specialists. Experimental evaluations were performed to compare the proposed method with other classifiers. The best results were achieved by CRF, which shows 76,78% of Precision and 54,33% of F-Measure. / O tratamento da informa??o textual torna-se cada vez mais relevante para muitos dom?nios. Nesse sentido, uma das primeira tarefas para Extra??o de Informa??es a partir de textos ? o Reconhecimento de Entidades Nomeadas (REN), que consiste na identifica??o de refer?ncias feitas a determinadas entidades e sua classifica??o. REN compreende muitos dom?nios, entre eles os mais usuais s?o medicina e biologia. Um dos dom?nios desafiadores no reconhecimento de EN ? o de Geologia, sendo essa uma ?rea carente de recursos lingu?sticos computacionais. A presente tese prop?e um m?todo para o reconhecimento de EN relevantes no dom?nio da Geologia, sub?rea Bacia Sedimentar Brasileira, em textos da l?ngua portuguesa. Definiram-se features gen?ricas e geol?gicas para a gera??o do modelo de aprendizado. Entre as abordagens autom?ticas para classifica??o de EN, a mais proeminente ? o modelo probabil?stico Conditional Random Fields (CRF). O CRF tem sido utilizado eficazmente no processamento de textos em linguagem natural. A fim de gerar um modelo de aprendizado foi criado o GeoCorpus, um corpus de refer?ncia para REN Geol?gicas, anotado por especialistas. Avalia??es experimentais foram realizadas com o objetivo de comparar o m?todo proposto com outros classificadores. Destacam-se os melhores resultados para o CRF, o qual alcan?ou 76,78% e 54,33% em Precis?o e Medida-F.

Submitted by Jadson Francisco de Jesus SILVA (jadson@uefs.br) on 2018-01-24T22:42:26Z No. of bitstreams: 1 JadsonDisst.pdf: 3499973 bytes, checksum: 5deaf9020f758e9c07f86e9e62890129 (MD5) / Made available in DSpace on 2018-01-24T22:42:26Z (GMT). No. of bitstreams: 1 JadsonDisst.pdf: 3499973 bytes, checksum: 5deaf9020f758e9c07f86e9e62890129 (MD5) Previous issue date: 2016-09-09 / The Named Entity Recognition (NER) process is the task of identifying relevant termsintextsandassigningthemalabel.Suchwordscanreferencenamesofpeople, organizations, and places. The variety of techniques that can be used in the named entityrecognitionprocessislarge.Thetechniquescanbeclassifiedintothreedistinct approaches: rule-based, machine learning and hybrid. Concerning to the machine learningapproaches,severalfactorsmayinfluenceitsaccuracy,includingtheselected classifier, the set of features extracted from the terms, the characteristics of the textual bases, and the number of entity labels. In this work, we compared classifiers that use machine learning applied to the NER task. The comparative study includes classifiers based on CRF (Conditional Random Fields), MEMM (MaximumEntropy Markov Model) and HMM (Hidden Markov Model), which are compared in two corpora in Portuguese derived from WikiNer, and HAREM, and two corporas in English derived from CoNLL-03 and WikiNer. The comparison of the classifiers shows that the CRF is superior to the other classifiers, both with Portuguese and English texts. This study also includes the comparison of the individual and joint contribution of features, including contextual features, besides the comparison ofthe NER per named entity labels, between classifiers andcorpora. / O processo de Reconhecimento de Entidades Nomeadas (REN) ? a tarefa de iden- tificar termos relevantes em textos e atribu?-los um r?tulo. Tais palavras podem referenciar nomes de pessoas, organiza??es e locais. A variedade de t?cnicas que podem ser usadas no processo de reconhecimento de entidades nomeadas ? grande. As t?cnicas podem ser classificadas em tr?s abordagens distintas: baseadas em regras, baseadas em aprendizagem de m?quina e h?bridas. No que diz respeito as abordagens de aprendizagem de m?quina, diversos fatores podem influenciar sua exatida?, incluindo o classificador selecionado, o conjunto de features extra?das dos termos, as caracter?sticas das bases textuais e o n?mero de r?tulos de entidades. Neste trabalho, comparamos classificadores que utilizam aprendizagem de m?quina aplicadas a tarefa do REN. O estudo comparativo inclui classificadores baseados no CRF (Condicional Random Fields), MEMM (Maximum Entropy Markov Model) e HMM (Hidden Markov Model), os quais s?o comparados em dois corporas em portugu?s derivados do WikiNer, e HAREM, e dois corporas em ingl?s derivados doCoNLL-03 e WikiNer. A compara??o dos classificadores demonstra que o CRF ? superior aos demais classificadores, tanto com textos em portugu?s, quanto ingl?s. Este estudo tamb?m inclui a compara??o da contribui??o, individual e em conjunto de features, incluindo features de contexto, al?m da compara??o do REN por r?otulos de entidades nomeadas, entre os classificadores e os corpora.

La thèse a deux objectifs : le premier est de développer un analyseur qui permet d'analyser automatiquement des sources textuelles en chinois simplifié afin de segmenter les textes en mots et de les étiqueter par catégories grammaticales, ainsi que de construire les relations syntaxiques entre les mots. Le deuxième est d'extraire des informations autour des entités et des actions qui nous intéressent à partir des textes analysés. Afin d'atteindre ces deux objectifs, nous avons traité principalement les problématiques suivantes : les ambiguïtés de segmentation, la catégorisation ; le traitement des mots inconnus dans les textes chinois ; l'ambiguïté de l'analyse syntaxique ; la reconnaissance et le typage des entités nommées. Le texte d'entrée est traité phrase par phrase. L'analyseur commence par un traitement typographique au sein des phrases afin d'identifier les écritures latines et les chiffres. Ensuite, nous segmentons la phrase en mots à l'aide de dictionnaires. Grâce aux règles linguistiques, nous créons des hypothèses de noms propres, changeons les poids des catégories ou des mots selon leur contextes gauches ou/et droits. Un modèle de langue n-gramme élaboré à partir d'un corpus d'apprentissage permet de sélectionner le meilleur résultat de segmentation et de catégorisation. Une analyse en dépendance est utilisée pour marquer les relations entre les mots. Nous effectuons une première identification d'entités nommées à la fin de l'analyse syntaxique. Ceci permet d'identifier les entités nommées en unité ou en groupe nominal et également de leur attribuer un type. Ces entités nommées sont ensuite utilisées dans l'extraction. Les règles d'extraction permettent de valider ou de changer les types des entités nommées. L'extraction des connaissances est composée des deux étapes : extraire et annoter automatiquement des contenus à partir des textes analysés ; vérifier les contenus extraits et résoudre la cohérence à travers une ontologie. / We have developed an automatic analyser and an extraction module for Chinese langage processing. The analyser performs automatic Chinese word segmentation based on linguistic rules and dictionaries, part-of-speech tagging based on n-gram statistics and dependency grammar parsing. The module allows to extract information around named entities and activities. In order to achieve these goals, we have tackled the following main issues: segmentation and part-of-speech ambiguity; unknown word identification in Chinese text; attachment ambiguity in parsing. Chinese texts are analysed sentence by sentence. Given a sentence, the analyzer begins with typographic processing to identify sequences of Latin characters and numbers. Then, dictionaries are used for preliminary segmentation into words. Linguistic-based rules are used to create proper noun hypotheses and change the weight of some word categories. These rules take into account word context. An n-gram language model is created from a training corpus and selects the best word segmentation and parts-of-speech. Dependency grammar parsing is used to annotate relations between words. A first step of named entity recognition is performed after parsing. Its goal is to identify single-word named entities and noun-phrase-based named entities and to determine their semantic type. These named entities are then used in knowledge extraction. Knowledge extraction rules are used to validate named entities or to change their types. Knowledge extraction consists of two steps: automatic content extraction and tagging from analysed text; extracted contents control and ontology-based co-reference resolution.

O Reconhecimento de Entidades Mencionadas (REM) é uma subtarefa da extração de informações e tem como objetivo localizar e classificar elementos do texto em categorias pré-definidas tais como nome de pessoas, organizações, lugares, datas e outras classes de interesse. Esse conhecimento obtido possibilita a execução de outras tarefas mais avançadas. O REM pode ser considerado um dos primeiros passos para a análise semântica de textos, além de ser uma subtarefa crucial para sistemas de gerenciamento de documentos, mineração de textos, extração da informação, entre outros. Neste trabalho, estudamos alguns métodos de Aprendizado de Máquina aplicados na tarefa de REM que estão relacionados ao atual estado da arte, dentre eles, dois métodos aplicados na tarefa de REM para a língua portuguesa. Apresentamos três diferentes formas de avaliação destes tipos de sistemas presentes na literatura da área. Além disso, desenvolvemos um sistema de REM para língua portuguesa utilizando Aprendizado de Máquina, mais especificamente, o arcabouço de máxima entropia. Os resultados obtidos com o nosso sistema alcançaram resultados equiparáveis aos melhores sistemas de REM para a língua portuguesa desenvolvidos utilizando outras abordagens de aprendizado de máquina. / Named Entity Recognition (NER), a task related to information extraction, aims to classify textual elements according to predefined categories such as names, places, dates etc. This enables the execution of more advanced tasks. NER is a first step towards semantic textual analysis and is also a crucial task for systems of information extraction and other types of systems. In this thesis, I analyze some Machine Learning methods applied to NER tasks, including two methods applied to Portuguese language. I present three ways of evaluating these types of systems found in the literature. I also develop an NER system for the Portuguese language utilizing Machine Learning that entails working with a maximum entropy framework. The results are comparable to the best NER systems for the Portuguese language developed with other Machine Learning alternatives.