Este trabalho faz parte do projeto de uma ferramenta denominada FIP (Ferramenta Inteligente de Apoio à Pesquisa) para recuperação, organização e mineração de grandes coleções de documentos. No contexto da ferramenta FIP, diversas técnicas de Recuperação de Informação, Mineração de Dados, Visualização de Informações e, em particular, técnicas de Extração de Informações, foco deste trabalho, são usadas. Sistemas de Extração de Informação atuam sobre um conjunto de dados não estruturados e objetivam localizar informações específicas em um documento ou coleção de documentos, extraí-las e estruturá-las com o intuito de facilitar o uso dessas informações. O objetivo específico desenvolvido nesta dissertação é induzir, de forma automática, um conjunto de regras para a extração de informações de artigos científicos. O sistema de extração proposto, inicialmente, analisa e extrai informações presentes no corpo dos artigos (título, autores, a filiação, resumo, palavras chaves) e, posteriormente, foca na extração das informações de suas referências bibliográficas. A proposta para extração automática das informações das referências é uma abordagem nova, baseada no mapeamento do problema de part-of-speech tagging ao problema de extração de informação. Como produto final do processo de extração, tem-se uma base de dados com as informações extraídas e estruturadas no formato XML, disponível à ferramenta FIP ou a qualquer outra aplicação. Os resultados obtidos foram avaliados em termos das métricas precisão, cobertura e F-measure, alcançando bons resultados comparados com sistemas similares / This dissertation is part of a project of a tool named FIP (an Intelligent Tool for Research Supporting). FIP is a tool for retrieval, organization, and mining large document collections. In the context of FIP diverse techniques from Information Retrieval, Data Mining, Information Visualization, and particularly Information Extraction, focus of this work, are used. Information Extraction systems deal with unstructured data looking for specific information in a document or document collection, extracting and structuring them in order to facilitate their use. The specific objective presented in this dissertation is automatically to induce a set of rules for information extraction from scientific articles. The proposed extraction system initially analyzes and extracts information from the body of the articles (heading, authors, affiliation, abstract, and keywords) and then extracts information from each reference in its bibliographical references. The proposed approach for information extraction from references is a new technique based on the strategy of part-of-speech tagging. As the outcome of the extraction process, a database with extracted and structured information in XML format is made available for the FIP or any other application. The system has been evaluated using measures of Precision, Recall and F-measure, reaching good results compared to similar systems

L’internet et les nouvelles formes de média de communication, d’information, et de divertissement ont entraîné une croissance massive de la quantité des données numériques. Le traitement et l’interprétation automatique de ces données permettent de créer des bases de connaissances, de rendre les recherches plus efficaces et d’effectuer des recherches sur les médias sociaux. Les travaux de recherche sur le traitement automatique du langage naturel concernent la conception et le développement d’algorithmes, qui permettent aux ordinateurs de traiter automatiquement le langage naturel dans les textes, les contenus audio, les images ou les vidéos, pour des tâches spécifiques. De par la complexité du langage humain, le traitement du langage naturel sous forme textuelle peut être divisé en 4 niveaux : la morphologie, la syntaxe, la sémantique et la pragmatique. Les technologies actuelles du traitement du langage naturel ont eu de grands succès sur les tâches liées auxdeux premiers niveaux, ce qui a permis la commercialisation de beaucoup d’applications comme les moteurs de recherche. Cependant, les moteurs de recherches avancés (structurels) nécessitent une interprétation du langage plus avancée. L’extraction d’information consiste à extraire des informations structurelles à partir des ressources non annotées ou semi-annotées, afin de permettre des recherches avancées et la création automatique des bases de connaissances. Cette thèse étudie le problème d’extraction d’information dans le domaine spécifique de l’extraction des événements biomédicaux. Nous proposons une solution efficace, qui fait un compromis entre deux types principaux de méthodes proposées dans la littérature. Cette solution arrive à un bon équilibre entre la performance et la rapidité, ce qui la rend utilisable pour traiter des données à grande échelle. Elle a des performances compétitives face aux meilleurs modèles existant avec une complexité en temps de calcul beaucoup plus faible. Lors la conception de ce modèle, nous étudions également les effets des différents classifieurs qui sont souvent proposés pour la résolution des problèmes de classification multi-classe. Nous testons également deux méthodes permettant d’intégrer des représentations vectorielles des mots appris par apprentissage profond (deep learning). Même si les classifieurs différents et l’intégration des vecteurs de mots n’améliorent pas grandement la performance, nous pensons que ces directions de recherche ont du potentiel et sont prometteuses pour améliorer l’extraction d’information. / Internet as well as all the modern media of communication, information and entertainment entails a massive increase of digital data quantities. Automatically processing and understanding these massive data enables creating large knowledge bases, more efficient search, social medial research, etc. Natural language processing research concerns the design and development of algorithms that allow computers to process natural language in texts, audios, images or videos automatically for specific tasks. Due to the complexity of human language, natural language processing of text can be divided into four levels: morphology, syntax, semantics and pragmatics. Current natural language processing technologies have achieved great successes in the tasks of the first two levels, leading to successes in many commercial applications such as search. However, advanced structured search engine would require computers to understand language deeper than at the morphology and syntactic levels. Information extraction is designed to extract meaningful structural information from unannotated or semi-annotated resources to enable advanced search and automatically create knowledge bases for further use. This thesis studies the problem of information extraction in the specific domain of biomedical event extraction. We propose an efficient solution, which is a trade-off between the two main trends of methods proposed in previous work. This solution reaches a good balance point between performance and speed, which is suitable to process large scale data. It achieves competitive performance to the best models with a much lower computational complexity. While designing this model, we also studied the effects of different classifiers that are usually proposed to solve the multi-class classification problem. We also tested two simple methods to integrate word vector representations learned by deep learning method into our model. Even if different classifiers and the integration of word vectors do not greatly improve the performance, we believe that these research directions carry some promising potential for improving information extraction.

Background: Traditionally text mention normalization corpora have normalized concepts to single ontology identifiers ("pre-coordinated concepts"). Less frequently, normalization corpora have used concepts with multiple identifiers ("post-coordinated concepts") but the additional identifiers have been restricted to a defined set of relationships to the core concept. This approach limits the ability of the normalization process to express semantic meaning. We generated a freely available corpus using post-coordinated concepts without a defined set of relationships that we term "compositional concepts" to evaluate their use in clinical text. Methods: We annotated 5397 disorder mentions from the ShARe corpus to SNOMED CT that were previously normalized as "CUI-less" in the "SemEval-2015 Task 14" shared task because they lacked a pre-coordinated mapping. Unlike the previous normalization method, we do not restrict concept mappings to a particular set of the Unified Medical Language System (UMLS) semantic types and allow normalization to occur to multiple UMLS Concept Unique Identifiers (CUIs). We computed annotator agreement and assessed semantic coverage with this method. Results: We generated the largest clinical text normalization corpus to date with mappings to multiple identifiers and made it freely available. All but 8 of the 5397 disorder mentions were normalized using this methodology. Annotator agreement ranged from 52.4% using the strictest metric (exact matching) to 78.2% using a hierarchical agreement that measures the overlap of shared ancestral nodes. Conclusion: Our results provide evidence that compositional concepts can increase semantic coverage in clinical text. To our knowledge we provide the first freely available corpus of compositional concept annotation in clinical text.

With the constantly growing amount of biomedical literature, methods for automatically distilling information from unstructured data, collectively known as information extraction, have become indispensable. Whilst most biomedical information extraction efforts in the last decade have focussed on the identification of gene products and interactions between them, the biomedical text mining community has recently extended their scope to capture associations between biomedical and chemical entities with the aim of supporting applications in drug discovery. This thesis is the first comprehensive study focussing on information extraction from pharmaceutical chemistry literature. In this research, we describe our work on (1) recognising names of chemical compounds and drugs, facilitated by the incorporation of domain knowledge; (2) exploring different coreference resolution paradigms in order to recognise co-referring expressions given a full-text article; and (3) defining drug-target interactions as events and distilling them from pharmaceutical chemistry literature using event extraction methods.

This dissertation addresses the problem of analysing digital data in digital forensics. It will be shown that text mining methods can be adapted and applied to digital forensics to aid analysts to more quickly, efficiently and accurately analyse data to reveal truly useful information. Investigators who wish to utilise digital evidence must examine and organise the data to piece together events and facts of a crime. The difficulty with finding relevant information quickly using the current tools and methods is that these tools rely very heavily on background knowledge for query terms and do not fully utilise the content of the data. A novel framework in which to perform evidence discovery is proposed in order to reduce the quantity of data to be analysed, aid the analysts' exploration of the data and enhance the intelligibility of the presentation of the data. The framework combines information extraction techniques with visual exploration techniques to provide a novel approach to performing evidence discovery, in the form of an evidence discovery system. By utilising unrestricted, unsupervised information extraction techniques, the investigator does not require input queries or keywords for searching, thus enabling the investigator to analyse portions of the data that may not have been identified by keyword searches. The evidence discovery system produces text graphs of the most important concepts and associations extracted from the full text to establish ties between the concepts and provide an overview and general representation of the text. Through an interactive visual interface the investigator can explore the data to identify suspects, events and the relations between suspects. Two models are proposed for performing the relation extraction process of the evidence discovery framework. The first model takes a statistical approach to discovering relations based on co-occurrences of complex concepts. The second model utilises a linguistic approach using named entity extraction and information extraction patterns. A preliminary study was performed to assess the usefulness of a text mining approach to digital forensics as against the traditional information retrieval approach. It was concluded that the novel approach to text analysis for evidence discovery presented in this dissertation is a viable and promising approach. The preliminary experiment showed that the results obtained from the evidence discovery system, using either of the relation extraction models, are sensible and useful. The approach advocated in this dissertation can therefore be successfully applied to the analysis of textual data for digital forensics Copyright / Dissertation (MSc)--University of Pretoria, 2010. / Computer Science / unrestricted

The automated extraction of semantic chemical data from the existing literature is demonstrated. For reasons of copyright, the work is focused on the patent literature, though the methods are expected to apply equally to other areas of the chemical literature. Hearst Patterns are applied to the patent literature in order to discover hyponymic relations describing chemical species. The acquired relations are manually validated to determine the precision of the determined hypernyms (85.0%) and of the asserted hyponymic relations (94.3%). It is demonstrated that the system acquires relations that are not present in the ChEBI ontology, suggesting that it could function as a valuable aid to the ChEBI curators. The relations discovered by this process are formalised using the Web Ontology Language (OWL) to enable re-use. PatentEye - an automated system for the extraction of reactions from chemical patents and their conversion to Chemical Markup Language (CML) - is presented. Chemical patents published by the European Patent Office over a ten-week period are used to demonstrate the capability of PatentEye - 4444 reactions are extracted with a precision of 78% and recall of 64% with regards to determining the identity and amount of reactants employed and an accuracy of 92% with regards to product identification. NMR spectra are extracted from the text using OSCAR3, which is developed to greatly increase recall. The resulting system is presented as a significant advancement towards the large-scale and automated extraction of high-quality reaction information. Extended Polymer Markup Language (EPML), a CML dialect for the description of Markush structures as they are presented in the literature, is developed. Software to exemplify and to enable substructure searching of EPML documents is presented. Further work is recommended to refine the language and code to publication-quality before they are presented to the community.

The purpose of this work was to deal with data extraction from various formats (HTML, XML, XLS) and transformation for further processing. As the data sources were used Czech public contracts and related code lists and classifications. Main goal was to implement periodic data extraction, RDF transformation and publishing the output in form of Linked Data using SPARQL endpoint. It was necessary to design and implement extraction modules for UnifiedViews tool as it was used for periodic extractions. Theoretical section of this thesis explains the principles of linked data and key tools used for data extraction and manipulation. Practical section deals with extractors design and implementation. Part describing extractor implementation shows methods for parsing data in various dataset formats and its transformation to RDF. The success of each extractor implementation is presented at the conclusion along with thought of usability in a real world.

La plus grande partie des informations disponibles librement sur le Web se présentent sous une forme textuelle, c'est-à-dire non-structurée. Dans un contexte comme celui de la veille, il est très utile de pouvoir présenter les informations présentes dans les textes sous une forme structurée en se focalisant sur celles jugées pertinentes vis-à-vis du domaine d'intérêt considéré. Néanmoins, lorsque l'on souhaite traiter ces informations de façon systématique, les méthodes manuelles ne sont pas envisageables du fait du volume important des données à considérer.L'extraction d'information s'inscrit dans la perspective de l'automatisation de ce type de tâches en identifiant dans des textes les informations concernant des faits (ou événements) afin de les stocker dans des structures de données préalablement définies. Ces structures, appelées templates (ou formulaires), agrègent les informations caractéristiques d'un événement ou d'un domaine d'intérêt représentées sous la forme d'entités nommées (nom de lieux, etc.).Dans ce contexte, le travail de thèse que nous avons mené s'attache à deux grandes problématiques : l'identification des informations liées à un événement lorsque ces informations sont dispersées à une échelle textuelle en présence de plusieurs occurrences d'événements de même type;la réduction de la dépendance vis-à-vis de corpus annotés pour la mise en œuvre d'un système d'extraction d'information.Concernant la première problématique, nous avons proposé une démarche originale reposant sur deux étapes. La première consiste en une segmentation événementielle identifiant dans un document les zones de texte faisant référence à un même type d'événements, en s'appuyant sur des informations de nature temporelle. Cette segmentation détermine ainsi les zones sur lesquelles le processus d'extraction doit se focaliser. La seconde étape sélectionne à l'intérieur des segments identifiés comme pertinents les entités associées aux événements. Elle conjugue pour ce faire une extraction de relations entre entités à un niveau local et un processus de fusion global aboutissant à un graphe d'entités. Un processus de désambiguïsation est finalement appliqué à ce graphe pour identifier l'entité occupant un rôle donné vis-à-vis d'un événement lorsque plusieurs sont possibles.La seconde problématique est abordée dans un contexte de peuplement de bases de connaissances à partir de larges ensembles de documents (plusieurs millions de documents) en considérant un grand nombre (une quarantaine) de types de relations binaires entre entités nommées. Compte tenu de l'effort représenté par l'annotation d'un corpus pour un type de relations donné et du nombre de types de relations considérés, l'objectif est ici de s'affranchir le plus possible du recours à une telle annotation tout en conservant une approche par apprentissage. Cet objectif est réalisé par le biais d'une approche dite de supervision distante prenant comme point de départ des exemples de relations issus d'une base de connaissances et opérant une annotation non supervisée de corpus en fonction de ces relations afin de constituer un ensemble de relations annotées destinées à la construction d'un modèle par apprentissage. Cette approche a été évaluée à large échelle sur les données de la campagne TAC-KBP 2010. / The major part of the information available on the web is provided in textual form, i.e. in unstructured form. In a context such as technology watch, it is useful to present the information extracted from a text in a structured form, reporting only the pieces of information that are relevant to the considered field of interest. Such processing cannot be performed manually at large scale, given the large amount of data available. The automated processing of this task falls within the Information extraction (IE) domain.The purpose of IE is to identify, within documents, pieces of information related to facts (or events) in order to store this information in predefined data structures. These structures, called templates, aggregate fact properties - often represented by named entities - concerning an event or an area of interest.In this context, the research performed in this thesis addresses two problems:identifying information related to a specific event, when the information is scattered across a text and several events of the same type are mentioned in the text;reducing the dependency to annotated corpus for the implementation of an Information Extraction system.Concerning the first problem, we propose an original approach that relies on two steps. The first step operates an event-based text segmentation, which identifies within a document the text segments on which the IE process shall focus to look for the entities associated with a given event. The second step focuses on template filling and aims at selecting, within the segments identified as relevant by the event-based segmentation, the entities that should be used as fillers, using a graph-based method. This method is based on a local extraction of relations between entities, that are merged in a relation graph. A disambiguation step is then performed on the graph to identify the best candidates to fill the information template.The second problem is treated in the context of knowledge base (KB) population, using a large collection of texts (several millions) from which the information is extracted. This extraction also concerns a large number of relation types (more than 40), which makes the manual annotation of the collection too expensive. We propose, in this context, a distant supervision approach in order to use learning techniques for this extraction, without the need of a fully annotated corpus. This distant supervision approach uses a set of relations from an existing KB to perform an unsupervised annotation of a collection, from which we learn a model for relation extraction. This approach has been evaluated at a large scale on the data from the TAC-KBP 2010 evaluation campaign.

This paper introduced a Computer-Aided Optically Scanned Document Information Extraction System. It could extract information including invoice No., issued date, buyer, etc., from the optically scanned document to meet the demand of customs declaration companies. The system output the structured information to a relational database. In detail, a software architecture for the information extraction of diverse-structure optically scanned document is designed. In this system, the original document is classified firstly. It would put into template-based extraction to improve the extraction performance if its template is pre-defined in the system. Then, a method for image enhancement to improve the image classification is proposed. This method aims to optimize the accuracy of neural network model by extracting the template-related feature and actively removing the unrelated feature. Lastly, the above system is implemented in this paper. This extraction are programed in Python which is a cross-platform languages. This system comprises three parts, classification module, template-based extraction and non-template extraction all of which have APIs and could be ran independently. This feature make this system flexible and easy to customization for the further demand. 445 real-world customs document images were input to evaluate the system. The result revealed that the introduced system ensured the diverse document support with non-template extraction and reached the overall high performance with template-based extraction showing the goal was basically achieved.

京都大学 / 0048 / 新制・課程博士 / 博士(情報学) / 甲第22574号 / 情博第711号 / 新制||情||122(附属図書館) / 京都大学大学院情報学研究科社会情報学専攻 / (主査)教授 吉川 正俊, 教授 鹿島 久嗣, 教授 田島 敬史, 特定准教授 JATOWT Adam Wladyslaw / 学位規則第4条第1項該当 / Doctor of Informatics / Kyoto University / DGAM