Mémoire numérisé par la Division de la gestion de documents et des archives de l'Université de Montréal

The advent of the era of big data on the Web has made automatic web information extraction an essential tool in data acquisition processes. Unfortunately, automated solutions are in most cases more error prone than those created by humans, resulting in dirty and erroneous data. Automatic repair and cleaning of the extracted data is thus a necessary complement to information extraction on the Web. This thesis investigates the problem of inducing cleaning rules on web extracted data in order to (i) repair and align the data w.r.t. an original target schema, (ii) produce repairs that are as generic as possible such that different instances can benefit from them. The problem is addressed from three different angles: replace cross-site redundancy with an ensemble of entity recognisers; produce general repairs that can be encoded in the extraction process; and exploit entity-wide relations to infer common knowledge on extracted data. First, we present ROSeAnn, an unsupervised approach to integrate semantic annotators and produce a unied and consistent annotation layer on top of them. Both the diversity in vocabulary and widely varying accuracy justify the need for middleware that reconciles different annotator opinions. Considering annotators as "black-boxes" that do not require per-domain supervision allows us to recognise semantically related content in web extracted data in a scalable way. Second, we show in WADaR how annotators can be used to discover rules to repair web extracted data. We study the problem of computing joint repairs for web data extraction programs and their extracted data, providing an approximate solution that requires no per-source supervision and proves effective across a wide variety of domains and sources. The proposed solution is effective not only in repairing the extracted data, but also in encoding such repairs in the original extraction process. Third, we investigate how relationships among entities can be exploited to discover inconsistencies and additional information. We present RuDiK, a disk-based scalable solution to discover first-order logic rules over RDF knowledge bases built from web sources. We present an approach that does not limit its search space to rules that rely on "positive" relationships between entities, as in the case with traditional mining of constraints. On the contrary, it extends the search space to also discover negative rules, i.e., patterns that lead to contradictions in the data.

Cette thèse aborde la problématique de l'accès à l'information scientifique et technique véhiculée par de grands ensembles documentaires. Pour permettre à l'utilisateur de trouver l'information qui lui est pertinente, nous avons oeuvré à la définition d'un modèle répondant à l'exigence de souplesse de notre contexte applicatif industriel ; nous postulons pour cela la nécessité de segmenter l'information tirée des documents en plans ontologiques. Le modèle résultant permet une immersion documentaire, et ce grâce à trois types de processus complémentaires : des processus endogènes (exploitant le corpus pour analyser le corpus), exogènes (faisant appel à des ressources externes) et anthropogènes (dans lesquels les compétences de l'utilisateur sont considérées comme ressource) sont combinés. Tous concourent à l'attribution d'une place centrale à l'utilisateur dans le système, en tant qu'agent interprétant de l'information et concepteur de ses connaissances, dès lors qu'il est placé dans un contexte industriel ou spécialisé. / This thesis adresses the issue of accessing scientific and technical information conveyed by large sets of documents. To enable the user to find his own relevant information, we worked on a model meeting the requirement of flexibility imposed by our industrial application context ; to do so, we postulated the necessity of segmenting information from documents into ontological facets. The resulting model enables a documentary immersion, thanks to three types of complementary processes : endogenous processes (exploiting the corpus to analyze the corpus), exogenous processes (using external resources) and anthropogenous ones (in which the user's skills are considered as a resource) are combined. They all contribute to granting the user a fundamental role in the system, as an interpreting agent and as a knowledge creator, provided that he is placed in an industrial or specialised context.

No description available.

This thesis describes the development and in-depth empirical investigation of a method, called BootMark, for bootstrapping the marking up of named entities in textual documents. The reason for working with documents, as opposed to for instance sentences or phrases, is that the BootMark method is concerned with the creation of corpora. The claim made in the thesis is that BootMark requires a human annotator to manually annotate fewer documents in order to produce a named entity recognizer with a given performance, than would be needed if the documents forming the basis for the recognizer were randomly drawn from the same corpus. The intention is then to use the created named en- tity recognizer as a pre-tagger and thus eventually turn the manual annotation process into one in which the annotator reviews system-suggested annotations rather than creating new ones from scratch. The BootMark method consists of three phases: (1) Manual annotation of a set of documents; (2) Bootstrapping – active machine learning for the purpose of selecting which document to an- notate next; (3) The remaining unannotated documents of the original corpus are marked up using pre-tagging with revision. Five emerging issues are identified, described and empirically investigated in the thesis. Their common denominator is that they all depend on the real- ization of the named entity recognition task, and as such, require the context of a practical setting in order to be properly addressed. The emerging issues are related to: (1) the characteristics of the named entity recognition task and the base learners used in conjunction with it; (2) the constitution of the set of documents annotated by the human annotator in phase one in order to start the bootstrapping process; (3) the active selection of the documents to annotate in phase two; (4) the monitoring and termination of the active learning carried out in phase two, including a new intrinsic stopping criterion for committee-based active learning; and (5) the applicability of the named entity recognizer created during phase two as a pre-tagger in phase three. The outcomes of the empirical investigations concerning the emerging is- sues support the claim made in the thesis. The results also suggest that while the recognizer produced in phases one and two is as useful for pre-tagging as a recognizer created from randomly selected documents, the applicability of the recognizer as a pre-tagger is best investigated by conducting a user study involving real annotators working on a real named entity recognition task.

Le traitement automatique de la parole est un domaine qui englobe un grand nombre de travaux : de la reconnaissance automatique du locuteur à la détection des entités nommées en passant par la transcription en mots du signal audio. Les techniques de traitement automatique de la parole permettent d’extraire nombre d’informations des documents audio (réunions, émissions, etc.) comme la transcription, certaines annotations (le type d’émission, les lieux cités, etc.) ou encore des informations relatives aux locuteurs (changement de locuteur, genre du locuteur). Toutes ces informations peuvent être exploitées par des techniques d’indexation automatique qui vont permettre d’indexer de grandes collections de documents. Les travaux présentés dans cette thèse s’intéressent à l’indexation automatique de locuteurs dans des documents audio en français. Plus précisément nous cherchons à identifier les différentes interventions d’un locuteur ainsi qu’à les nommer par leur prénom et leur nom. Ce processus est connu sous le nom d’identification nommée du locuteur (INL). La particularité de ces travaux réside dans l’utilisation conjointe du signal audio et de sa transcription en mots pour nommer les locuteurs d’un document. Le prénom et le nom de chacun des locuteurs est extrait du document lui même (de sa transcription enrichie plus exactement), avant d’être affecté à un des locuteurs du document. Nous commençons par rappeler le contexte et les précédents travaux réalisés sur l’INL avant de présenter Milesin, le système développé lors de cette thèse. L’apport de ces travaux réside tout d’abord dans l’utilisation d’un détecteur automatique d’entités nommées (LIA_NE) pour extraire les couples prénom / nom de la transcription. Ensuite, ils s’appuient sur la théorie des fonctions de croyance pour réaliser l’affectation aux locuteurs du document et prennent ainsi en compte les différents conflits qui peuvent apparaître. Pour finir, un algorithme optimal d’affectation est proposé. Ce système obtient un taux d’erreur compris entre 12 et 20 % sur des transcriptions de référence (réalisées manuellement) en fonction du corpus utilisé. Nous présentons ensuite les avancées réalisées et les limites mises en avant par ces travaux. Nous proposons notamment une première étude de l’impact de l’utilisation de transcriptions entièrement automatiques sur Milesin. / The automatic processing of speech is an area that encompasses a large number of works : speaker recognition, named entities detection or transcription of the audio signal into words. Automatic speech processing techniques can extract number of information from audio documents (meetings, shows, etc..) such as transcription, some annotations (the type of show, the places listed, etc..) or even information concerning speakers (speaker change, gender of speaker). All this information can be exploited by automatic indexing techniques which will allow indexing of large document collections. The work presented in this thesis are interested in the automatic indexing of speakers in french audio documents. Specifically we try to identify the various contributions of a speaker and nominate them by their first and last name. This process is known as named identification of the speaker. The particularity of this work lies in the joint use of audio and its transcript to name the speakers of a document. The first and last name of each speaker is extracted from the document itself (from its rich transcription more accurately), before being assigned to one of the speakers of the document. We begin by describing the context and previous work on the speaker named identification process before submitting Milesin, the system developed during this thesis. The contribution of this work lies firstly in the use of an automatic detector of named entities (LIA_NE) to extract the first name / last name of the transcript. Afterwards, they rely on the theory of belief functions to perform the assignment to the speakers of the document and thus take into account the various conflicts that may arise. Finally, an optimal assignment algorithm is proposed. This system gives an error rate of between 12 and 20 % on reference transcripts (done manually) based on the corpus used.We then present the advances and limitations highlighted by this work.We propose an initial study of the impact of the use of fully automatic transcriptions on Milesin.

This dissertation addresses the problem of classification of entities in text represented by noun phrases. The goal of this thesis is to develop a method for automated classification of entities appearing in datasets consisting of short textual fragments. The emphasis is on unsupervised and semi-supervised methods that will allow for fine-grained character of the assigned classes and require no labeled instances for training. The set of target classes is either user-defined or determined automatically. Our initial attempt to address the entity classification problem is called Semantic Concept Mapping (SCM) algorithm. SCM maps the noun phrases representing the entities as well as the target classes to WordNet. Graph-based WordNet similarity measures are used to assign the closest class to the noun phrase. If a noun phrase does not match any WordNet concept, a Targeted Hypernym Discovery (THD) algorithm is executed. The THD algorithm extracts a hypernym from a Wikipedia article defining the noun phrase using lexico-syntactic patterns. This hypernym is then used to map the noun phrase to a WordNet synset, but it can also be perceived as the classification result by itself, resulting in an unsupervised classification system. SCM and THD algorithms were designed for English. While adaptation of these algorithms for other languages is conceivable, we decided to develop the Bag of Articles (BOA) algorithm, which is language agnostic as it is based on the statistical Rocchio classifier. Since this algorithm utilizes Wikipedia as a source of data for classification, it does not require any labeled training instances. WordNet is used in a novel way to compute term weights. It is also used as a positive term list and for lemmatization. A disambiguation algorithm utilizing global context is also proposed. We consider the BOA algorithm to be the main contribution of this dissertation. Experimental evaluation of the proposed algorithms is performed on the WordSim353 dataset, which is used for evaluation in the Word Similarity Computation (WSC) task, and on the Czech Traveler dataset, the latter being specifically designed for the purpose of our research. BOA performance on WordSim353 achieves Spearman correlation of 0.72 with human judgment, which is close to the 0.75 correlation for the ESA algorithm, to the author's knowledge the best performing algorithm for this gold-standard dataset, which does not require training data. The advantage of BOA over ESA is that it has smaller requirements on preprocessing of the Wikipedia data. While SCM underperforms on the WordSim353 dataset, it overtakes BOA on the Czech Traveler dataset, which was designed specifically for our entity classification problem. This discrepancy requires further investigation. In a standalone evaluation of THD on Czech Traveler dataset the algorithm returned a correct hypernym for 62% of entities.

The diploma thesis deals with the issue of distinguishing and denominating of geometric models by children aged five to six years. The aim of the work is to find out to what extent children are currently able to distinguish the world of the plane and the world of space and to what extent these two worlds merge for children. There are three hypotheses. H1: "Children are able to identify at least half of objects from a set of spatial bodies selected from a construction set.", hypothesis H2: "More than a half of children are confused naming planar and spatial objects." and hypothesis H3: "When identifying some of the objects familiar to children, they come up with their own names for them." To verify them, quantitative research was used, which also used audio recording to get data and to be able analyze the observed phenomena. The research confirmed all of those three these hypotheses. Two thousand two hundred data were analyzed. The work on a sample of hundred children shows what the level of object distinction and naming them is currently among preschool children in the Pilsen region. It opens up a number of other areas that need to be further addressed and lots of questions to be answered. These include mixing the world of space and the plane. KEYWORDS Space; plane; identification of 3D objects;...

This master thesis deals with the extraction of relationships between named entities in the text. In the theoretical part of the thesis, the issue of natural language representation for machine processing is discussed. Subsequently, two partial tasks of relationship extraction are defined, namely named entities recognition and classification of relationships between them, including a summary of state-of-the-art solutions. In the practical part of the thesis, system for automatic extraction of relationships between named entities from downloaded pages is designed. The classification of relationships between entities is based on the pre-trained transformers. In this thesis, four pre-trained transformers are compared, namely BERT, XLNet, RoBERTa and ALBERT.

In this master thesis are described the history and theoretical background of named-entity recognition and implementation of the system in C++ for named entity recognition and disambiguation. The system uses local disambiguation method and statistics generated from the  Wikilinks web dataset. With implemented system and with alternative implementations are performed various experiments and tests. These experiments show that the system is sufficiently successful and fast. System participates in the Entity Recognition and Disambiguation Challenge 2014.