Entity Extraction from Unstructured Data on the Web

Abstract

A large number of web pages contain information about entities in lists where the lists are represented in textual form. Textual lists contain implicit records of entities. However, the field values of such records cannot easily be separated or extracted by automatic processes. This, therefore, remains a challenging research problem in the literature. Previous studies in the literature relied mainly on probabilistic graph-based models to capture the attributes and the likely structures of implicit records in a list. However, one of the important limitations of existing methods is that the structures of the records in input lists were implicitly encoded via training data which was manually created. This thesis aims to investigate novel techniques to acquire automatically information about entities from implicit records embedded in textual lists on the web.This thesis introduces a self-supervised learning framework which exploits both existing data in a knowledge base and the structural similarity between sequences in lists to build an extraction model automatically. In the proposed framework, initial labels for candidate field values are created and assigned to generate label sequences. Then, the structure of implicit records is captured via a graphical model to assign unmatched labels and rectify mismatched labels. As a result of which, the process of entity extraction from lists can be completely unsupervised and automated without user intervention. In order to attain that outcome, we address three substantive research problems that need to be solved in this thesis.Firstly, the text segments in input lists need to be assigned labels precisely so that their statistical information is then used to build an extraction model. However, previous studies have not considered completely both the format and content of field values when performing the text segmentation and assigning labels. By viewing the problem of assigning labels for text segments as the problem of membership checking in set theory, we identify and propose a dyadic representation of semantic relations between a text segment and an attribute by using its extensional and intensional representations. We incorporate those representations to define a novel format-enhanced labelling technique to assign labels for text segments.Secondly, the labels of identical concepts with differing sequences in an input list are iii often located in similar positions but the positions of the labels may vary somewhat in different sequences. However, until this point, there has been no information extraction system designed to capture the distribution of labels in differing positions in order to enhance extraction results. To capture the positional information of labels, we are proposing a proximitybased positional model, which is combined with a sequential model to improve the quality of the label-refinement phase in our framework.Thirdly, in order to reduce dependence on the overlap between knowledge bases and input lists, we exploit structural similarity between t...

Enseigner la littérature numérique au secondaire, entre innovation et sédimentation : analyse de cas autour d’une recherche collaborative

“…La dernière étape de la séquence prévoit une production écrite portant sur l'appréciation de l'expérience de lecture vécue. On demande aux élèves de comparer les effets esthétiques provoqués par les ressources sémiotiques de trois oeuvres numériques (Florence(Mountains, 2018), Déprise(Bouchardon, 2019(Bouchardon, /2019 et The boat(Huynh & Le, 2015)), puis de justifier leur préférence pour l'une des trois.25 La séquence didactique de Samar a pour but de faire évoluer la conception des élèves à l'égard de la lecture qui, à son avis, est essentiellement monomodale et centrée sur l'écrit.Articulée en cinq étapes et déployée en huit séances, elle prévoit (1) une réflexion sur la définition et les périmètres de la lecture, (2) une lecture interprétative de l'oeuvre, (3) une analyse d'autres oeuvres et productions numériques, (4) une activité d'écriture créative, consistant à scénariser un nouveau texte numérique et multimodal et, enfin (5) une présentation orale des productions élaborées lors de l'étape précédente. Dans ce cas aussi, les élèves doivent remplir un questionnaire de lecture qui les invite, d'une part, à interroger les images, les couleurs, les musiques, la symbolique des illustrations, d'autre part, à écrire de nouveaux textes inspirés de quelques éléments esthétiques, techniques et sémiotiques de Florence.…”