<> "The repository administrator has not yet configured an RDF license."^^ . <> . . "Aspects of Coherence for Entity Analysis"^^ . "Natural language understanding is an important topic in natural language proces-\r\nsing. Given a text, a computer program should, at the very least, be able to under-\r\nstand what the text is about, and ideally also situate it in its extra-textual context\r\nand understand what purpose it serves. What exactly it means to understand what a\r\ntext is about is an open question, but it is generally accepted that, at a minimum, un-\r\nderstanding involves being able to answer questions like “Who did what to whom?\r\nWhere? When? How? And Why?”. Entity analysis, the computational analysis of\r\nentities mentioned in a text, aims to support answering the questions “Who?” and\r\n“Whom?” by identifying entities mentioned in a text. If the answers to “Where?”\r\nand “When?” are specific, named locations and events, entity analysis can also pro-\r\nvide these answers. Entity analysis aims to answer these questions by performing\r\nentity linking, that is, linking mentions of entities to their corresponding entry in\r\na knowledge base, coreference resolution, that is, identifying all mentions in a text\r\nthat refer to the same entity, and entity typing, that is, assigning a label such as\r\nPerson to mentions of entities.\r\nIn this thesis, we study how different aspects of coherence can be exploited to\r\nimprove entity analysis. Our main contribution is a method that allows exploiting\r\nknowledge-rich, specific aspects of coherence, namely geographic, temporal, and\r\nentity type coherence. Geographic coherence expresses the intuition that entities\r\nmentioned in a text tend to be geographically close. Similarly, temporal coherence\r\ncaptures the intuition that entities mentioned in a text tend to be close in the tem-\r\nporal dimension. Entity type coherence is based in the observation that in a text\r\nabout a certain topic, such as sports, the entities mentioned in it tend to have the\r\nsame or related entity types, such as sports team or athlete. We show how to integrate\r\nfeatures modeling these aspects of coherence into entity linking systems and esta-\r\nblish their utility in extensive experiments covering different datasets and systems.\r\nSince entity linking often requires computationally expensive joint, global optimi-\r\nzation, we propose a simple, but effective rule-based approach that enjoys some of\r\nthe benefits of joint, global approaches, while avoiding some of their drawbacks.\r\nTo enable convenient error analysis for system developers, we introduce a tool for \r\nvisual analysis of entity linking system output. Investigating another aspect of co-\r\nherence, namely the coherence between a predicate and its arguments, we devise a\r\ndistributed model of selectional preferences and assess its impact on a neural core-\r\nference resolution system. Our final contribution examines how multilingual entity\r\ntyping can be improved by incorporating subword information. We train and make\r\npublicly available subword embeddings in 275 languages and show their utility in\r\na multilingual entity typing task"^^ . "2019" . . . . . . . "Benjamin"^^ . "Heinzerling"^^ . "Benjamin Heinzerling"^^ . . . . . . "Aspects of Coherence for Entity Analysis (PDF)"^^ . . . "th.pdf"^^ . . . "Aspects of Coherence for Entity Analysis (Other)"^^ . . . . . . "indexcodes.txt"^^ . . . "Aspects of Coherence for Entity Analysis (Other)"^^ . . . . . . "small.jpg"^^ . . . "Aspects of Coherence for Entity Analysis (Other)"^^ . . . . . . "medium.jpg"^^ . . . "Aspects of Coherence for Entity Analysis (Other)"^^ . . . . . . "preview.jpg"^^ . . . "Aspects of Coherence for Entity Analysis (Other)"^^ . . . . . . "lightbox.jpg"^^ . . "HTML Summary of #26117 \n\nAspects of Coherence for Entity Analysis\n\n" . "text/html" . . . "004 Informatik"@de . "004 Data processing Computer science"@en . .