• Refine Query
  • Source
  • Publication year
  • to
  • Language
  • 851
  • 186
  • 86
  • 58
  • 34
  • 24
  • 16
  • 12
  • 11
  • 10
  • 8
  • 6
  • 5
  • 4
  • 4
  • Tagged with
  • 1593
  • 1593
  • 1378
  • 555
  • 520
  • 432
  • 354
  • 344
  • 240
  • 227
  • 218
  • 217
  • 210
  • 204
  • 195
  • About
  • The Global ETD Search service is a free service for researchers to find electronic theses and dissertations. This service is provided by the Networked Digital Library of Theses and Dissertations.
    Our metadata is collected from universities around the world. If you manage a university/consortium/country archive and want to be added, details can be found on the NDLTD website.
191

Linguistic Refactoring of Business Process Models

Pittke, Fabian 11 1900 (has links) (PDF)
In the past decades, organizations had to face numerous challenges due to intensifying globalization and internationalization, shorter innovation cycles and growing IT support for business. Business process management is seen as a comprehensive approach to align business strategy, organization, controlling, and business activities to react flexibly to market changes. For this purpose, business process models are increasingly utilized to document and redesign relevant parts of the organization's business operations. Since companies tend to have a growing number of business process models stored in a process model repository, analysis techniques are required that assess the quality of these process models in an automatic fashion. While available techniques can easily check the formal content of a process model, there are only a few techniques available that analyze the natural language content of a process model. Therefore, techniques are required that address linguistic issues caused by the actual use of natural language. In order to close this gap, this doctoral thesis explicitly targets inconsistencies caused by natural language and investigates the potential of automatically detecting and resolving them under a linguistic perspective. In particular, this doctoral thesis provides the following contributions. First, it defines a classification framework that structures existing work on process model analysis and refactoring. Second, it introduces the notion of atomicity, which implements a strict consistency condition between the formal content and the textual content of a process model. Based on an explorative investigation, we reveal several reoccurring violation patterns are not compliant with the notion of atomicity. Third, this thesis proposes an automatic refactoring technique that formalizes the identified patterns to transform a non-atomic process models into an atomic one. Fourth, this thesis defines an automatic technique for detecting and refactoring synonyms and homonyms in process models, which is eventually useful to unify the terminology used in an organization. Fifth and finally, this thesis proposes a recommendation-based refactoring approach that addresses process models suffering from incompleteness and leading to several possible interpretations. The efficiency and usefulness of the proposed techniques is further evaluated by real-world process model repositories from various industries. (author's abstract)
192

Predicting the programming language of questions and snippets of stack overflow using natural language processing

Alrashedy, Kamel 11 September 2018 (has links)
Stack Overflow is the most popular Q&A website among software developers. As a platform for knowledge sharing and acquisition, the questions posted in Stack Over- flow usually contain a code snippet. Stack Overflow relies on users to properly tag the programming language of a question and assumes that the programming language of the snippets inside a question is the same as the tag of the question itself. In this the- sis, a classifier is proposed to predict the programming language of questions posted in Stack Overflow using Natural Language Processing (NLP) and Machine Learning (ML). The classifier achieves an accuracy of 91.1% in predicting the 24 most popular programming languages by combining features from the title, body and code snippets of the question. We also propose a classifier that only uses the title and body of the question and has an accuracy of 81.1%. Finally, we propose a classifier of code snip- pets only that achieves an accuracy of 77.7%.Thus, deploying ML techniques on the combination of text and code snippets of a question provides the best performance. These results demonstrate that it is possible to identify the programming language of a snippet of only a few lines of source code. We visualize the feature space of two programming languages Java and SQL in order to identify some properties of the information inside the questions corresponding to these languages. / Graduate
193

Supporting Process Model Validation through Natural Language Generation

Leopold, Henrik, Mendling, Jan, Polyvyanyy, Artem 29 May 2014 (has links) (PDF)
The design and development of process-aware information systems is often supported by specifying requirements as business process models. Although this approach is generally accepted as an effective strategy, it remains a fundamental challenge to adequately validate these models given the diverging skill set of domain experts and system analysts. As domain experts often do not feel confident in judging the correctness and completeness of process models that system analysts create, the validation often has to regress to a discourse using natural language. In order to support such a discourse appropriately, so-called verbalization techniques have been defined for different types of conceptual models. However, there is currently no sophisticated technique available that is capable of generating natural-looking text from process models. In this paper, we address this research gap and propose a technique for generating natural language texts from business process models. A comparison with manually created process descriptions demonstrates that the generated texts are superior in terms of completeness, structure, and linguistic complexity. An evaluation with users further demonstrates that the texts are very understandable and effectively allow the reader to infer the process model semantics. Hence, the generated texts represent a useful input for process model validation.
194

Vyvozování v přirozeném jazyce s využitím obrazových dat / Grounding Natural Language Inference on Images

Vu Trong, Hoa January 2018 (has links)
Grounding Natural Language Inference on Images Hoa Trong VU July 20, 2018 Abstract Despite the surge of research interest in problems involving linguistic and vi- sual information, exploring multimodal data for Natural Language Inference remains unexplored. Natural Language Inference, regarded as the basic step towards Natural Language Understanding, is extremely challenging due to the natural complexity of human languages. However, we believe this issue can be alleviated by using multimodal data. Given an image and its description, our proposed task is to determined whether a natural language hypothesis contra- dicts, entails or is neutral with regards to the image and its description. To address this problem, we develop a multimodal framework based on the Bilat- eral Multi-perspective Matching framework. Data is collected by mapping the SNLI dataset with the image dataset Flickr30k. The result dataset, made pub- licly available, has more than 565k instances. Experiments on this dataset show that the multimodal model outperforms the state-of-the-art textual model. References 1
195

Advancing Biomedical Named Entity Recognition with Multivariate Feature Selection and Semantically Motivated Features

January 2013 (has links)
abstract: Automating aspects of biocuration through biomedical information extraction could significantly impact biomedical research by enabling greater biocuration throughput and improving the feasibility of a wider scope. An important step in biomedical information extraction systems is named entity recognition (NER), where mentions of entities such as proteins and diseases are located within natural-language text and their semantic type is determined. This step is critical for later tasks in an information extraction pipeline, including normalization and relationship extraction. BANNER is a benchmark biomedical NER system using linear-chain conditional random fields and the rich feature set approach. A case study with BANNER locating genes and proteins in biomedical literature is described. The first corpus for disease NER adequate for use as training data is introduced, and employed in a case study of disease NER. The first corpus locating adverse drug reactions (ADRs) in user posts to a health-related social website is also described, and a system to locate and identify ADRs in social media text is created and evaluated. The rich feature set approach to creating NER feature sets is argued to be subject to diminishing returns, implying that additional improvements may require more sophisticated methods for creating the feature set. This motivates the first application of multivariate feature selection with filters and false discovery rate analysis to biomedical NER, resulting in a feature set at least 3 orders of magnitude smaller than the set created by the rich feature set approach. Finally, two novel approaches to NER by modeling the semantics of token sequences are introduced. The first method focuses on the sequence content by using language models to determine whether a sequence resembles entries in a lexicon of entity names or text from an unlabeled corpus more closely. The second method models the distributional semantics of token sequences, determining the similarity between a potential mention and the token sequences from the training data by analyzing the contexts where each sequence appears in a large unlabeled corpus. The second method is shown to improve the performance of BANNER on multiple data sets. / Dissertation/Thesis / Ph.D. Computer Science 2013
196

ucsCNL A controlled natural language for use case specifications

HORI, Érica Aguiar Andrade 31 January 2010 (has links)
Made available in DSpace on 2014-06-12T15:57:41Z (GMT). No. of bitstreams: 2 arquivo3220_1.pdf: 1307302 bytes, checksum: 42435c33fd14be36778e3c202d24fd2d (MD5) license.txt: 1748 bytes, checksum: 8a4605be74aa9ea9d79846c1fba20a33 (MD5) Previous issue date: 2010 / A maioria das empresas utiliza a linguagem natural livre para documentar software, desde os seus requisitos, até os casos de uso e testes usados para verificar o produto final. Visto que as fases de análise, projeto, implementação e teste do sistema dependem essencialmente dessa documentação, é preciso assegurar inicialmente a qualidade desses textos. Contudo, textos escritos em linguagem natural nem sempre são precisos, devido ao fenômeno da ambigüidade (léxica e estrutural), podendo dar margem a diferentes interpretações. Uma alternativa para se minimizar esse problema é o uso de uma Linguagem Natural Controlada - um subconjunto de alguma língua natural, que usa um vocabulário restrito a um domínio particular, e regras gramaticais que guiam a construção de sentenças com redução de ambigüidade semântica visando padronização e precisão dos textos. Este trabalho, na área de Teste de Software, apresenta a ucsCNL (Use Case Specification CNL), uma Linguagem Natural Controlada para escrever especificações de casos de uso no domínio de dispositivos móveis. A ucsCNL foi integrada à TaRGeT (Test and Requirements Generation Tool), uma ferramenta para geração automática de casos de teste funcionais baseados em cenários de casos de uso escritos em Inglês. A ucsCNL provê um ambiente para geração de casos de teste mais claros, com ambigüidade reduzida, influindo diretamente na qualidade dos testes e na produtividade dos testadores. A ucsCNL já está em uso e tem alcançado resultados satisfatórios
197

Concept Maps Mining for Text Summarization

AGUIAR, C. Z. 31 March 2017 (has links)
Made available in DSpace on 2018-08-02T00:03:48Z (GMT). No. of bitstreams: 1 tese_11160_CamilaZacche_dissertacao_final.pdf: 5437260 bytes, checksum: 0c96c6b2cce9c15ea234627fad78ac9a (MD5) Previous issue date: 2017-03-31 / 8 Resumo Os mapas conceituais são ferramentas gráficas para a representação e construção do conhecimento. Conceitos e relações formam a base para o aprendizado e, portanto, os mapas conceituais têm sido amplamente utilizados em diferentes situações e para diferentes propósitos na educação, sendo uma delas a represent ação do texto escrito. Mes mo um gramá tico e complexo texto pode ser representado por um mapa conceitual contendo apenas conceitos e relações que represente m o que foi expresso de uma forma mais complicada. No entanto, a construção manual de um mapa conceit ual exige bastante tempo e esforço na identificação e estruturação do conhecimento, especialmente quando o mapa não deve representar os conceitos da estrutura cognitiva do autor. Em vez disso, o mapa deve representar os conceitos expressos em um texto. Ass im, várias abordagens tecnológicas foram propostas para facilitar o processo de construção de mapas conceituais a partir de textos. Portanto, esta dissertação propõe uma nova abordagem para a construção automática de mapas conceituais como sumarização de t extos científicos. A sumarização pretende produzir um mapa conceitual como uma representação resumida do texto, mantendo suas diversas e mais importantes características. A sumarização pode facilitar a compreensão dos textos, uma vez que os alunos estão te ntando lidar com a sobrecarga cognitiva causada pela crescente quantidade de informação textual disponível atualmente. Este crescimento também pode ser prejudicial à construção do conhecimento. Assim, consideramos a hipótese de que a sumarização de um text o representado por um mapa conceitual pode atribuir características importantes para assimilar o conhecimento do texto, bem como diminuir a sua complexidade e o tempo necessário para processá - lo. Neste contexto, realizamos uma revisão da literatura entre o s anos de 1994 e 2016 sobre as abordagens que visam a construção automática de mapas conceituais a partir de textos. A partir disso, construímos uma categorização para melhor identificar e analisar os recursos e as características dessas abordagens tecnoló gicas. Além disso, buscamos identificar as limitações e reunir as melhores características dos trabalhos relacionados para propor nossa abordagem. 9 Ademais, apresentamos um processo Concept Map Mining elaborado seguindo quatro dimensões : Descrição da Fonte de Dados, Definição do Domínio, Identificação de Elementos e Visualização do Mapa. Com o intuito de desenvolver uma arquitetura computacional para construir automaticamente mapas conceituais como sumarização de textos acadêmicos, esta pesquisa resultou na ferramenta pública CMBuilder , uma ferramenta online para a construção automática de mapas conceituais a partir de textos, bem como uma api java chamada ExtroutNLP , que contém bibliotecas para extração de informações e serviços públicos. Para alcançar o objetivo proposto, direcionados esforços para áreas de processamento de linguagem natural e recuperação de informação. Ressaltamos que a principal tarefa para alcançar nosso objetivo é extrair do texto as proposições do tipo ( conceito, rela ção, conceito ). Sob essa premissa, a pesquisa introduz um pipeline que compreende: regras gramaticais e busca em profundidade para a extração de conceitos e relações a partir do texto; mapeamento de preposição, resolução de anáforas e exploração de entidad es nomeadas para a rotulação de conceitos; ranking de conceitos baseado na análise de frequência de elementos e na topologia do mapa; e sumarização de proposição baseada na topologia do grafo. Além disso, a abordagem também propõe o uso de técnicas de apre ndizagem supervisionada de clusterização e classificação associadas ao uso de um tesauro para a definição do domínio do texto e construção de um vocabulário conceitual de domínios. Finalmente, uma análise objetiva para validar a exatidão da biblioteca Extr outNLP é executada e apresenta 0.65 precision sobre o corpus . Além disso, uma análise subjetiva para validar a qualidade do mapa conceitual construído pela ferramenta CMBuilder é realizada , apresentando 0.75/0.45 para precision / recall de conceitos e 0.57/ 0.23 para precision/ recall de relações em idioma inglês e apresenta ndo 0.68/ 0.38 para precision/ recall de conceitos e 0.41/ 0.19 para precision/ recall de relações em idioma português. Ademais , um experimento para verificar se o mapa conceitual sumarizado pe lo CMBuilder tem influência para a compreensão do assunto abordado em um texto é realizado , atingindo 60% de acertos para mapas extraídos de pequenos textos com questões de múltipla escolha e 77% de acertos para m apas extraídos de textos extensos com quest ões discursivas
198

Computer Forensic Text Analysis with Open Source Software / Kriminaltekniska textundersökningar med hjälp av öppen källkod

Johansson, Christian January 2003 (has links)
Detta papper koncentrerar sig på kriminaltekniska undersökningar av text, med fokus på användande av mjukvara med öppen källkod. Pappret diskuterar och undersöker olika tekniker för framtida automatisering av dessa undersökningar.
199

25 Challenges of Semantic Process Modeling

Mendling, Jan, Leopold, Henrik, Pittke, Fabian January 2014 (has links) (PDF)
Process modeling has become an essential part of many organizations for documenting, analyzing and redesigning their business operations and to support them with suitable information systems. In order to serve this purpose, it is important for process models to be well grounded in formal and precise semantics. While behavioural semantics of process models are well understood, there is a considerable gap of research into the semantic aspects of their text labels and natural language descriptions. The aim of this paper is to make this research gap more transparent. To this end, we clarify the role of textual content in process models and the challenges that are associated with the interpretation, analysis, and improvement of their natural language parts. More specifically, we discuss particular use cases of semantic process modeling to identify 25 challenges. For each challenge, we identify prior research and discuss directions for addressing them.
200

Komunikace s robotem přirozeným jazykem / Natural language communication with Robots

Pišl, Bedřich January 2017 (has links)
Interpreting natural language actions in a simulated world is the first step towards robots controlled by natural language commands. In this work we present several models for interpreting unrestricted natural language commands in a simple block world. We present and compare rule-based models and recurrent neural network models of various architectures. We also discuss strategies to deal with errors in natural language data and compare them. On the Language Grounding dataset, our models outperform the previous state-of-the-art results in both source and location prediction reaching source accuracy 98.8% and average distance 0.71 between the correct and predicted location.

Page generated in 0.0405 seconds