scholarly journals Capturing Logical Structure of Visually Structured Documents with Multimodal Transition Parser

Author(s):  
Yuta Koreeda ◽  
Christopher Manning
Author(s):  
M. LALMAS ◽  
T. ROLLEKE

Structured documents are composed of objects with a content and a logical structure. The effective retrieval of structured documents requires models that provide for a content-based retrieval of objects that takes into account their logical structure, so that the relevance of an object is not solely based on its content, but also on the logical structure among objects. This paper proposes a formal model for representing structured documents where the content of an object is viewed as the knowledge contained in that object, and the logical structure among objects is capture by a process of knowledge augmentation: the knowledge contained in an object is augmented with that of its structurally related objects. The knowledge augmentation process takes into account the fact that knowledge can be incomplete and become inconsistent.


Author(s):  
Ludovic Denoyer

Document classification developed over the last ten years, using techniques originating from the pattern recognition and machine learning communities. All these methods do operate on flat text representations where word occurrences are considered independents. The recent paper (Sebastiani, 2002) gives a very good survey on textual document classification. With the development of structured textual and multimedia documents, and with the increasing importance of structured document formats like XML, the document nature is changing. Structured documents usually have a much richer representation than flat ones. They have a logical structure. They are often composed of heterogeneous information sources (e.g. text, image, video, metadata, etc). Another major change with structured documents is the possibility to access document elements or fragments. The development of classifiers for structured content is a new challenge for the machine learning and IR communities. A classifier for structured documents should be able to make use of the different content information sources present in an XML document and to classify both full documents and document parts. It should easily adapt to a variety of different sources (e.g. to different Document Type Definitions). It should be able to scale with large document collections.


Author(s):  
Ludovic Denoyer ◽  
Patrick Gallinari

Document classification developed over the last 10 years, using techniques originating from the pattern recognition and machine-learning communities. All these methods operate on flat text representations, where word occurrences are considered independents. The recent paper by Sebastiani (2002) gives a very good survey on textual document classification. With the development of structured textual and multimedia documents and with the increasing importance of structured document formats like XML, the document nature is changing. Structured documents usually have a much richer representation than flat ones. They have a logical structure. They are often composed of heterogeneous information sources (e.g., text, image, video, metadata, etc.). Another major change with structured documents is the possibility to access document elements or fragments. The development of classifiers for structured content is a new challenge for the machine-learning and IR communities. A classifier for structured documents should be able to make use of the different content information sources present in an XML document and to classify both full documents and document parts. It should adapt easily to a variety of different sources (e.g., different document type definitions). It should be able to scale with large document collections.


1993 ◽  
Vol 32 (04) ◽  
pp. 272-273 ◽  
Author(s):  
A. L. Rector

Response to: Essin DJ. Intelligent processing of loosely structured documents as a strategy for organizing electronic health care records. Meth Inform Med 1993; 32: 265.


1993 ◽  
Vol 32 (04) ◽  
pp. 265-268 ◽  
Author(s):  
D. J. Essin

AbstractLoosely structured documents can capture more relevant information about medical events than is possible using today’s popular databases. In order to realize the full potential of this increased information content, techniques will be required that go beyond the static mapping of stored data into a single, rigid data model. Through intelligent processing, loosely structured documents can become a rich source of detailed data about actual events that can support the wide variety of applications needed to run a health-care organization, document medical care or conduct research. Abstraction and indirection are the means by which dynamic data models and intelligent processing are introduced into database systems. A system designed around loosely structured documents can evolve gracefully while preserving the integrity of the stored data. The ability to identify and locate the information contained within documents offers new opportunities to exchange data that can replace more rigid standards of data interchange.


2019 ◽  
Vol 48 (1) ◽  
pp. 57-89
Author(s):  
Mareike Schildmann

Abstract This article traces some of the fundamental poetological changes that the traditional crime novel undergoes in the work of the Swiss author Friedrich Glauser at the beginning of the 20th century. The rational-analytical, conservative approach of the criminal novel in the 19th century implied – according to Luc Boltanski – the separation of an epistemologically structured, institutionalized order of “reality” and a chaotic, unruly, unformatted “world” – a separation that is questioned, but reestablished in the dramaturgy of crime and its resolution. By shifting the attention from the logical structure of ‘whodunnit’ to the sensual material culture and “atmosphere” that surrounds actions and people, Glauser’s novels blur these epistemological and ontological boundaries. The article shows how in Die Fieberkurve, the second novel of Glauser’s famous Wachtmeister Studer-series, material and sensual substances develop a specific, powerful dynamic that dissipates, complicates, crosslinks, and confuses the objects and acts of investigation as well as its narration. The material spoors, dust, fibers, fingerprints, intoxicants and natural resources like oil and gas – which lead the investigation from Switzerland to North Africa – trigger a new sensual mode of perception and reception that replaces the reassuring criminological ideal of solution by the logic of “dissolution”. The novel thereby demonstrates the poetic impact of the slogan of modernity: matter matters.


Sign in / Sign up

Export Citation Format

Share Document