Ferilli S. Automatic Digital Document Processing and Management. Problems, Algorithms and Techniques

Springer, 2011. — 325 p.Automatic document processing plays a crucial role in the present society, due to the progressive spread of computer-readable documents in everyday life, from informal uses to more official exploitations. This holds not only for new documents, typically born digital, but also for legacy ones that undergo a digitization process in order to be exploited in computer-based environments. In turn, the increased availability of digital documents has caused a corresponding increase in users’ needs and expectations. It is a very hot topic in these years, for both academy and industry, as witnessed by several flourishing research areas related to it and by the ever-increasing number and variety of applications available on the market. Indeed, the broad range of document kinds and formats existing today makes this subject a many-faceted and intrinsically multi-disciplinary one that joins the most diverse branches of knowledge, covering the whole spectrum of humanities, science and technology. It turns out to be a fairly complex domain even focusing on the Computer Science perspective alone, since almost all of its branches come into play in document processing, management, storage and retrieval, in order to support the several concerns involved in, and to solve the many problems raised from, application to real-world tasks. The resulting landscape calls for a reference text where all involved aspects are collected, described and related to each other.
This book concerns Automatic Digital Document Processing and Management, where the adjective ‘digital’ is interpreted as being associated to ‘processing and management’ rather than to ‘document’, thus including also digitized documents in the focus of interest, in addition to born-digital ones. It is conceived as a survey on the different issues involved in the principal stages of a digital document’s life, aimed at providing a sufficiently complete and technically valid idea of the whole range of steps occurring in digital document handling and processing, instead of focusing particularly on any specific one of them. For many of such steps, fundamentals and established technology (or current proposals for questions still under investigation) are presented. Being the matter too wide and scattered, a complete coverage of the significant literature is infeasible. More important is making the reader acquainted of the main problems involved, of the Computer Science branches suitable for tackling them, and of some research milestones and interesting approaches available. Thus, after introducing each area of concern, a more detailed description is given of selected algorithms and techniques proposed in this field along the past decades. The choice was not made with the aim of indicating the best solutions available in the state-of-the-art (indeed, no experimental validation result is reported), but rather for the purpose of comparing different perspectives on how the various problems can be faced, and possibly complementary enough to give good chance of fruitful integration.Digital Documents
Documents
Digital Formats
Legal and Security Aspects
Document Analysis
Image Processing
Document Image Analysis
Content Processing
Natural Language Processing
Information Management
A A Case Study: DOMINUS
B Machine Learning Notions