Peter A. Ng

A document segmentation, classification and recognition system

A discussion is given on a document segmentation, classification and recognition system for automatically reading daily-received office documents that have complex layout structures, such as multiple columns and mixed-mode contents of texts, graphics and half-tone pictures. First, the block segmentation employs a two-step run-length smoothing algorithm for decomposing any document into single-mode blocks. Next, based on clustering rules the block classification classifies each block into one of text, horizontal or vertical lines, graphics, and pictures. The text block is separated into isolated characters using projection profiles, and which are translated into ASCII codes through a font- and size-independent character recognition subsystem. Logo pictures discriminated from half-tone pictures are identified and converted into symbolic words. The experimental results show that the proposed system is capable of correctly reading different styles of mixed-mode printed documents.<<ETX>>

TEXPROS: AN INTELLIGENT DOCUMENT PROCESSING SYSTEM

This paper presents the design of an intelligent document processing system, called TEXPROS. The system is a combination of filing and retrieval systems, which supports storing, extracting, classifying, categorizing, retrieving and browsing information from a variety of documents. TEXPROS is built based on object-oriented programming and rule-based specification techniques. In this paper, we describe main design goals of the system, its data model, logical file structure, and strategies for document classification and categorization. We also illustrate various retrieval methods and query processing techniques through examples. Finally applications of TEXPROS are presented, where we suggest ways in which the use of the system may alter the software process model.

Nested segmentation: an approach for layout analysis in document classification

Office information systems (OISs) are employed to support office workers in their management of information and to assist them in their daily work. In the OISs, document classification is one of the major functional capabilities. Classifying a document can be facilitated through the layout analysis of the document. A new approach to the layout analysis, called nested segmentation, is introduced. The layout relationships of components of a document are defined in terms of the adjacency of blocks. Given the adjacency of blocks, an adjacent block graph is introduced where the problem of the nested segmentation is transformed to a classic minimal cut problem for the graph. Also, an ordered labeled tree structure (L-S-Tree) is introduced to represent the segmented document for document classification.<<ETX>>

Further Analysis of the Entity-Relationship Approach to Database Design

The nondeterministic or deterministic entity-relationship model of a database is formaRy defined as a users view of that database in terms of a collection of time-varying relations: the regular or weak entity relations, or the regular or weak relationship relations. Both nondeterministic and deterministic entity-relationship models have the same strength to characterize information concerning entities and relationships which exist in our minds. An improved table form of the relations is introduced to provide a clear and concise users view of databases. The basic concept of the entity-relationship approach to the logical database design is provided, and is used to derive 3NF relations. FinaUly, a method of representing physicaly these relations, which are generated by the use of the entity-relationship approach to the logical database design, is presented. Thus, the entity-relationship approach to the logical and physical database design can also be realized.

Some thoughts on systems integration: A conceptual framework

This article presents a conceptual framework for systems integration in the area of computing systems. The framework identifies a hierarchical structure of concepts and methods that allows us to discuss the different elements of systems integration. The kernel of this approach is the concept of an integration architecture, specifying the general layout of the integration framework. Such an integration architecture is based on enabling technologies, like networking, and forms the basis for global integration activities. These global integration activities are targeted to preserve a common universe of discourse for the user of the integrated architecture and to agree upon a conceptual model of the application domain. Based on these prerequisites, two models for managing the process of systems integration, the post factor approach and the phased top-down approach, are discussed.

A tool for classifying office documents

The authors present the design of a tool for classifying office documents. They represent a documents layout structure using an ordered labeled tree, called the layout structure tree (L-S-tree), based on a nested segmentation procedure. The tool uses a sample-based approach for learning, where concepts are learned by retaining samples and new documents are classified by matching their L-S-trees with samples. The matching process involves both computing the edit distance between two trees using a previously developed pattern matching toolkit, and calculating the degree of conceptual closeness between the documents and samples. The experimental results show that the tool is capable of classifying various types of office documents, even with very few samples in the sample base.

A predicate-driven document filing system

This paper presents a predicate-driven document filing system for organizing and automatically filing documents. A document model consists of two basic elements: frame templates representing document classes, and folders which are repositories of frame instances. The frame templates can be organized to form a document type hierarchy, which helps classify and file documents. Frame instances are grouped into a folder on the basis of user-defined criteria, specified as predicates which determine whether a frame instance belongs to a folder. Folders can naturally organized into a folder organization which represents the users real world document filing system. The predicate consistency problem is discussed to eliminate two abnormalities from a folder organization: inapplicable edges (filing paths) and redundant folders. An evaluating net (including an association dictionary, an instantiation component and a production system) is then proposed for evaluating whether a frame instance satisfies the predicate of a folder during document filing. And the concept of consistency a rule base is also discussed.

A browser of supporting vague query processing in an office document system

DOCPROS (DOCument PROcessing System) is an intelligent document processing and retrieval system; it supports storing, extracting, classifying, categorizing, retrieving and browsing information from a variety of office documents. This article presents a portion of the retrieval subsystem of DOCPROS for processing vague queries in which the user may only have a vague retrieval target. That is, the user does not know exactly what kinds of information he needs until some kind of description is displayed to him. A browsing mechanism is employed for such situations to assist the user in the retrieval process.

HEURISTIC CLASSIFICATION OF OFFICE DOCUMENTS

Document Processing Systems (DPSs) support office workers to manage information. Document classification is a major function of DPSs. By analyzing a document’s layout and conceptual structures, we present in this paper a sample-based approach to document classification. We represent a document’s layout structure by an ordered labeled tree through a procedure known as nested segmentation and represent the document’s conceptual structure by a set of attribute type pairs. The layout similarities between the document to be classified and sample documents are determined by a previously developed approximate tree matching toolkit. The conceptual similarities between the documents are determined by analyzing their contents and by calculating the degree of conceptual closeness. The document type is identified by computing both the layout and conceptual similarities between the document to be classified and the samples in the document sample base. Some experimental results are presented, which demonstrate the effectiveness of the proposed techniques.

Analysis on access fairness and a technique to extend distance for 802.6

Even though IEEE 802.6 can cover longer distances than those of local area networks (LANs), it is limited by unfairness in terms of average access delay. The effect of distance on average access delay is analyzed based on homogeneous assumptions, and a technique to extend 802.6 over a wider area is proposed. The technique makes possible simplification of topology for wide area networks (WANs), which in turn simplifies routing and network management tasks.<<ETX>>