基于SVM的网络文本信息自动分类

介绍基于SVM的网络文本信息自动分类算法,该算法在训练阶段将一个大型数据集分成许多不相交的子集,按批次对各个训练子集中的样本进行训练而得到多个分类器,利用误差纠错输出编码优化分类器,从而减少较深层次训练需要学习的文档.     

基于关键词抽取的hypertext自动建立方法

随着Internet的发展,电子文档的数量成指数级增长,大量的文档之间存在密切的联系。将这些电子文档发布到WWW上需要有效地建立这些大量文档之间的链接,从而为用户提供一个更加友好的导航界面。对于以超文拳形式产生出来的大量文档,用手工的方式为其指定超链接,不但需要领域知识,而且将是一项极为繁重的劳动。因此,实现超文本建立的自动化是一项很有意义的工作。目前的各种超链建立方法存在着自动化程度不高和准确率低的缺点。本文基于关键词自动抽取提出了一种为文档自动建立超链接的方法。实验证明该方法取得了较好的效果。     

Automatic Presentations and Semigroup Constructions

An automatic presentation for a relational structure is, informally, an abstract representation of the elements of that structure by means of a regular language such that the relations can all be recognized by finite automata. A structure admitting an automatic presentation is said to be FA-presentable. This paper studies the interaction of automatic presentations and certain semigroup constructions, namely: direct products, free products, finite Rees index extensions and subsemigroups, strong semilattices of semigroups, Rees matrix semigroups, Bruck–Reilly extensions, zero-direct unions, semidirect products, wreath products, ideals, and quotient semigroups. For each case, the closure of the class of FA-presentable semigroups under that construction is considered, as is the question of whether the FA-presentability of the semigroup obtained from such a construction implies the FA-presentability of the original semigroup[s]. Classifications are also given of the FA-presentable finitely generated Clifford semigroups, completely simple semigroups, and completely 0-simple semigroups.     

基于高斯混合模型的现代汉语构式成分自动标注方法

现代汉语构式成分自动标注作为文本自动标注任务之一,其最大的困难在于,当不存在标注语料作为训练数据时,如何从生语料中挖掘不同类型的构式成分相关的知识并进行标注,特别是面对构式序列在句中的边界难以判断的情况。该文试图借助高斯混合模型聚类方法,结合句中每一个字的位置特征与构式形式本身的语言学特征,融合正则表达式匹配结果信息,挖掘句子中的构式实例序列,并对构式内部成分进行自动标注。相较于仅基于正则表达式匹配和词性匹配的自动标注结果,该方法的F₁分别至少提高了17.9%(半凝固型构式)、19.3%(短语型构式)、14.9%(复句型构式)。     

利用超链分析自动获取网页的主题词描述

本文介绍超级链接（Hypertext Link）中超链文字的作用，探讨如何利用超链文字的信息自动获取网页文件的主题词描述，以及该算法在搜索引擎中的相关度评价（relevance judgement）中的重要作用。     

Hypertext: A new information technology

The hypertext technology is described, including a survey of hypertext systems and their artificial-intelligence tools. The design principles of the first Ukrainian hypertext system HYPSY are discussed.Translated from Kibernetika i Sistemnyi Analiz, No. 5, pp. 116–135, September–October, 1992.     

Hypertext functionality: A theoretical framework

This article presents a theoretical framework useful for understanding the functionality of Hypertext systems in terms of their ability to satisfy cognitive requirements for expressing concepts and their relationships. This theoretical framework is derived from a re‐interpretation of Guilford's theory of the Structure of Intellect.     

A Study of Approaches to Hypertext Categorization

Hypertext poses new research challenges for text classification. Hyperlinks, HTML tags, category labels distributed over linked documents, and meta data extracted from related Web sites all provide rich information for classifying hypertext documents. How to appropriately represent that information and automatically learn statistical patterns for solving hypertext classification problems is an open question. This paper seeks a principled approach to providing the answers. Specifically, we define five hypertext regularities which may (or may not) hold in a particular application domain, and whose presence (or absence) may significantly influence the optimal design of a classifier. Using three hypertext datasets and three well-known learning algorithms (Naive Bayes, Nearest Neighbor, and First Order Inductive Learner), we examine these regularities in different domains, and compare alternative ways to exploit them. Our results show that the identification of hypertext regularities in the data and the selection of appropriate representations for hypertext in particular domains are crucial, but seldom obvious, in real-world problems. We find that adding the words in the linked neighborhood to the page having those links (both inlinks and outlinks) were helpful for all our classifiers on one data set, but more harmful than helpful for two out of the three classifiers on the remaining datasets. We also observed that extracting meta data from related Web sites was extremely useful for improving classification accuracy in some of those domains. Finally, the relative performance of the classifiers being tested provided insights into their strengths and limitations for solving classification problems involving diverse and often noisy Web pages.     

超文本形式化模型研究

本文谰述了目前超文本系统中常用的形式化模型，它们分别基于图论，网论，逻辑和集论。     

一种基于语料库的日语动词格框架自动构造技术

结合一个基于格语法的日汉机器翻译系统,针对现有系统日语动词格框架缺乏的现状,提出一种日语动词格框架自动构造方法.该方法基于类比与统计思想,从现有的少量人工构造的动词格框架和大量生语料出发,自动构造日语动词格框架.     

PROBLEM-SOLVING SUPPORT FOR TQM A Hypertext Approach

New IS technology is contributing to continuous quality improvement throughout the organization. Hypertext and hypermedia offer new ways for decision makers to gain easy and relevant access to information. By combining hypertext with total quality tools, decision makers are able to explore the richness of data bases in more creative and useful ways. This article describes how you can develop a hypertext system that can help employees improve their job performance through better problem solving.     

Hypertext and CASE

Computer-aided software engineering (CASE) systems require a database system that must let designers logically associate documentation and source code, make annotations to record explanations and assumptions, and manage different versions of the software. The author reports the development of a database, Neptune, which demonstrates that hypertext provides an appropriate data model for CASE systems. Hypertext is a medium-grained, entity-relationship-like data model that lets information be structured arbitrarily and keeps a complete version history of both information and structure. Neptune achieves application independence by using a layered system architecture; at the bottom is a transaction-based server, the Hypertext Abstract Machine, and layered above it are applications and the user interface. The use of Neptune in the Dynamic Design CASE environment is discussed     

The Rhetoric of Hypertext

This study adopts as a fundamental orientation the view that hypertext may eventually bring about a paradigm shift in text delivery and in human information processing. However, paradigm shifts do not occur overnight; they are evolutionary rather than revolutionary. Because of the considerable commitment of Western knowledge and culture to the written word and to linear text, it seems likely that successful hypertext systems will — at least in the beginning — electronically emulate many of the strategies a sophisticated reader uses in dealing with hard copy. This article describes a project which modeled the functionality an intelligent reader would bring to paper text (specifically, a maintenance manual), and designed and implemented a hypertext version of the document using Apple Computer's HyperCard.

Since few fully specified hypertext systems have been deployed, the research described here was undertaken to build a platform on which notions about the efficacy of non-linear text processing for a specific application could be tried out. This article describes a Stackware version of a portion of an aircraft maintenance manual and considers issues of (1) data storage models, (2) user interface, and (3) information retrieval methods for online text. These elements — among others — are the essence of the new rhetoric for non-sequential text.     

Meta-Interpretation and Hypertext Fiction: A Critical Response

Traditional discourses upon literature havebeen predicated upon the ability to refer to atext that others may consult (Landow, 1994,p. 33). Texts that involve elements of feedbackand non-trivial decision-making on the part ofthe reader (Aarseth, 1997, p. 1) therefore presenta challenge to readers and critics alike. Sincea persuasive case has been made against acritical method that sets out to ``identify thetask of interpretation as a task of territorialexploration and territorial mastery' (Aarseth,p. 87), this paper proposes the use of readers inan empirically based approach to hypertextfiction. Meta-interpretation, a method thatcombines individual responses to a text,reading logs, screen recordings and limitedqualitative/quantitative analysis, and criticalinterpretation is outlined. By analysingreaders' responses it is possible to suggestboth the ways that textual elements may haveinfluenced or determined readers' choices andthe ways that readers' choices ``configure' thetext. The method thus addresses Espen Aarseth'sconcerns and illuminates interesting featuresof interactive processes in fictionalenvironments. The paper is divided into twoparts: the first part sketches outmeta-interpretation through consideration ofthe main problems confronting the literarycritic; the second part describes readingresearch aimed at generating data for theliterary critic.     

Interacting With Hypertext: A Meta-Analysis of Experimental Studies

The meta-analysis compared and synthesized the results of 23 experimental studies on hypertext. The analysis was based on 56 pairs of effect sizes and significance levels of the impact of users, tasks, and tools on interactions with hypertext. This analysis focused on three factors that prevailingly influence the use of hypertext: the cognitive styles and spatial ability of users; the complexity of tasks; and the structure of information organization and the visualization of the structure. The meta-analysis found that this group of experimental studies reported significantly discrepant findings, indicating that substantial differences exist among individual experiments. Individual differences in cognition did not yield enough evidence to conclude that the effect sizes are significantly apart from zero. The meta-analysis showed that the overall performance of hypertext users tended to be more effective than that of nonhypertext users, but the differences in efficiency measures were consistently in favor of nonhypertext users. Users benefited more from hypertext tools for open tasks. Overall, the complexity of tasks has the largest combined effect sizes. Graphical maps that visualize the organization of hypertext have significant impact on the usefulness of a hypertext system. This meta-analysis raised two issues concerned with the present hypertext literature: (a) the absence of a taxonomy of tasks for analyzing and comparing hypertext usability across studies, and (b) the weaknesses of the connections between abstract hypertext reference models and specific hypertext systems. These weaknesses may considerably undermine the significance of individual findings on hypertext usability. Results of the meta-analysis suggest that the discrepancies among empirical findings are related to these weaknesses. Future work on hypertext usability should emphasize task taxonomies along with longitudinal and ethnographic studies for a deep understanding of the interactions between users and hypertext. Recommended research issues for the future are highlighted in Section 5.     

Issues in Hypertext Interchange

An important area of hypertext research focuses on the establishment of reference models and standards. In addition to defining the essential elements of hypertext systems, these standards can be elaborated as data structures and used as an intermediate format for hypertext interchange. The resulting interchange format is then used by a set of program modules which perform hypertext interchange between existing hypertext systems. This paper discusses the major issues in static and dynamic interchange of hypertexts. Definitions, methodology and goals of hypertext interchange are given. Static issues discussed include: ideal mappings, node size, node contents, multiple destinations, overlapping anchors, link directionality, partial hypertexts, system attributes and other attributes. Dynamic issues discussed include: system support, access to remote systems, identical remapping and partial hypertexts. Practical solutions to these issues are given where possible. A case study discusses the results of a research project that used the Dexter Hypertext Reference Model as the intermediate format in the transfer of hypertexts from Intermedia to KMS. An overview of the Dexter model is given and the Dexter interchange format, an implementation of the data modelling aspects of the Dexter model, is described. Issues and difficulties with hypertext interchange that arose in this implementation are presented and discussed. A list of open research issues concludes the paper.     

Tailoring Automatically Generated Hypertext

This paper describes an approach for tailoring the content and structure of automatically generated hypertext. The implemented system HYLITE is based on applied Natural Language Generation (NLG) techniques, a re-usable user modelling component (VIEWGEN), and a flexible architecture with module feedback. The user modelling component is used by the language generation modules to adapt the hypertext content and links to user beliefs and preferences and to the previous interaction. Unlike previous adaptive NLG systems, which have their own, application-specific user models, HYLITE has re-used a generic agent modelling framework (VIEWGEN) instead. Apart from avoiding the development costs of a new model, this also enabled a more extendable system architecture. Another distinct feature of our approach is making NLG techniques adaptable by the user, i.e., providing users with control over the user model and the hypertext adaptivity.