Maintaining a Federated Search Service

ABSTRACT

A federated search service does not stand still. Software changes on a fairly predictable schedule but content is constantly in flux as vendors make changes in their products and platforms and libraries and librarians make changes in their selection of products and vendors. It is important to have a plan for distribution of maintenance responsibilities and a workflow that integrates the maintenance of the federated search tool into existing routines. The extent to which these routines can be automated is a focus of this article. doi:10.1300/J136v12n03_05     

Protocol Analysis of a Federated Search Tool: Designing for Users

SUMMARY

Librarians at Springfield College conducted usability testing of Endeavor's federated search tool, ENCompass for Resource Access. The purpose of the testing was to make informed decisions prior to customizing the look and function of the software's interface in order to make the product more usable for their patrons. Protocol, or think-aloud, analysis was selected as a testing and analysis method. Subjects from the general college community were recruited and given a list of tasks to perform on ENCompass, and they were asked to speak all of their thoughts out loud as they worked. Upon analyzing the test results, researchers found that subjects' problems fell into certain categories, such as unfamiliarity with terms or navigation from screen to screen. The researchers were able to use their findings to recommend extensive revisions to the interface, which improved usability for this library's patrons.     

Librarian Perspectives on Teaching Metasearch and Federated Search Technologies

ABSTRACT

This article explores three assessment projects: two focused on librarians' reactions to metasearch technologies from a reference and information literacy perspective, and one user survey that attempts to capture student experiences, understanding, and satisfaction with metasearching at California State University Northridge. Exploring both librarian and user reactions to metasearch and federated search interface technologies in academic libraries, the authors review what the literature reveals about these technologies and their impact on information literacy initiatives. Emphasis focuses on how federated searching technologies affect librarians, students, the marketing of existing library and electronic databases, and information literacy instructional programming. doi:10.1300/J136v12n03_02     

个性化搜索引擎综述

分析了现有搜索引擎的基本原理和基本分类,研究了个性化搜索引擎的概念和基本分类,最后对个性化搜索引擎的发展提出了一些建议。     

浅析智能搜索引擎技术及发展

文章介绍了智能搜索引擎的特点,对智能搜索引擎中自然语言理解、智能搜索代理、多媒体信息检索等关键技术进行了详尽的分析,并展望智能搜索引擎未来的发展方向。     

一种解决联邦数据库系统查询处理的模式转换方法

本文在分析联邦数据库系统成员异构的特点后，论述了一种基于输入／输出模式和概念模式的转换方法，通过该转换处理可以有效地解决模式异构的问题，这种方法已在ＳＵ－ＦＤＢＳ原型中实现。     

User Perceptions of MetaLib Combined Search: An Investigation of How Users Make Sense of Federated Searching

SUMMARY

This article presents an investigation of users' understanding of MetaLib Combined Search (MCS), a federated search system implemented for the Washington Research Library Consortium. Through a survey instrument, librarians and students reported their experience, usage, and opinions of the system. Upon responding to process-related questions about a search simulation included in the questionnaire, participants described and illustrated their understanding of MCS operation. Data show that students considered MCS primarily as a tool for locating full text, while librarians viewed it as a secondary search tool with disappointing performance. In discussing MCS operation, students focused largely on full-text retrieval capability and search efficiency whereas librarians paid more attention to search strategies and retrieval quality. Both students and librarians indicated that the lack of background information about MCS operation was problematic. Both groups identified knowledge and skills related to federated searching for information literacy programs. Implications of the findings for information literacy, interface design, and system development were explored.     

桌面搜索引擎设计

针对目前主流桌面搜索引擎用户体验不好,索引实时性不高,系统资源占用率过高,查全率不高等缺点。该文设计出一款短小精悍,功能强大,性能卓越,界面友好的桌面搜索引擎。支持检索结果再检索,同时保证系统运行时cpu、内存和磁盘等较低的系统资源消耗。采用三种索引机制保证索引的实时性和系统的查全率。     

基于多Agent的搜索引擎设计与实现

搜索引擎的出现改变了人们获取信息的方式,利用搜索引擎可以快速地找到需要的信息,为我们在Intemet上获取信息提供了一种有效的手段。但随着Intemet的发展和网上信息量的激增,人们在使用中却发现要准确、快速地查找自己所需的信息是越来越困难。文章依据搜索引擎、Agent技术,提出了基于多Agent技术的智能搜索引擎概念,能够有效地提高搜索引擎的搜索质量和用户服务,为解决当前搜索引擎存在的一些问题提供了一种新的有效的方法     

Planning and Implementing a Federated Searching System: An Examination of the Crucial Roles of Technical,Functional, and Usability Testing

SUMMARY

Federated searching software offers much promise to users as a convenient way of accessing the wealth of electronic information resources libraries provide. But metasearching is not the same as Google searching; care must be taken in organizing and presenting search options and results so that they are comprehensible to users. From software installation to usability testing and creating documentation, most of the work of implementation is behind-the-scenes and hidden from most library staff and users; however, the decisions made during implementation greatly affect staff and user experiences with the product as well as its overall utility and usability. Systematic testing of the product is necessary to make informed and defensible decisions. This article details three layers of testing (technical, functional, and usability) recommended during implementation of a federated search product, based on best practices in the literature, metasearch standards, and the authors' own experiences with implementing a locally developed broadcast search system and the federated search system WebFeat.     

社交网络相关机制应用于搜索引擎的研究综述

首先对社交网络相关机制应用于搜索引擎的研究背景做简单陈述,然后对现有社交化搜索算法按照应用环境的不同进行分类综述与分析,总结其优势、不足以及影响因素。进而提出限制社交化搜索研究发展的主要原因,展望未来开展社交化搜索的相关研究时应当重点关注的方向,着重讨论利用个性化推荐系统连结社交网络与搜索引擎的思想,以及搜索引擎合法获取社交数据的可行的模式。     

搜索引擎的Web Robot技术与优化

针对目前如何使搜索引擎快速准确地从庞大的网页数据中获取所需资源的需求，文中对搜索引擎作了概述，重点阐述了搜索引擎的Web Robot搜索器的搜索策略、文档提取以及搜索优化措施，提出了改进搜索引擎的Web Rohot的方法，改进后的Web Robot能够更有效地发现和搜集信息。为搜索引擎进行信息搜集，高效、稳定的Web Rohot保证了为用户提供的网上信息的全面性和实时性。     

基于XML的搜索引擎相关度计算的改进

XML在网络上越来越广泛的应用,必然要求传统的搜索引擎技术作出相应的改进。本文在对XML文档结构特性分析的基础上,主要在索引词相关度计算方法方面进行了改进。     

基于知网与搜索引擎的词汇语义相似度计算

提出一种基于知网与搜索引擎的词汇语义相似度计算方法。利用义原在层次体系树的深度、密度、信息量优化义原的相似性计算。将逐点共有信息（PMI）算法与归一化谷歌距离（NGD）算法结合优化基于搜索引擎的词汇语义相似度计算。将词汇的词性作为权重因子融合知网与搜索引擎的词汇相似度计算结果。实验结果表明,与基于知网和基于搜索引擎的语义相似度计算方法相比,所提出的方法在NLPCC测试集上的平均相似度更接近于测试集的评测标准,在汽车票务领域的词汇相似度计算中具有较好的应用效果。     

企业专用搜索引擎的搜索策略

分析了现有的通用搜索引擎技术,并指出了其局限性。针对企业用户的需求,提出构建面向企业用户的专用搜索引擎的设计思想。该设计采用人工智能启发式搜索的思想,在企业用户设定的种子站点集和关键词集的基础上,根据网页和链接的评价值进行有针对性的搜索,提高了专用搜索引擎的效率,并为个性化的服务提供了基础。     

搜索引擎的现状及分析

在介绍了搜索引擎的工作原理、一般概念及分类后,较为详细地分析了目前搜索引擎存在的问题和发展趋势,提出了搜索引擎当前的主要研究内容。     

多元搜索引擎的现状及其典型系统的介绍

本文介绍了搜索引擎发展过程中出现的一种新技术——多元搜索引擎。分析了多元搜索引擎的现状，并对典型的多元搜索引擎系统进行了介绍。同时提出了多元搜索引擎发展过程中有待改进的一面。     

多元搜索引擎的现状及其典型系统的介绍

本文介绍了搜索引擎发展过程中出现的一种新技术——多元搜索引擎。分析了多元搜索引擎的现状,并对典型的多元搜索引擎系统进行了介绍。同时提出了多元搜索引擎发展过程中有待改进的一面。     

User Expectations in the Time of Google: Usability Testing of Federated Searching

SUMMARY

As libraries expand their services into the world of federated searching, librarians need to work with users to discover what their expectations are and how the library can customize the software to meet users' expectations. This article describes the user testing performed at Texas A&M University during 2005 as the libraries implemented a new federated search service called Search Now (ExLibris' MetaLib). Over fifty volunteers–including undergraduates, graduate students, faculty, and library faculty and staff–helped to test the new system and offered suggestions for improvements. Problems were noted and, where possible, modifications were made to improve results. These modifications were then tested again. Major issues noted during the usability testing included: user expectations of search performance; information included in and the layout of the search results; availability of advanced search options; and lack of ability to limit by format, scholarly nature of journal, date and full-text availability. Suggestions for further development are also presented.     

一种面向商业领域的搜索引擎索引结构设计及实现

本文通过分析传统搜索引擎技术不能满足商业领域搜索要求的灵活排序、特殊单字段搜索等需求,提出用新的索引结构和查询算法构建面向商业的搜索引擎系统并予以实现。另外,文章还讨论了其中的索引结构部分,是整个搜索引擎系统的基石。