首页> 外国专利> METHOD OF INTERACTIVE SEARCH IN DISTRIBUTED COMPUTER NETWORKS AND INFORMATION RETRIEVAL SYSTEM TO THIS EFFECT

METHOD OF INTERACTIVE SEARCH IN DISTRIBUTED COMPUTER NETWORKS AND INFORMATION RETRIEVAL SYSTEM TO THIS EFFECT

机译:分布式计算机网络的交互搜索方法及信息检索系统

摘要

FIELD: automation.;SUBSTANCE: invention relates to automation of training and scientific researches and can be used in interactive systems of automation of research works. The method of interactive search in distributed computer networks includes integration of processes of formation of interactive search inquiries and digital conversions of the results of measurements of corresponding signals in every channel of formation of search inquiries. Here, input signals of search inquiries are separated into two groups, the first group incorporating time-independent interactive search inquiries and the second one including those time-dependent. Now, the channels of measurement are selected for the aforesaid both search inquiries signals, a combinatory polling is effected, for example, exhaustive search of combinations, an appropriate number of channels of measurement and measurement of parameters of signals in each channel of search inquiries formation. Here, measurement of signals from the first group is effected on the basis of values of scatter of their lower and upper limits of variations allowing for tolerable, for example, absolute error of measurement in corresponding time intervalTj=ti1+1-ti1 of measurement for all t∈[ti, ti1+1], i1=1, 2, etc. Now, the duration of the temporary intervalTi1 of formation of interactive search inquiries is corrected on the basis of values of the signal of mismatch of the affinity of the document content to a query or information need in previous interval Ti1-1 of formation of search inquiries. Also, the system of implementation of the proposed method is claimed.;EFFECT: higher efficiency of interactive search processes, higher accuracy, reliability and objectivity of search results.;3 cl, 16 dwg
机译:技术领域本发明涉及培训和科学研究的自动化,并且可以用于研究工作自动化的交互式系统。分布式计算机网络中的交互式搜索方法包括交互式搜索查询的形成过程的集成以及搜索查询形成的每个通道中相应信号的测量结果的数字转换。在此,搜索查询的输入信号被分为两组,第一组包括与时间无关的交互式搜索查询,第二组包括与时间有关的交互式搜索查询。现在,为上述两个搜索查询信号选择测量通道,进行组合轮询,例如,组合的穷举搜索,适当数量的测量通道和每个搜索查询通道中信号参数的测量。 。在此,来自第一组的信号的测量是基于它们的变化的上下限的分散值来实现的,例如允许在相应的时间间隔T j =中测量的绝对误差所有t&isin的测量值的t i1 + 1 -t i1 ; [t i ,t i1 + 1 ] ,i 1 = 1、2等。现在,根据的信号值校正交互式搜索查询形成的临时间隔T i1 的持续时间。搜索内容形成的先前间隔T i1-1 中文档内容与查询或信息需求的亲和力不匹配。效果:交互式搜索过程的效率更高,搜索结果的准确性,可靠性和客观性更高。3cl,16 dwg

著录项

相似文献

  • 专利
  • 外文文献
  • 中文文献
获取专利

客服邮箱:kefu@zhangqiaokeyan.com

京公网安备:11010802029741号 ICP备案号:京ICP备15016152号-6 六维联合信息科技 (北京) 有限公司©版权所有
  • 客服微信

  • 服务号