METHOD AND SYSTEM FOR CREATING BOTH INITIATIVE MULTI-MODAL DIALOGUE AND RELATED BROWSING MECHANISM

    公开(公告)号:JP2004005676A

    公开(公告)日:2004-01-08

    申请号:JP2003142156

    申请日:2003-05-20

    Applicant: IBM

    Abstract: PROBLEM TO BE SOLVED: To provide a both initiative multi-modal dialogue for using in a dialogue system, declarative type programming relevant to natural language understanding, preparation technique for an application using other programming and other information browsing mechanism relevant to it. SOLUTION: This system comprises a step expressing a dialogue which a user can perform with a dialogue system as one data model or as a one or a plurality of user dialogue components transplanting an application condition of the data model and bound to it, so that the dialogue which the user performs with the dialogue system is a both initiative dialogue. At least a part of one or a plurality of user dialogue components can be converted to or correlated with rendering peculiar to a one or a plurality of forms in the application. The rendering can be shown to the user, and is selected or generated by a dialogue manager algorithm. COPYRIGHT: (C)2004,JPO

    Method for merging search results, computer program and system
    3.
    发明专利
    Method for merging search results, computer program and system 有权
    搜索结果,计算机程序和系统的方法

    公开(公告)号:JP2011081802A

    公开(公告)日:2011-04-21

    申请号:JP2010226260

    申请日:2010-10-06

    Abstract: PROBLEM TO BE SOLVED: To disclose a method for merging search results, system, and computer program product. SOLUTION: In one embodiment, the method includes steps of identifying a query, splitting the query into sub-queries, and calculating information content for each of the sub-queries. This method also includes the steps of executing each of the sub-queries to obtain a plurality of search results, and combining the search results based on the information content calculated for sub-queries. In an embodiment, the execution of each of the sub-queries includes a step of identifying a multitude of search results for at least one of the sub-queries, and the combining step includes the steps of grouping the multitude of search results into a plurality of clusters, and computing a relevance score for each of the clusters. In the embodiment, the combining step further includes merging the clusters based on the relevance scores computed for the clusters as well as the information content calculated for the sub-queries. COPYRIGHT: (C)2011,JPO&INPIT

    Abstract translation: 要解决的问题:公开一种合并搜索结果,系统和计算机程序产品的方法。 解决方案:在一个实施例中,该方法包括识别查询,将查询分解为子查询以及计算每个子查询的信息内容的步骤。 该方法还包括以下步骤:执行每个子查询以获得多个搜索结果,并且基于为子查询计算的信息内容来组合搜索结果。 在一个实施例中,每个子查询的执行包括识别至少一个子查询的多个搜索结果的步骤,并且组合步骤包括以下步骤:将多个搜索结果分组为多个 的集群,并计算每个集群的相关性分数。 在实施例中,组合步骤还包括基于针对集群计算的相关性分数以及为子查询计算的信息内容来合并集群。 版权所有(C)2011,JPO&INPIT

    Beheben von Ressourcenüberlastung

    公开(公告)号:DE102013206172A1

    公开(公告)日:2013-10-24

    申请号:DE102013206172

    申请日:2013-04-09

    Applicant: IBM

    Abstract: Ein Verfahren, eine Vorrichtung und ein Herstellungsartikel zum Beheben von Überlastung in einer überbelegten Datenverarbeitungsumgebung. Das Verfahren beinhaltet Messen der Ressourcennutzung jeder der vielen virtuellen Maschinen in jedem von mindestens einem Hypervisor in einer Datenverarbeitungsumgebung, bei Erkennung einer Ressourcenüberlastung in einem von mindestens einem Hypervisor Ermitteln von mindestens einer Operation, die für mindestens eine der vielen virtuellen Maschinen im Hypervisor vorzunehmen ist, um die Ressourcenüberlastung zu beheben und dabei die Werte der im Betrieb befindlichen virtuellen Maschinen zu steigern, und Senden eines Befehls an den Hypervisor, um die mindestens eine Operation auszugeben.

Patent Agency Ranking