利索-全球专利检索

  1. 登录
  2. 注册

搜索结果

6 条记录 (耗时 0.016 秒)

    Aggregating results from named entity recognition services
    1.
    发明授权
    Aggregating results from named entity recognition services 有权

    公开(公告)号：US09721002B2

    公开(公告)日：2017-08-01

    申请号：US14093227

    申请日：2013-11-29

    申请人： Katja Pfeifer ,  Johannes Meinecke

    发明人： Katja Pfeifer ,  Johannes Meinecke

    IPC分类号： G06F17/30 ,  G06F17/27

    CPC分类号： G06F17/30684 ,  G06F17/278 ,  G06F17/30657 ,  G06F17/30696

    摘要： An aggregation service aggregates extraction results from diverse named entity recognition (“NER”) services, which can help improve the quality of extracted information. In some cases, the aggregation service considers differences in entity type classifications when aggregating extraction results from different NER services. The aggregation service can also consider performance characteristics (e.g., error rates) for the different NER services. For example, the aggregation service receives extraction results generated for a document corpus according to an entity type schema for each of multiple different NER services, then aggregates the extraction results based at least in part on relations between entity types for the NER services. For a given annotation area, the computing system can identify hypotheses and rank the hypotheses according to an aggregation approach. For some types of aggregation approach, the computing system uses weight values, error path values and/or other performance characteristics determined during training for the NER services.

    Flexibly performing reallocations in databases
    3.
    发明授权
    Flexibly performing reallocations in databases 有权
    灵活地执行数据库中的重新分配

    公开(公告)号：US09460130B2

    公开(公告)日：2016-10-04

    申请号：US14259938

    申请日：2014-04-23

    申请人： Johannes Meinecke ,  Gregor Hackenbroich ,  Lars Dannecker ,  Dave Parsons ,  Paul Binks ,  Simon Moscrop ,  Jeremie Brunet

    发明人： Johannes Meinecke ,  Gregor Hackenbroich ,  Lars Dannecker ,  Dave Parsons ,  Paul Binks ,  Simon Moscrop ,  Jeremie Brunet

    IPC分类号： G06F17/30

    CPC分类号： G06F17/30339 ,  G06F17/303 ,  G06F17/30498 ,  G06F17/30507

    摘要： A reallocation processing block including a computing system including one or more data processors receives a base table, a reference table, and at least one assignment path table. Subsequently, rules from the at least one assignment path table are applied to the base table and the reference table by reallocating values between at least two existing data objects. A results table is generated with the reallocated values in the at least two existing data objects. A reallocated value is compared with a threshold value to determine the need for an iteration. At least one of the activities described is implemented using at least one data processor. Related apparatus, systems, techniques and articles are also described.

    摘要翻译： 包括包括一个或多个数据处理器的计算系统的重新分配处理块接收基表，参考表和至少一个分配路径表。 随后，通过在至少两个现有数据对象之间重新分配值，将来自至少一个分配路径表的规则应用于基准表和参考表。 使用至少两个现有数据对象中的重新分配值生成结果表。 将重新分配的值与阈值进行比较以确定迭代的需要。 使用至少一个数据处理器来实现描述的活动中的至少一个。 还描述了相关设备，系统，技术和物品。

    FLEXIBLY PERFORMING REALLOCATIONS IN DATABASES
    5.
    发明申请
    FLEXIBLY PERFORMING REALLOCATIONS IN DATABASES 有权
    灵活地执行数据库中的重新安装

    公开(公告)号：US20150154186A1

    公开(公告)日：2015-06-04

    申请号：US14259938

    申请日：2014-04-23

    申请人： Johannes Meinecke ,  Gregor Hackenbroich ,  Lars Dannecker ,  Dave Parsons ,  Paul Binks ,  Simon Moscrop ,  Jeremie Brunet

    发明人： Johannes Meinecke ,  Gregor Hackenbroich ,  Lars Dannecker ,  Dave Parsons ,  Paul Binks ,  Simon Moscrop ,  Jeremie Brunet

    IPC分类号： G06F17/30

    CPC分类号： G06F17/30339 ,  G06F17/303 ,  G06F17/30498 ,  G06F17/30507

    摘要： A reallocation processing block including a computing system including one or more data processors receives a base table, a reference table, and at least one assignment path table. Subsequently, rules from the at least one assignment path table are applied to the base table and the reference table by reallocating values between at least two existing data objects. A results table is generated with the reallocated values in the at least two existing data objects. A reallocated value is compared with a threshold value to determine the need for an iteration. At least one of the activities described is implemented using at least one data processor. Related apparatus, systems, techniques and articles are also described.

    摘要翻译： 包括包括一个或多个数据处理器的计算系统的重新分配处理块接收基表，参考表和至少一个分配路径表。 随后，通过在至少两个现有数据对象之间重新分配值，将来自至少一个分配路径表的规则应用于基准表和参考表。 使用至少两个现有数据对象中的重新分配值生成结果表。 将重新分配的值与阈值进行比较以确定迭代的需要。 使用至少一个数据处理器来实现描述的活动中的至少一个。 还描述了相关设备，系统，技术和物品。

    AGGREGATING RESULTS FROM NAMED ENTITY RECOGNITION SERVICES
    6.
    发明申请
    AGGREGATING RESULTS FROM NAMED ENTITY RECOGNITION SERVICES 有权
    来自匿名实体识别服务的结果

    公开(公告)号：US20150154284A1

    公开(公告)日：2015-06-04

    申请号：US14093227

    申请日：2013-11-29

    申请人： Katja Pfeifer ,  Johannes Meinecke

    发明人： Katja Pfeifer ,  Johannes Meinecke

    IPC分类号： G06F17/30

    CPC分类号： G06F17/30684 ,  G06F17/278 ,  G06F17/30657 ,  G06F17/30696

    摘要： An aggregation service aggregates extraction results from diverse named entity recognition (“NER”) services, which can help improve the quality of extracted information. In some cases, the aggregation service considers differences in entity type classifications when aggregating extraction results from different NER services. The aggregation service can also consider performance characteristics (e.g., error rates) for the different NER services. For example, the aggregation service receives extraction results generated for a document corpus according to an entity type schema for each of multiple different NER services, then aggregates the extraction results based at least in part on relations between entity types for the NER services. For a given annotation area, the computing system can identify hypotheses and rank the hypotheses according to an aggregation approach. For some types of aggregation approach, the computing system uses weight values, error path values and/or other performance characteristics determined during training for the NER services.

    摘要翻译： 聚合服务聚合来自不同命名实体识别（“NER”）服务的提取结果，这有助于提高提取信息的质量。 在某些情况下，聚合服务会考虑在不同NER服务的聚合提取结果时实体类型分类中的差异。 聚合服务还可以考虑不同NER服务的性能特征（例如，错误率）。 例如，聚合服务接收根据用于多个不同NER服务中的每一个的实体类型模式为文档语料库生成的提取结果，然后至少部分地基于NER服务的实体类型之间的关系聚合提取结果。 对于给定的注释区域，计算系统可以根据聚合方法识别假设并对假设进行排序。 对于某些类型的聚合方法，计算系统使用在训练期间为NER服务确定的权重值，误差路径值和/或其他性能特征。