Event identification in sensor analytics
    4.
    发明授权
    Event identification in sensor analytics 有权
    传感器分析中的事件识别

    公开(公告)号:US08620624B2

    公开(公告)日:2013-12-31

    申请号:US12241266

    申请日:2008-09-30

    IPC分类号: G06F11/00

    CPC分类号: H04L41/0618 H04L41/12

    摘要: A method of detecting an event anomaly includes receiving one or more data points, in which each data point represents a spatial or temporal event; associating a unique identifier with each of the one or more data points to obtain one or more individualized data points; distributing the one or more individualized data points across a grid, in which the grid includes one or more cells; determining an event likelihood ratio for one or more of the grid cells; identifying one or more event clusters, in which each event cluster includes one or more of the grid cells; and storing in a data repository an event cluster having a significance level above a threshold significance level.

    摘要翻译: 检测事件异常的方法包括接收一个或多个数据点,其中每个数据点表示空间或时间事件; 将唯一标识符与所述一个或多个数据点中的每一个相关联以获得一个或多个个性化数据点; 在网格中分布一个或多个个体化数据点,其中网格包括一个或多个单元; 确定一个或多个网格单元的事件似然比; 识别一个或多个事件集群,其中每个事件集群包括一个或多个网格单元; 以及在数据存储库中存储具有高于阈值显着性水平的显着性水平的事件簇。

    Event Identification In Sensor Analytics
    5.
    发明申请
    Event Identification In Sensor Analytics 有权
    传感器分析中的事件识别

    公开(公告)号:US20100082301A1

    公开(公告)日:2010-04-01

    申请号:US12241266

    申请日:2008-09-30

    IPC分类号: G06F11/00

    CPC分类号: H04L41/0618 H04L41/12

    摘要: A method of detecting an event anomaly includes receiving one or more data points, in which each data point represents a spatial or temporal event, associating a unique identifier with each of the one or more data points to obtain one or more individualized data points, distributing the one or more individualized data points across a grid, in which the grid includes one or more cells, determining an event likelihood ratio for one or more of the grid cells, identifying one or more event clusters, in which each event cluster includes one or more of the grid cells, and storing in a data repository an event cluster having a significance level above a threshold significance level.

    摘要翻译: 检测事件异常的方法包括接收一个或多个数据点,其中每个数据点表示空间或时间事件,将唯一标识符与所述一个或多个数据点中的每一个相关联以获得一个或多个个性化数据点,分发 网格中的一个或多个个性化数据点,其中网格包括一个或多个小区,确定一个或多个网格单元的事件似然比,识别一个或多个事件簇,其中每个事件簇包括一个或多个 更多的网格单元,以及在数据存储库中存储具有高于阈值显着性水平的显着性水平的事件簇。