公开(公告)号：US20140325417A1

公开(公告)日：2014-10-30

申请号：US13871924

申请日：2013-04-26

Applicant: HEWLETT-PACKARD DEVELOPMENT COMPANY, L.P.

Inventor： Ming C. Hao ,  Manish Marwah ,  Michael Hund ,  Sebastian Mittelstaedt ,  Umeshwar Dayal ,  Meichun Hsu ,  Cullen E. Bash ,  Daniel Keim

IPC: G06F3/0484

CPC classification number: G06T11/206 ,  G06F3/048 ,  G06F3/0484

Abstract: Visual analytics using multivariate concentric rings with a visual start time mechanism includes displaying an interactive graph where the interactive graph has multiple concentric rings that have multiple cells that represent sequential time periods. The concentric rings form a time unit that starts at an origin and ends at a time unit end and also has a pre-nonorigin starting section and a post-nonorigin starting section. A color is displayed in the cells to represent measurements associated with time stamps corresponding to cells in the post-nonorigin starting section. Further, a background color is displayed in cells of the pre-nonorigin starting section. The cells in the pre-nonorigin starting section are reused by displaying a color to represent metrics associated with time stamps belonging to a subsequent time unit.

Abstract translation: 使用具有可视启动时间机制的多变量同心环的视觉分析包括显示交互图，其中交互图具有多个同心环，其具有表示连续时间段的多个单元。 同心环形成时间单位，起始于原点并在时间单位末端结束，并且还具有前非诺里金起始部分和后非原始启动部分。 在单元格中显示颜色以表示与后非诺里金起始部分中的细胞相对应的时间标记相关联的测量。 此外，在前非原始码开始部分的单元格中显示背景颜色。 通过显示颜色以代表与属于后续时间单位的时间标记相关联的度量来重新使用前非诺里金开始部分中的单元。

公开(公告)号：US20140172861A1

公开(公告)日：2014-06-19

申请号：US13714871

申请日：2012-12-14

Applicant: HEWLETT-PACKARD DEVELOPMENT COMPANY, L.P.

Inventor： Ming C. Hao ,  Sebastian Mittelstaedt ,  Umeshwar Dayal ,  Meichun Hsu ,  Daniel Keim

IPC: G06F17/30

CPC classification number: G06F17/30994 ,  G06F17/30598

Abstract: For plural categorical values of a first attribute, more relevant categorical values of a second attribute are selected from among candidate categorical values of the second attribute, where the selecting uses a relevance determination process that considers frequencies of occurrence of respective pairs of the categorical values of the first and second attributes. A visualization that includes groups of cells corresponding to the respective pairs of categorical values of the first and second attributes is generated. At least one of the groups includes cells assigned visual indicators based on values of a third attribute.

Abstract translation: 对于第一属性的多个分类值，从第二属性的候选分类值中选择第二属性的更相关的分类值，其中选择使用考虑各个对的分类值对的出现频率的相关性确定过程 第一和第二属性。 生成包括与第一和第二属性的相应分类值对应的单元组的可视化。 至少一个组包括基于第三属性的值分配视觉指示符的小区。

公开(公告)号：US09280612B2

公开(公告)日：2016-03-08

申请号：US13714871

申请日：2012-12-14

Applicant: HEWLETT-PACKARD DEVELOPMENT COMPANY, L.P.

Inventor： Ming C. Hao ,  Sebastian Mittelstaedt ,  Umeshwar Dayal ,  Meichun Hsu ,  Daniel Keim

IPC: G06F17/30

CPC classification number: G06F17/30994 ,  G06F17/30598

Abstract: For plural categorical values of a first attribute, more relevant categorical values of a second attribute are selected from among candidate categorical values of the second attribute, where the selecting uses a relevance determination process that considers frequencies of occurrence of respective pairs of the categorical values of the first and second attributes. A visualization that includes groups of cells corresponding to the respective pairs of categorical values of the first and second attributes is generated. At least one of the groups includes cells assigned visual indicators based on values of a third attribute.

Abstract translation: 对于第一属性的多个分类值，从第二属性的候选分类值中选择第二属性的更相关的分类值，其中选择使用考虑各个对的分类值对的出现频率的相关性确定过程 第一和第二属性。 生成包括与第一和第二属性的相应分类值对应的单元组的可视化。 至少一个组包括基于第三属性的值分配视觉指示符的小区。

公开(公告)号：US20140365149A1

公开(公告)日：2014-12-11

申请号：US13912090

申请日：2013-06-06

Applicant: Hewlett-Packard Development Company, L.P.

Inventor： Ming C Hao ,  Halldor Janetzko ,  Manish Marwah ,  Umeshwar Dayal ,  Meichun Hsu ,  Daniel Keim

IPC: G01R13/02

CPC classification number: G01R13/02 ,  G01D4/004 ,  G06Q50/06 ,  Y02B90/242 ,  Y02B90/245 ,  Y02B90/246 ,  Y04S20/322 ,  Y04S20/40 ,  Y04S20/42 ,  Y04S20/46

Abstract: Example embodiments relate to providing visual analytics of spatial time series data. In example embodiments, sensors may be located at regions within a building for collecting sensor data at regular time intervals. A sensor hierarchy can be generated including sensor nodes that are hierarchically arranged according to a physical infrastructure of the building, where each of the sensor nodes corresponds to a sensor. Sensor data can be obtained from the sensors, and a pixel calendar tree can be generated based on the sensor data and the sensor hierarchy, where the pixel calendar tree is recursively subdivided into tree portions according to a proportion of the sensor data attributable to each of the sensors. The pixel calendar tree can be displayed, where each of the tree portions includes time series sensor data of a corresponding region that is generated based on the sensor data.

Abstract translation: 示例性实施例涉及提供空间时间序列数据的视觉分析。 在示例实施例中，传感器可以位于建筑物内的区域，以便以规则的时间间隔收集传感器数据。 可以生成传感器层级，其包括根据建筑物的物理基础设施分层布置的传感器节点，其中每个传感器节点对应于传感器。 可以从传感器获得传感器数据，并且可以基于传感器数据和传感器层级来生成像素日历树，其中像素日历树根据归因于每个传感器数据的传感器数据的比例被递归地细分为树部分 传感器。 可以显示像素日历树，其中每个树部分包括基于传感器数据生成的相应区域的时间序列传感器数据。

公开(公告)号：US20140325329A1

公开(公告)日：2014-10-30

申请号：US13869675

申请日：2013-04-24

Applicant: HEWLETT-PACKARD DEVELOPMENT COMPANY, L.P.

Inventor： Ming C. Hao ,  Halldor Janetzko ,  Manish Marwah ,  Umeshwar Dayal ,  Meichun Hsu ,  Daniel Keim

IPC: G06F17/24

CPC classification number: G06F17/246

Abstract: Visual analytics of multivariate data using a cell based calendar matrix having a visual folding mechanism can include forming a time based layout that is divided into cells where the cells represent measurement intervals and a color of the cells represents a measurement value, folding the time based layout into a cell based calendar matrix with other time based layouts that include other cells that represent corresponding measurement intervals in different calendar units of the cell based calendar matrix, and displaying the cell based calendar matrix in a display such that the cells of the time based layout align by time with the other cells of the other time based layouts.

Abstract translation: 使用具有视觉折叠机构的基于单元格的日历矩阵的多变量数据的视觉分析可以包括形成基于时间的布局，其被划分为单元格，其中单元表示测量间隔，并且单元格的颜色表示测量值，折叠基于时间的布局 进入具有其他基于时间的布局的基于单元的日历矩阵，其包括表示基于单元的日历矩阵的不同日历单元中的对应测量间隔的其他单元，以及在显示器中显示基于单元的日历矩阵，使得基于时间的布局 时间与其他基于时间的布局的其他单元格对齐。

公开(公告)号：US20140204090A1

公开(公告)日：2014-07-24

申请号：US13745985

申请日：2013-01-21

Applicant: HEWLETT-PACKARD DEVELOPMENT COMPANY, L.P.

Inventor： Ming C. Hao ,  Sebastian Mittelstaedt ,  Umeshwar Dayal ,  Meichun Hsu ,  Daniel Keim

IPC: G06T11/20

CPC classification number: G06T11/206

Abstract: Using a contingency calculation based on a number of events sharing a collection of values of plural attributes, a discriminative metric is computed representing a statistical significance of the events that share the collection of values of the plural attributes. A visualization is generated that includes cells representing respective events, the visualization including a region containing a subset of the cells corresponding to the collection of values of the plural attributes, and the visualization including a significance visual indicator associated with the region to indicate the statistical significance of the events sharing the collection of values of the plural attributes.

Abstract translation: 使用基于共享多个属性的值的集合的事件的数量的事件计算，计算表示共享多个属性的值的集合的事件的统计重要性的歧视度量。 生成包括表示相应事件的单元的可视化，所述可视化包括包含对应于所述多个属性的值的集合的单元的子集的区域，并且所述可视化包括与所述区域相关联的显着性视觉指示符，以指示统计学意义 的事件共享多个属性的值的集合。

公开(公告)号：US09465678B2

公开(公告)日：2016-10-11

申请号：US14418307

申请日：2012-10-25

Applicant: HEWLETT-PACKARD DEVELOPMENT COMPANY, L.P.

Inventor： Chetan Kumar Gupta ,  Craig Peter Sayers ,  Umeshwar Dayal ,  Krishnamurthy Viswanathan

IPC: G06F9/54 ,  G06F11/00 ,  G06F9/44

CPC classification number: G06F9/542 ,  G06F9/448 ,  G06F11/008 ,  G06F2201/86

Abstract: A method for event correlation includes capturing events and arranging the events sequentially in at least one dimension. An event correlator implemented by a computational device convolves a kernel density function with each of the events to produce a convolved function for each event. Co-occurrences between events are found by calculating overlap between convolved functions.

Abstract translation: 用于事件相关的方法包括捕获事件并且在至少一个维度中顺序排列事件。 由计算设备实现的事件相关器将核心密度函数与每个事件相结合，以产生每个事件的卷积函数。 通过计算卷积函数之间的重叠来发现事件之间的共同事件。