发明名称 Visualizing a relationship of attributes using a relevance determination process to select from candidate attribute values
摘要 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.
申请公布号 US9280612(B2) 申请公布日期 2016.03.08
申请号 US201213714871 申请日期 2012.12.14
申请人 Hewlett Packard Enterprise Development LP 发明人 Hao Ming C.;Mittelstaedt Sebastian;Dayal Umeshwar;Hsu Meichun;Keim Daniel
分类号 G06F17/30 主分类号 G06F17/30
代理机构 Hewlett Packard Enterprise Development LP 代理人 Hewlett Packard Enterprise Development LP
主权项 1. A computer-implemented method of visualizing a relationship of attributes in data records, comprising: for plural categorical values of a first attribute, selecting, by a system having a processor, more relevant categorical values of a second attribute from among candidate categorical values of the second attribute, wherein 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; and generating, by the system, a visualization that includes groups of cells corresponding to the respective pairs of categorical values of the first and second attributes, each of the cells corresponding to a pair of categorical values of the first and second attributes in a respective data record, wherein at least one of the groups includes cells assigned visual indicators based on values of a third attribute, wherein generating the visualization comprises: placing cells sharing a given pair of a first categorical value of the first attribute and a second categorical value of the second attribute in a region corresponding to an intersection of the first categorical value and the second categorical value in the visualization, wherein the placing avoids overlaying of the placed cells.
地址 Houston TX US