Interactive data visualization for trend analysis

Patent metadata

Field	Value
Publication number	US-8959003-B2
Application number	US-201113177721-A
Country	US
Kind code	B2
Filing date	Jul 7, 2011
Priority date	Jul 7, 2011
Publication date	Feb 17, 2015
Grant date	Feb 17, 2015

Abstract

A method and system for identifying a trend in data of heat maps. An initial set of comparison heat maps included in the repository is determined and displayed based on how closely initial comparison area(s) of each comparison heat map match initial target area(s) of a first heat map. A review of the initial set of comparison heat maps determines additional comparison area(s) corresponding to additional target area(s) of the first heat map. In an iterative process, an additional set of comparison heat maps is determined and displayed based on how closely initial and additional comparison areas match respective initial and additional target areas of the first heat map. A trend relating initial and additional target areas is identified based on the displayed additional set of comparison heat maps and based on how closely comparison areas match their respective target areas.

First claim

What is claimed is:

1. A method of identifying a trend in data of heat maps, said method comprising:
a computer displaying a first heat map in a user interface and receiving user-configured shape(s) whose border(s) positioned over said displayed first heat map in said user interface specify respective portion(s) of said displayed first heat map, said portion(s) being initial target area(s) included in said first heat map, said initial target area(s) having respective color composition(s), and said initial target area(s) being less than an entirety of said first heat map;
said computer determining initial values indicating respective first measures of similarity in color composition between (1) color composition(s) of initial comparison area(s) included in each comparison heat map included in a plurality of comparison heat maps stored in a repository and (2) said color composition(s) of said initial target area(s), said initial comparison area(s) being respective portion(s) of each comparison heat map and being in position(s) in each comparison heat map which are identical to position(s) of said initial target area(s) in said first heat map;
said computer determining that at least some of said first measures of similarity in color composition exceed a first threshold value which indicates said color composition(s) of said initial target area(s) are similar to said color composition(s) of said initial comparison area(s) included in at least some of said comparison heat maps;
based on said first measures of similarity in color composition exceeding said first threshold value, said computer determining an initial set of comparison heat maps included in said plurality of comparison heat maps;
said computer determining an initial ranking of comparison heat maps included in said initial set of comparison heat maps based on said first measures of similarity in color composition;
based on said initial ranking, said computer initiating a display of said initial set of comparison heat maps and indications of said initial values;
said computer receiving user-configured shape(s) whose border(s) specify respective additional portion(s) of said displayed first heat map, said additional portion(s) being additional target area(s) included in said first heat map, said additional target area(s) having respective color composition(s), and said additional target area(s) being less than said entirety of said first heat map, wherein said additional target area(s) have position(s) in said first heat map that are identical to position(s) of additional comparison area(s) in each comparison heat map included in said initial set of comparison heat maps, and said additional comparison area(s) being respective additional portion(s) of each comparison heat map;
said computer determining additional values indicating respective second measures of similarity in color composition between said color composition(s) of said initial comparison area(s) and said color composition(s) of said initial target area(s) and between color composition(s) of said additional comparison area(s) and color composition(s) of said additional target area(s);
said computer determining that at least some of said second measures of similarity in color composition exceed a second threshold value which indicates said color composition(s) of said initial target area(s) are similar to said color composition(s) of said initial comparison area(s) included in said at least some of said comparison heat maps and further indicates said color composition(s) of said additional target area(s) are similar to said color composition(s) of said additional comparison area(s) included in some of said comparison heat maps;
based on said second measures of similarity in color composition exceeding said second threshold value, said computer determining an additional set of comparison heat maps included in said plurality of comparison heat maps;
said computer determining an additional ranking of comparison heat maps included in said additional set of comparison heat maps based on said second measures of similarity in color composition;
based on said additional ranking, said computer initiating an additional display of said additional set of comparison heat maps and indications of said additional values;
said computer determining one or more colors in said initial target area(s) indicate a first anomaly in data included in said first heat map, said first anomaly being a first problem associated with a performance of a first hardware portion of a computer network;
said computer determining one or more colors in said additional comparison area(s) and in said additional target area(s) indicate a second anomaly in data included in each comparison heat map of said additional set of comparison heat maps and in said first heat map, said second anomaly being a second problem associated with a performance of a second hardware portion of said computer network;
attempting to identify a trend relating said initial and said additional target areas based on said additional display of said additional set of comparison heat maps and said indications of said additional values;
if a result of said attempting is said trend is not identified, repeating said receiving said user-configured shape(s) whose border(s) specify respective additional portion(s) of said displayed first heat map, said determining additional values, said determining that at least some of said second measures of similarity in color composition exceed said second threshold value, said determining an additional set of comparison heat maps, said determining an additional ranking, said initiating an additional display of said additional set of comparison heat maps and said indications of said additional values, and said attempting to identify said trend until said result of said attempting is said trend is identified;
subsequent to said attempting and based on said additional display of said additional set of comparison heat maps and said indications of said additional values, identifying said trend relating said initial and said additional target areas, wherein said identifying said trend includes determining a relationship between said first anomaly and said second anomaly; and
identifying a cause of said first problem based on said identified trend.

2. The method of claim 1 , wherein said determining said initial values indicating said respective first measures of similarity in color composition between said color composition(s) of said initial comparison area(s) and said color composition(s) of said initial target area(s) includes determining an initial percentage of colors of said initial comparison area(s) included in a first comparison heat map included in said plurality of comparison heat maps that match colors of said initial target area(s) included in said first heat map, and
wherein said determining said additional values indicating said respective second measures of similarity between said color composition(s) of said initial comparison area(s) and said color composition(s) of said initial target area(s) and between said color composition(s) of said additional comparison area(s) and color composition(s) of said additional target area(s) includes:
said determining said initial percentage of colors of said initial comparison area(s) that match said colors of said initial target area(s); and
determining an additional percentage of colors of said additional comparison area(s) included in said first comparison heat map included in said plurality of comparison heat maps that match colors of said additional target area(s) included in said first heat map.

3. The method of claim 2 , further comprising:
converting data associated with said initial target area(s) to said colors of said initial target area(s);
converting data associated with said additional target area(s) to said co

Assignees

• Brugler Thomas S
• Craig Randall A
• Conzola Vincent C
• Eischeid Todd M
• Molander Mark E
• IBM

Inventors

• Brugler Thomas S
• Craig Randall A
• Conzola Vincent C
• Eischeid Todd M
• Molander Mark E

Classifications

• G06F11/3452

  Performance evaluation by statistical analysis · CPC title

• G06F11/3082 Primary

  the data filtering being achieved by aggregating or compressing the monitored data · CPC title