Graph similarity analytics

A computer-implemented method, comprising:

• generating, by a system operatively coupled to a processor, a first information index indicative of a first entropy measure for a first graph-structured dataset that represents a first dynamic network that changes over time;
• generating, by the system, a second information index indicative of a second entropy measure for a second graph-structured dataset that represents a second dynamic network that changes over the time;
• receiving, by the system, first change data that represents a first set of sequential changes to the first graph-structured dataset over the time;
• updating, by the system, the first information index based on the first change data, resulting in an updated first information index;
• receiving, by the system, second change data that represents a second set of sequential changes to the second graph-structured dataset over the time;
• updating, by the system, the second information index based on the second change data, resulting in an updated second information index;
• determining, by the system, based on a similarity computation between the updated first information index and the updated second information index, a measure of similarity between the first graph-structured dataset and the second graph-structured dataset;
• employing, by the system, an artificial intelligence system to select and perform one of a plurality of task-dependent analyses for the first set of sequential changes and the second set of sequential changes based on the measure of similarity;
• and
• generating, by the system on a display device, visual data comprising dynamic visual elements indicative of visual statistics associated with the measure of similarity, wherein visual characteristics of the dynamic visual elements are displayed in an altered format based on the measure of similarity, and wherein the visual characteristics are selected from a group consisting of:
• content, visual property, size, hue and shading.

A computer program product comprising a non-transitory computer readable medium having program instructions embodied therewith, the program instructions executable by a processor to cause the processor to:

• generate, by the processor, a first graph data structure for a first graph-structured dataset that represents a first dynamic network that changes over time;
• generate, by the processor, a second graph data structure for a second graph-structured dataset that represents a second dynamic network that changes over the time;
• receive, by the processor, first change data that represents a first set of sequential changes to the first graph-structured dataset over the time;
• update, by the processor, a first information index of the first graph data structure based on the first change data, resulting in an updated first information index;
• receive, by the processor, second change data that represents a second set of sequential changes to the second graph-structured dataset over the time;
• update, by the processor, a second information index of the second graph data structure based on the second change data, resulting in an updated second information index;
• determine, by the processor, a measure of similarity between the first graph-structured dataset and the second graph-structured dataset based on a similarity computation between the updated first information index and the updated second information index;
• employ, by the processor, an artificial intelligence system to select and perform one of a plurality of task-dependent analyses for the first set of sequential changes and the second set of sequential changes based on the measure of similarity;
• and
• generate on a display device, visual data comprising dynamic visual elements indicative of visual statistics associated with the measure of similarity, wherein visual characteristics of the dynamic visual elements are displayed in an altered format based on the measure of similarity, and wherein the visual characteristics are selected from a group consisting of:
• content, visual property, size, hue and shading.