Visualization of relationships and strengths between data nodes
View Patent ↗One or more processors receive a dataset that includes a plurality of nodes. One or more processors identify relationships between a plurality of interacting nodes within the dataset. One or more processors determine relationship strength values between a plurality of interacting node pairs within the dataset. One or more processors generate a graphical representation that represents the relationship strength values between the plurality of interacting nodes within the dataset. Interacting node pairs are connected by edges and the edges have a length that correlates with the relationship strength value between the interacting node pairs.
1. A method for visualizing relationships and relationship strengths of data comprising:
receiving, by one or more processors, a dataset that includes a plurality of nodes;
identifying, by one or more processors, relationships between a plurality of interacting nodes within the dataset;
determining, by one or more processors, relationship strength values between a plurality of interacting node pairs within the dataset;
generating, by one or more processors, a graphical representation that represents the relationship strength values between the plurality of interacting nodes within the dataset, wherein interacting node pairs are connected by edges, wherein the edges have a length that correlates with the relationship strength value between the interacting node pairs, and wherein a calculation to determine the length of the edges includes an equation:
l
(
e
i
)
=
L
·
1.0
0.5
+
ⅇ
-
(
5
·
(
s
i
-
s
avg
)
)
where l(e i ) is the length of the edges corresponding to a record i; L is a scaling factor specific to a visualization and chosen by a user; s i is a numeric strength-of-relationship for the record i; and s avg is an average numeric strength-of-relationship;
counting, by one or more processors, a number of interactions between interacting node pairs;
increasing, by one or more processors, the relationship strength value based, at least in part, on each interaction counted;
generating, by one or more processors, an animation showing a change in relationship strength over time using a plurality of graphical representations of graphs that represent a specified time period, wherein one or more distances between two or more nodes visually changes over time in the animation;
predicting, by one or more processors, a future relationship strength value for at least one of the interacting node pairs;
providing, by one or more processors, a degree of confidence value for the future relationship strength value based, at least in part, on a statistical analysis of data included in the dataset;
generating, by one or more processors, a graphical representation that includes a visualization of one or both of trends and patterns in a change of relationship strength between interacting nodes; and
generating, by one or more processors, the graphical representations using a modified force-directed algorithm or a separate layout algorithm.