Computer-implemented system and method for cluster spine group arrangement
A computer-implemented system and method for cluster spine group arrangement is provided. A set of spine groups each having one or more spines of clusters and at least one singleton cluster is obtained. Unique spine groups are identified within the set and placed in a center of a display. At least a portion of the remaining spine groups in the set are placed to extend radially from the unique spine groups in the display center.
1. A computer-implemented system for cluster spine group arrangement, comprising:
a set of spine groups each comprising one or more spines of clusters and at least one singleton cluster;
a display; and
a server comprising a central processing unit, memory, an input port to receive the spine groups, and an output port, wherein the central processing unit is configured to:
identify unique spine groups within the set;
place the unique spine groups in a center of a display; and
arrange at least a portion of the remaining spine groups in the set extending radially from the unique spine groups in the display center.
2. A system according to claim 1 , wherein the central processing unit places at least one of the remaining spine groups adjacent to one of the unique spine groups.
3. A system according to claim 1 , wherein the central processing unit moves one or more of the unique spine groups to prevent overlap with one or more of the placed remaining spine groups.
4. A system according to claim 1 , wherein the central processing unit determines the placement of each of the remaining spine groups to prevent overlap with one or more of the unique spine groups.
5. A system according to claim 1 , wherein the central processing unit generates one such spine group by arranging two or more clusters of documents along a spine based on concepts of the clusters and grafting at least one other spine of clusters to one of the clusters arranged along the spine.
6. A system according to claim 5 , wherein the central processing unit selects the two or more clusters for placement along the spine based on a best fit comprising a minimum similarity.
7. A system according to claim 5 , wherein the cluster on which the other spine of clusters is grafted is most similar to the other spine of clusters.
8. A system according to claim 5 , wherein the central processing unit places a different spine of clusters having less similarity to the clusters arranged along the spine adjacent one of the clusters arranged along with spine that is most similar to the different spine of clusters.
9. A system according to claim 1 , wherein the central processing unit identifies the unique spine groups by determining a group concept score vector for each spine group based on concepts of the clusters in that group and identifying as the unique spine groups those spine groups that are dissimilar to the other spine groups based on the group concept scores.
10. A system according to claim 1 , wherein the central processing unit places one or more labels adjacent to one or more of the remaining spine groups placed in the display and extending away from the display center.
11. A computer-implemented method for cluster spine group arrangement, comprising:
obtaining a set of spine groups each comprising one or more spines of clusters and at least one singleton cluster;
identifying unique spine groups within the set;
placing the unique spine groups in a center of a display; and
placing at least a portion of the remaining spine groups in the set extending radially from the unique spine groups in the display center.
12. A method according to claim 11 , further comprising:
placing at least one of the remaining spine groups adjacent to one of the unique spine groups.
13. A method according to claim 11 , further comprising:
moving one or more unique spine groups to prevent overlap with one or more of the placed remaining spine groups.
14. A method according to claim 11 , further comprising:
determining the placement of each of the remaining spine groups to prevent overlap with one or more of the unique spine groups.
15. A method according to claim 11 , further comprising:
generating one such spine group, comprising:
arranging two or more clusters of documents along a spine based on concepts of the clusters; and
grafting at least one other spine of clusters to one of the clusters arranged along the spine.
16. A method according to claim 15 , further comprising:
selecting the two or more clusters for placement along the spine based on a best fit comprising a minimum similarity.
17. A method according to claim 15 , wherein the cluster on which the other spine of clusters is grafted is most similar to the other spine of clusters.
18. A method according to claim 15 , further comprising:
placing a different spine of clusters having less similarity to the clusters arranged along the spine adjacent one of the clusters arranged along with spine that is most similar to the different spine of clusters.
19. A method according to claim 11 , further comprising:
identifying the unique spine groups, comprising:
determining a group concept score vector for each spine group based on concepts of the clusters in that group; and
identifying as the unique spine groups those spine groups that are dissimilar to the other spine groups based on the group concept scores.
20. A method according to claim 11 , further comprising:
placing one or more labels adjacent to one or more of the placed remaining spine groups and extending away from the display center.