Automatic hot area detection in heat map visualizations
The present disclosure involves systems, software, and computer implemented methods for automatically detecting hot areas in heat map visualizations. One example method includes identifying a two-dimensional heat map. The identified two-dimensional heat map is converted to a one-dimensional heat map. Cells of the one-dimensional heat map are clustered using a density-based clustering algorithm to generate at least one dense region of cells. A mean value of cells in each dense region is calculated and the dense regions are sorted by mean value in descending order. An approach for identifying hot areas is selected and the selected approach is used to identify at least one dense region as a hot area of the one-dimensional heat map.
1. A computer-implemented method comprising:
identifying a two-dimensional heat map;
converting the identified two-dimensional heat map to a single one-dimensional heat map;
clustering cells of the single one-dimensional heat map using a density-based clustering algorithm to generate a plurality of dense regions of cells, wherein the clustering includes:
identifying clustering algorithm parameters for a current iteration of clustering, wherein the clustering algorithm parameters include:
an epsilon parameter that specifies a closeness value of cells for inclusion in a same cluster; and
a minimum-samples parameter that defines a minimum number of cells for forming a dense region; and
performing the clustering according to the identified clustering algorithm parameters to generate the plurality of dense regions in the single one-dimensional heat map:
calculating a mean value of cells in each dense region of the single one-dimensional heat map:
sorting the plurality of dense regions according to the mean value in descending order;
selecting an approach for identifying hot areas; and
using the selected approach to identify a single hot area that includes multiple dense regions of the plurality of dense regions in the single one-dimensional heat map, wherein using the selected approach comprises:
identifying a dense-region count parameter that specifies how many dense regions of the plurality of dense regions to include in a given single hot area;
selecting, according to the selected approach, a number of dense regions equal to the dense-region count parameter; and
merging a plurality of dense regions of the selected number to form the single hot area.
2. The method of claim 1 , wherein the density-based clustering algorithm is a DBSCAN (Density-Based Spatial Clustering of Applications with Noise) algorithm.
3. The method of claim 1 , wherein the approach is identifying positive peaks.
4. The method of claim 3 , wherein using the selected approach comprises:
selecting one or more regions with highest mean values; and
merging the selected one or more regions to form the hot area.
5. The method of claim 1 , wherein the approach is identifying negative peaks.
6. The method of claim 5 , wherein using the selected approach
comprises: selecting one or more regions with lowest mean values; and
merging the selected one or more regions to form the hot area.
7. The method of claim 1 , wherein the approach is identifying positive and negative peaks.
8. The method of claim 5 , wherein using the selected approach comprises:
selecting one or more lowest-mean regions having lowest mean values;
merging the selected one or more lowest-mean regions to form a first hot area;
selecting one or more highest-mean regions having highest mean values; and
merging the selected one or more highest-mean regions to form a second hot area.
9. The method of claim 1 , further comprising, before sorting the dense regions, replacing the cell values for all cells in each dense region with the mean value for the dens e region.
10. A system comprising:
one or more computers; and
a non-transitory computer-readable medium coupled to the one or more computers having instructions stored thereon which, when executed by the one or more computers, caus e the one or more computers to perform operations comprising:
identifying a two-dimensional heat map;
converting the identified two-dimensional heat map to a single one-dimensional heat map;
clustering cells of the single one-dimensional heat map using a density-based clustering algorithm to generate a plurality of dense regions of cells, wherein the clustering includes:
identifying clustering algorithm parameters for a current iteration of clustering, wherein the clustering algorithm parameters include:
an epsilon parameter that specifies a closeness value of cells for inclusion in a same cluster; and
a minimum-samples parameter that defines a minimum number of cells for forming a dense region; and
performing the clustering according to the identified clustering algorithm parameters to generate the plurality of dense regions in the single one-dimensional heat calculating a mean value of cells in each dense region of the single one-dimensional heat map;
sorting the plurality of dense regions according to the mean value in descending order;
selecting an approach for identifying hot areas; and
using the selected approach to identify a single hot area that includes multiple dense regions of the plurality of dense regions in the single one-dimensional heat map, wherein using the selected approach comprises:
identifying a dense-region count parameter that specifies how many dense regions of the plurality of dense regions to include in a given single hot area;
selecting, according to the selected approach, a number of dense regions equal to the dense-region count parameter; and
merging a plurality of dense regions of the selected number to form the single hot area.
11. The system of claim 10 , wherein the density-based clustering algorithm is a DBSCAN (Density-Based Spatial Clustering of Applications with Noise) algorithm.
12. The system of claim 10 , wherein the approach is identifying positive peaks.
13. The system of claim 12 , wherein using the selected approach comprises:
selecting one or more regions with highest mean values; and
merging the selected one or more regions to form the hot area.
14. The system of claim 10 , wherein the approach is identifying negative peaks.
15. The system of claim 14 , wherein using the selected approach comprises:
selecting one or more regions with lowest mean values; and
merging the selected one or more regions to form the hot area.
16. A computer program product encoded on a non-transitory storage medium, the product comprising non-transitory, computer readable instructions for causing one or more processors to perform operations comprising:
identifying a two-dimensional heat map;
converting the identified two-dimensional heat map to a single one-dimensional heat map;
clustering cells of the single one-dimensional heat map using a density-based clustering algorithm to generate a plurality of dense regions of cells, wherein the clustering includes:
identifying clustering algorithm parameters for a current iteration of clustering, wherein the clustering algorithm parameters include:
an epsilon parameter that specifies a closeness value of cells for inclusion in a same cluster; and
a minimum-samples parameter that defines a minimum number of cells for forming a dense region; and
performing the clustering according to the identified clustering algorithm parameters to generate the plurality of dense regions in the single one-dimensional heat
calculating a mean value of cells in each dense region of the single one-dimensional heat map;
sorting the plurality of dense regions according to the mean value in descending order;
selecting an approach for identifying hot areas; and
using the selected approach to identify a single hot area that includes multiple dense regions of the plurality of dense regions in the single one-dimensional heat map, wherein using the selected approach comprises:
identifying a dense-region count parameter that specifies how many dense regions of the plurality of dense regions to include in a given single hot area;
selecting, according to the selected approach, a number of dense regions equal to the dense-region count parameter; and
merging a plurality of dense regions of the selected number to form the single hot area.
17. The computer program product of claim 16 , wherein the density-based clustering algorithm is a DBSCAN (Density-Based Spatial Clustering of Applications with Noise) algorithm.
18. The computer program product of claim 16 , wherein the approach is identifying positive peaks.
19. The computer program product of claim 18 , wherein using the selected approach comprises:
selecting one or more regions with highest mean values; and
merging the selected one or more regions to form the hot area.
20. The computer program product of claim 16 , wherein the approach is identifying negative peaks.