Create a umap object from the uwot::umap() function.
Usage
hai_umap_list(.data, .kmeans_map_tbl, .k_cluster = 5)
umap_list(.data, .kmeans_map_tbl, .k_cluster = 5)Arguments
- .data
The data from the
hai_kmeans_user_item_tbl()function.- .kmeans_map_tbl
The data from the
hai_kmeans_mapped_tbl().- .k_cluster
Pick the desired amount of clusters from your analysis of the scree plot.
Details
This takes in the user item table/matix that is produced by
hai_kmeans_user_item_tbl() function. This function uses the defaults of
uwot::umap().
See also
https://github.com/jlmelville/uwot (GitHub)
https://github.com/jlmelville/uwot (arXiv paper)
Other UMAP:
hai_umap_plot()
Examples
library(healthyR.data)
library(dplyr)
library(broom)
data_tbl <- healthyR_data %>%
filter(ip_op_flag == "I") %>%
filter(payer_grouping != "Medicare B") %>%
filter(payer_grouping != "?") %>%
select(service_line, payer_grouping) %>%
mutate(record = 1) %>%
as_tibble()
uit_tbl <- hai_kmeans_user_item_tbl(
.data = data_tbl,
.row_input = service_line,
.col_input = payer_grouping,
.record_input = record
)
kmm_tbl <- hai_kmeans_mapped_tbl(uit_tbl)
umap_list(.data = uit_tbl, kmm_tbl, 3)
#> New names:
#> • `` -> `...1`
#> • `` -> `...2`
#> Joining with `by = join_by(service_line)`
#> $umap_obj
#> [,1] [,2]
#> [1,] 0.39184968 -1.8896158
#> [2,] 0.16395013 -2.2863021
#> [3,] -1.33942778 1.4722737
#> [4,] -0.68681033 0.3922719
#> [5,] -0.14532168 1.7592033
#> [6,] -0.82508357 1.8600784
#> [7,] 1.36565988 -1.0231871
#> [8,] 0.86666410 -1.1346344
#> [9,] -1.11591799 0.9673515
#> [10,] 0.31026094 1.1925840
#> [11,] 0.83419626 -1.5810329
#> [12,] -0.39045889 1.3636999
#> [13,] -0.22018607 -1.6445454
#> [14,] 1.13742368 -0.6410403
#> [15,] 0.05296152 0.1861346
#> [16,] 0.79893129 -0.3175221
#> [17,] -0.77310927 1.2788252
#> [18,] 0.01746671 -1.1095703
#> [19,] 0.58316629 -0.6341155
#> [20,] 0.16815202 0.8797567
#> [21,] -0.16529353 0.7925861
#> [22,] 0.11417787 -1.6938942
#> [23,] -1.14325125 1.8106948
#> attr(,"scaled:center")
#> [1] -11.4538671 0.6038069
#>
#> $umap_results_tbl
#> # A tibble: 23 × 3
#> x y service_line
#> <dbl> <dbl> <chr>
#> 1 0.392 -1.89 Alcohol Abuse
#> 2 0.164 -2.29 Bariatric Surgery For Obesity
#> 3 -1.34 1.47 CHF
#> 4 -0.687 0.392 COPD
#> 5 -0.145 1.76 CVA
#> 6 -0.825 1.86 Carotid Endarterectomy
#> 7 1.37 -1.02 Cellulitis
#> 8 0.867 -1.13 Chest Pain
#> 9 -1.12 0.967 GI Hemorrhage
#> 10 0.310 1.19 Joint Replacement
#> # ℹ 13 more rows
#>
#> $kmeans_obj
#> K-means clustering with 3 clusters of sizes 5, 6, 12
#>
#> Cluster means:
#> Blue Cross Commercial Compensation Exchange Plans HMO Medicaid
#> 1 0.1495475 0.03679700 0.0003066332 0.020729565 0.16252855 0.13072521
#> 2 0.1170278 0.03141187 0.0101665392 0.013865190 0.09822472 0.08557952
#> 3 0.0783745 0.02182129 0.0043244347 0.006202137 0.04493860 0.03684344
#> Medicaid HMO Medicare A Medicare HMO No Fault Self Pay
#> 1 0.31446157 0.1318675 0.03192357 0.001364577 0.019748398
#> 2 0.14652195 0.3535395 0.10524131 0.007067791 0.031353724
#> 3 0.08001653 0.5625037 0.15152338 0.003475542 0.009976485
#>
#> Clustering vector:
#> [1] 1 1 3 3 3 3 2 2 3 3 1 3 1 2 3 2 3 2 2 3 3 1 3
#>
#> Within cluster sum of squares by cluster:
#> [1] 0.19152559 0.08456928 0.09625399
#> (between_SS / total_SS = 73.6 %)
#>
#> Available components:
#>
#> [1] "cluster" "centers" "totss" "withinss" "tot.withinss"
#> [6] "betweenss" "size" "iter" "ifault"
#>
#> $kmeans_cluster_tbl
#> # A tibble: 23 × 2
#> service_line .cluster
#> <chr> <fct>
#> 1 Alcohol Abuse 1
#> 2 Bariatric Surgery For Obesity 1
#> 3 CHF 3
#> 4 COPD 3
#> 5 CVA 3
#> 6 Carotid Endarterectomy 3
#> 7 Cellulitis 2
#> 8 Chest Pain 2
#> 9 GI Hemorrhage 3
#> 10 Joint Replacement 3
#> # ℹ 13 more rows
#>
#> $umap_kmeans_cluster_results_tbl
#> # A tibble: 23 × 4
#> x y service_line .cluster
#> <dbl> <dbl> <chr> <fct>
#> 1 0.392 -1.89 Alcohol Abuse 1
#> 2 0.164 -2.29 Bariatric Surgery For Obesity 1
#> 3 -1.34 1.47 CHF 3
#> 4 -0.687 0.392 COPD 3
#> 5 -0.145 1.76 CVA 3
#> 6 -0.825 1.86 Carotid Endarterectomy 3
#> 7 1.37 -1.02 Cellulitis 2
#> 8 0.867 -1.13 Chest Pain 2
#> 9 -1.12 0.967 GI Hemorrhage 3
#> 10 0.310 1.19 Joint Replacement 3
#> # ℹ 13 more rows
#>