Skripsi
PERBANDINGAN ALGORITMA PARTITIONING AROUND MEDOID (PAM) DAN DENSITY BASED SPATIAL CLUSTERING OF APPLICATION WITH NOISE (DBSCAN) DALAM KLASTERISASI KASUS BALITA STUNTING DI INDONESIA
The high prevalence of stunting in various regions in Indonesia has become a major focus of the government. Although various programs have been implemented, their effectiveness is still insufficient in addressing this issue. Therefore, it is necessary to cluster regions in Indonesia based on stunting prevalence rates, so that areas with similar stunting rates can be identified and separated from areas with different stunting rates. This study aims to cluster districts/cities in Indonesia based on the prevalence value of stunting among children under five and compare the performance of clustering methods using the silhouette coefficient. The clustering methods used are PAM and DBSCAN. The data used is secondary data from the publication of the Ministry of Home Affairs of the Republic of Indonesia related to stunting cases. The clustering results showed that PAM produced two clusters, namely cluster 0 and cluster 1 with members of 49 and 62 districts/cities respectively. Meanwhile, DBSCAN also produces two clusters, namely 0 and 1. Cluster 0 consists of 84 districts/cities and cluster 1 consists of 5 districts/cities. The PAM and DBSCAN methods produced silhouette coefficient values of 0.4546 and 0.2867, respectively. Therefore, in this study, the PAM method provided better clustering results.
Inventory Code | Barcode | Call Number | Location | Status |
---|---|---|---|---|
2407004063 | T149940 | T1499402024 | Central Library (REFERENCES) | Available but not for loan - Not for Loan |
No other version available