Penerapan Data Mining Pengelompokan Data Pengguna Air Bersih Berdasarkan Keluhannya Menggunakan Metode Clustering Pada PDAM Langkat

Abstract

Customer problems are indeed very complex, therefore they must be handled properly, clearly, and thoroughly. Good service from a company can show the professionalism of the company itself, meaning that seriousness, certainty of time, punctuality and work results that can be accounted for in solving all problems can prove the quality of a company. Clustering is the process of partitioning a set of data objects into subsets called clusters. Objects in the cluster have similar characteristics to each other and are different from other clusters. Clustering is very useful and can find unknown groups or groups in the data. From 2056 customer complaint data, the results obtained are Cluster 1, namely 12, 5, 5, in cluster 2, namely 4, 5, 5 and cluster 3, namely 8, 2, 2. With the number of cluster members 1 883 members, cluster 2 635 members and cluster 3 namely 538 members. From the results of the Matlab cluster, there are similar results, namely the types of complaints in cluster 1 and cluster 2, namely code 5 types of leaking pipe complaints with handling damage to connecting water pipes (gibout join).

Keywords : Clustering, Custome Problems, Matlab