Application Of The K-Means Algorithm in Clustering Medical Records Of BPJS Participants At Bhayangkara Hospital In Bengkulu

Abstract

Grouping is to separate labels from unknown data and grouping is expected to be able to identify data groups to then be labeled as desired. Cluster analysis is a multivariate analysis technique to find and organize information about variables so that they can be relatively grouped into homogeneous groups or "clusters" can be formed.The purpose of data clustering work can be divided into two, namely grouping for understanding and grouping for use. If the goal is for understanding, the formed groups must capture the natural structure of the data, usually the grouping process in this goal is only an initial process to then be continued with core work such as summarization (average, standard deviation), class labeling in each group to then be used as classification training data and so on. K-Means is one of the clustering algorithms included in the Unsupervised Learning group which is used to divide data into several groups with a partition system. This algorithm accepts input in the form of data without class labels. In the K-Means algorithm, the computer groups the data that is its input without first knowing the target class. The input received is data or objects and k desired groups (clusters).