darnila eva, Maryana, Martunis

Infant Routine Monitoring System During 0-12 Months Immunization Using Agglomerative Hierarchical Clustering Algorithm

Introduction

Infant routine monitoring system during 0-12 months immunization using agglomerative hierarchical clustering algorithm. Group infant immunization data (0-12 months) in Bireuen Regency using Agglomerative Hierarchical Clustering. Analyzes gender, address, and immunization types to monitor awareness.

Abstract

Data mining is the process of finding patterns from large data sets using description, estimation, prediction, classification, clustering and association techniques. One of the mininng data techniques used to group is the Agglomerative Hierarchical Clustering algorithm. The process of grouping data using Agglomerative Hierarchical Clustering aims to group objects based on the distance between the two clusters. There are several methods on Agglomerative Hierarchical Clustering algorithm namely Single Linkage, Average Linkage and Complete Linkage. In this study, the authors used the Average Linkage method. By using average linkage the author is interested to conduct research on infant immunization data in Kotajuang Health Center and Gandapura Health Center in Bireuen Regency. Immunization is the process to make a person immune or immune to a disease. This process is carried out by administering a vaccine that stimulates the immune system to be immune to the disease. The purpose of this study was to look at the grouping of infant immunizations based on 3 data variables namely gender, address and type of immunization. From the results of the study, the results were obtained for one of the groups consisting of 3 villages, namely Pante Sikumbang Village, Palohme Village and Ie Rhop Village. It has a male infant immunization rate of 54% and a female infant immunization rate of 77%. The information shown is that awareness of immunizations for baby boys and baby girls in this group "has started well".

Review

This paper presents a timely and relevant application of data mining techniques, specifically Agglomerative Hierarchical Clustering (AHC) with the Average Linkage method, to analyze infant immunization data. The focus on identifying patterns in immunization rates within specific health centers (Kotajuang and Gandapura Health Centers in Bireuen Regency) is commendable, offering a data-driven approach to understanding community health behaviors. The selection of AHC for grouping infants based on variables like gender, address, and immunization type is a suitable methodology for exploratory analysis, aiming to identify distinct segments within the target population. This initiative has the potential to contribute valuable insights for public health planning and resource allocation in local communities. While the premise is strong, the abstract raises several questions regarding the depth and scope of the study. The title suggests an "Infant Routine Monitoring System," but the abstract primarily describes a clustering *analysis* rather than the development or implementation of a functional monitoring system. Further clarification is needed on how the identified clusters directly contribute to or form part of such a system. The presentation of results is also quite limited, offering only one example cluster with generalized percentages and a subjective interpretation ("has started well"). To enhance the value of the research, the abstract should provide more details on the characteristics of *other* identified clusters, the overall number of clusters found, and how these groupings yield more nuanced and actionable insights for health interventions or policy. Methodological specifics such as the distance metric used and how the number of clusters was determined would also strengthen the scientific rigor. In conclusion, this study demonstrates a promising initial step towards leveraging data mining for public health monitoring. To elevate its impact, the authors are encouraged to elaborate on the transition from data clustering to a functional "routine monitoring system," perhaps by outlining how these clusters could trigger specific interventions or resource reallocation based on identified needs. A more comprehensive discussion of the clustering results, including a comparative analysis of different clusters and their implications for public health strategies, would significantly enhance the paper's contribution. Further, a clearer articulation of the study's generalizability and potential for broader application beyond the immediate health centers would be beneficial. Addressing these points would transform a descriptive analysis into a more robust and actionable piece of research.

Full Text

Comments

You need to be logged in to post a comment.