IMPLEMENTATION OF DATA MINING BY USING K-MEANS TO CLASSIFY MARRIAGE AGE

: Marriage is a husband and wife relationship between a man and a woman to form a family. There are several conditions in marriage that must be fulfilled both religiously and legally in force in Indonesia. To carry out the marriage, the prospective bride and groom must register at the nearest Religious Affairs Office (KUA), KUA is an institution established by the government to handle marriage matters. At marriages, various age groups are often found registering at the KUA. This research was conducted using the Data Mining technique through the K-Means Clustering Model to determine the age grouping of marriage which aims to make it easier for the Office of Religious Affairs in educating the prospective bride and groom from a future perspective and an economic perspective in terms of having a child. The research dataset is data on prospective wedding brides at KUA Rawang Lama, Panca Arga in 2022 with a total of 102 samples, by forming 3 clusters, namely: the Ideal cluster of 76 prospective wedding brides (age 19-30 based on husband's age and age 18-25 based on age wife), a good cluster of 20 prospective marriage brides (age 28-44 based on husband's age and age 24-37 based on wife's age), and a risky cluster of 6 prospective marriage brides (age 49-72 based on husband's age and age 39-58 based on wife's age), and produces a Silhouette Score of 0.57


INTRODUCTION
Information technology is always developing rapidly.This development is the right opportunity to obtain more effective and efficient but diverse data.To process this data, a technique is needed so that the processing results or information obtained are appropriate.One technique that can be used is data mining [1].
The Office of Religious Affairs (KUA) is the frontline work unit of the Ministry of Religion which carries out governmental duties in the field of Islamic Religion, in the District area.It is said to be the foremost work unit because the Office of Religious Affairs (KUA) directly deals with the community.Because of that, it is only natural that the existence of the Office of Religious Affairs (KUA) is considered very urgent along with the existence of the Ministry of Religion [2].
The fact, there are still some people who do not understand the duties and functions of the Office of Religious Affairs (KUA).The result is not surprising, there is an impression that the duties and functions of the Office of Religious Affairs (KUA) are only limited to reading prayers and marrying off [3].
KUA Rawang Lama, Panca Arga is one of the religious affairs offices that handles various marriages where many age groups register their marriages both religiously and officially in the state.However, from the information available, the age of marriage is 19 years old who can legally marry in the country which has been regulated in the laws currently in force.Marriage under the age of 19 can only be married according to religion with various conditions and considerations that have been agreed upon.In 2022 at the KUA Rawang Lama, Panca Arga, there are 102 marriage registrars, the age groups who register marriages at the KUA are: ages 19-72 for men and ages 18-58 for women.
The issue of marriage related to age limits, namely underage or even below the minimum age for marriage is a complex discourse related to both legal and non-legal aspects.In this regard, the question of marriage collides diametrically with legal provisions that stipulate a minimum age for marriage [4].This problem lies in how mentally prepared one has to live a household life with a partner.Not ready mentally and materially, it's better to postpone wedding plans in advance.Instead of living life as an unhappy husband and wife.
It is common knowledge that the ideal age for marriage in Indonesia is less than 20 years, especially for women [5].With the data owned by the KUA Rawang Lama, Panca Arga, it fits the grouping of three categories, so the categories are ideal, good, and risky in determining the division of marriage ages.
Tahaga (Ketahanan Keluarga) is also known as the strength or resilience family.This is related to personal and family abilities to utilize their potential to face life's challenges, including the ability to restore family functions to their original state in facing challenges and crises [6].This problem affects the need for housing to the cost of raising a child and reducing divorce cases, which often occur due to economic problems and readiness for a household.

METHOD
Data mining and machine learning techniques can be used to make predictions based on past data.Data mining is the process of finding useful patterns in large data sets.From other sources, data mining is the study of collecting, cleaning, processing, analyzing, and obtaining useful insights from data [7].This data was obtained from KUA Rawang Lama, Panca Arga and has not gone through a cleaning process, so a lot of data is private and must be deleted.

Image 1. Decription of Data KUA
There are several different approaches classified as information seeking techniques in KDD.There are quantitative approaches, such as proballistic and statistical approaches.
Several approaches make use of visualization techniques, classification approaches such as inductive logic, decision tree analysis and pattern finding.Other approaches include genetic algorithms, trend analysis, artificial neural networks, deviations and a mixed approach of two or more of the existing approaches [8].
The first step in the data mining process is data cleaning.the activities carried out in this process are checking inconsistent data, correcting errors in data and removing data duplication [9].
The dataset below has gone through a cleaning process, which deleted some unnecessary data such as tanggal daftar, NIK suami, tempat lahir suami, NIK istri, tempat lahir istri where this data is very private and may not be published.

Image 2. Decription of Data Used
Clustering model analysis is a technique of multivariable analysis that is used to group objects (variables or data) that have similarities into one group so that they can produce information in testing the object and then present a hypothesis based on the relationships that occure [10].
Centroid is the data center point for calculating the vector mean as a centroid.In applying the K-means algorithm, the midpoint or centroid value is generated from the data obtained from each cluster [11].Clustering steps contained in the K-Means algorithm [12].

Image 3. K-Means Clustering Method
The first step of the K-Means algorithm is to determine the number of clusters, in this study 3 clusters were determined [13].Namely the ideal cluster (C0), the risky cluster (C1), and the good cluster (C2).
use the Euclidean Distance formula to calculate the distance of each input data to each centroid until the closest Based on Figure 8, the results of Silhouette Score is 0.57 for the K-Means method.The closer to 1, the better the model.

CONCLUSION
Based on analysis by using the K-Means Algorithm, there are 3  based on wife's age) then it produces a Silhouette Score of 0.57 for the K-Means method.The Silhouette Score model has a value between -1 to 1, the closer to 1, the better the model, which is included in the Good Classification category.Therefore, the K-Means method is a model that is categorized as good and is implemented to predict clusters based on previous historical experience to simplify the process of classifying the marriage age of the bride and groom.
clusters, namely: The Ideal Cluster of 76 prospective wedding brides (age 19-30 based on husband's age and age 18-25 based on age wife), A Good Cluster of 20 prospective marriage brides (age 28-44 based on husband's age and age 24-37 based on wife's age), and A Risky Cluster of 6 prospective marriage brides (age 49-72 based on husband's age and age 39-58