Based on observations made by researchers, FinTech which is common and is being used by the people of Indonesia is the first generation for payment, purchase of goods and services, one of which is the Fund Application. In choosing the application to be used, it is usually considered, comfort, safety, accuracy of the transaction, comfort, and various promotions. But some users are hesitant in using the application because of some reviews from the application that show positive and negative ratings. With the number of reviews paid in the comments column provided by the Google Play Store in the Funds Application needed to classify the reviews given as positive or negative. The research used was an experimental method using the Naive Bayes Classifier Algorithm and K-Nearest Neighbor. Regarding testing on the Fund Application has the best testing value of 84.76%. Then it can be considered a review that can be approved from positive reviews by other users and if there are negative reviews it will be input to the company to develop and develop the product.
<p>Hampir setiap pelajar di Indonesia terdaftar dengan atribut profil yang lengkap, seperti : Nama, Jenis Kelamin, Jenis Tinggal, Alat Transportasi, Usia Orangtua, Pendidikan Orangtua, Pekerjaan Orangtua, Penghasilan Orangtua dan atribut lainnya. Dari data atribut profil tersebut dapat diklasterisasi berdasarkan kedekatan nilai antara atribut yang dimiliki masing-masing siswa. Disisi lain siswa juga memiliki data yang berisi nilai akademis yang juga dapat dibuat klasterisasi.</p><p>Data yang dipakai dalam penelitian ini melibatkan 512 instances yang didapat dari sebuah Sekolah Menengah Kejuruan (SMK) di Jakarta. Metode yang pakai untuk klasterisasi menggunakan algoritma <em>K-Means.</em> Penelitian ini akan mencari korelasi klasterisasi profil siswa terhadap nilai akademisnya.</p><p>Tahapan penelitian diawali dengan persiapan dataset profil dan dataset nilai siswa, atribut dari dataset profil yang dipakai hanya atribut yang dianggap dapat merepresentasikan profil siswa dan keluarganya. Tahap berikutnya adalah mentrasformasi data atribut non numerik (kategorik dan interval) menjadi numerik. Dilanjutkan dengan tahap perhitungan jarak antar data dan tahap terakhir mencari pola korelasi antara klaster profil dan klaster nilai akademis yang terbentuk.</p><p>Dengan metode <em>elbow</em> jumlah klaster yang paling ideal dalam penelitian ini adalah antara 3 dan 4 klaster, dimana nilai <em>Silhoutte Coefficient</em> tertinggi adalah 0,8103 untuk penglompokan 3 klaster.</p><p> </p><p><em><strong>Abstract</strong></em></p><p class="Judul2"><em>Almost every student in Indonesia is registered with complete profile attributes, such as: Name, Gender, Type of Stay, Transportation Equipment, Parents' Age, Parental Education, Parents' Work, Parents' Earnings and other attributes. From the profile attribute data it can be clustered based on the closeness of the values between the attributes possessed by each student. On the other hand students also have attribute data that contains academic values that can also be clustered.</em></p><p class="Judul2"><em>The data used in this study involved 512 instances obtained from a Vocational High School (SMK) in Jakarta. The method used for clustering is using the K-Means algorithm. This research will look for correlation of student profile clustering to its academic value.</em></p><p class="Judul2"><em>The stages of the research began with the preparation of the profile dataset and the student value dataset, the attributes of the profile dataset used were only those attributes that were considered to represent the profiles of students and their families. The next step is to transform non-numeric attribute data (categorical and interval) into numeric. Followed by the stage of calculating the distance between data and the final stage looking for patterns of correlation between profile clusters and academic value clusters that are formed.</em></p><p class="Judul2"><em>With the elbow method, the most ideal number of clusters in this study is between 3 and 4 clusters, where the highest Silhoutte Coe</em><em>f</em><em>ficient value is 0.8103 for grouping 3 clusters.</em></p><p><em><strong><br /></strong></em></p>
scite is a Brooklyn-based organization that helps researchers better discover and understand research articles through Smart Citations–citations that display the context of the citation and describe whether the article provides supporting or contrasting evidence. scite is used by students and researchers from around the world and is funded in part by the National Science Foundation and the National Institute on Drug Abuse of the National Institutes of Health.
hi@scite.ai
10624 S. Eastern Ave., Ste. A-614
Henderson, NV 89052, USA
Copyright © 2024 scite LLC. All rights reserved.
Made with 💙 for researchers
Part of the Research Solutions Family.