Bulletin of Information System Research

Penerapan Sistem Pendukung Keputusan Desa Terbaik untuk Pemasangan Jaringan Internet Menggunakan Metode MAUT

Novriansyah Putra — 2024-12-23

The internet is currently a very important sector for everyday life, internet connection is not only needed in urban areas, but also in rural areas, especially for village government information systems, the cost of providing internet in rural areas is relatively high, however, the demand for creating an internet network is always high, this condition is utilized by internet service providers (ISP) internet service providers. SPK is a system that plays a role in the process of making a decision in a school, organization or company. Therefore, to determine the best village in installing a village internet network, the Multi Attribute utility method is needed. This MAUT method is a quantitative comparison method that usually combines measurements of different risk costs and benefits, each existing criterion has several alternatives that can provide solutions, to find alternatives that are close to the user's wishes, to identify them, multiplication is carried out on the predetermined priority scale, so that the best and closest results from these alternatives will be taken as a solution.

Implementasi Data Mining Menggunakan Metode Deskripsi Untuk Mengetahui Pola Penentuan Penerima Bantuan Siswa Berprestasi

Piki Ashari Pasaribu — 2024-12-23

Student achievement assistance is one of the programs run at MAN 2 Tapteng to support and motivate its students to continue to excel in academics. However, so far the determination of recipients of student achievement assistance at MAN 2 Tapteng still uses a manual method so that it takes a long time and process so that it is less effective. For that, the author conducted research and created a program to determine the pattern of determining recipients of student achievement assistance at MAN 2 Tapteng to be more efficient. The determination of the pattern of recipients of student achievement assistance is carried out by utilizing student report card data at the end of each semester. The method that is appropriate to this problem is the Description Method. Description is one of the functions in data mining to explore and collect a lot of data. The description method aims to explain or describe a situation, event and object. The results of this study are in the form of a pattern of determining recipients of student achievement assistance from each class at MAN 2 Tapteng. This program is expected to be more efficient in determining the pattern of recipients of student achievement assistance at MAN 2 Tapteng.

Sistem Pendukung Keputusan Penilaian Kinerja Pegawai Menerapkan Metode COPRAS

Muhammad Andika Hasibuan — 2024-12-23

PT. Nusira is a company engaged in the processing of crumb rubber. Crumb Rubber is a plantation trade that has a very important role in Indonesia. This trade can affect economic growth in Indonesia. In the process of improving the quality of work of each employee working at PT. Nusira, it is necessary to conduct an employee performance assessment. In assessing employee performance, a decision support system is needed. The method used in this study is the Complex Proportional Assessment (COPRAS) method using Rank Order Centroid (ROC) weighting. The COPRAS method is a method that can assume direct and proportional dependence of the level of significance and utility of existing alternatives with conflicting criteria, while ROC weighting is used to assign weight values to each criterion. From the calculation of the COPRAS method with ROC weighting, a ranking is obtained in assessing employee performance at PT. Nusira must meet the above criteria by getting the highest score, then the one with the best employee performance is Alternative A2 on behalf of "Suwito" = 100.

Sistem Pendukung Keputusan Memilih Facial Foam untuk Kulit Berminyak Pria dengan UTA

Darwis William Nainggolan — 2024-12-31

Facial skin care in men, especially for those with oily skin, is gaining more and more attention in today's society. The problem of oily skin often leads to various problems, including acne and inflammation, which in turn can affect one's appearance and self-confidence. One solution to this problem is the use of Facial Foam, but choosing the right product is often a challenge, especially among men who are less familiar with skincare products. This research aims to develop a Decision Support System (DSS) using the Utility Theory Additive (UTA) Method to recommend Facial Foam that suits the characteristics of oily skin in men. The methodology used includes data collection regarding criteria, sub-criteria, and alternatives through observation and interviews with users. The five main criteria analysed were Product Effectiveness, Side Effects, Price, Packaging, and Availability. The results of the data analysis show that the Pond's Facial Foam product gets the highest rating of 3,443 so that this product makes it a recommended choice for men with oily skin. This research shows that the UTA Method proves to be effective in simplifying the multicriteria decision-making process related to skincare product selection, providing clear guidance for consumers to choose the product that best suits their needs.

Komparasi Metode Decision Tree dan K-Nearest Neighbor (KNN) dalam Memprediksi Costumer Churn Pada Perusahaan Telekomunikasi

Khadisah Syah Riebhan Palluvi — 2024-12-31

Prediksi customer churn bertujuan untuk mengklasifikasikan data pelanggan sebelumnya menjadi dua kategori: pelanggan yang akan berhenti berlangganan dan pelanggan yang akan terus berlangganan. Prediksi tersebut memanfaatkan ilmu data mining peran klasifikasi yang merupakan menempatkan variabel atau objek ke dalam beberapa kategori relevan yang telah ditetapkan sebelumnya. Dalam proses eksekusi data mining, diperlukan sebuah algoritma yang dapat mengklasifikasikan apakah customer churn atau tidak churn. Data yang digunakan terdiri dari 7043 rows dan 21 columns. Didalam data tersebut salah satu kolom akan dijadikan label yaitu kolom ‘Churn’. Dalam proses prediksi churn, algoritma yang digunakan yaitu Decision Tree dan K-Nearest Neighbor. Dari hasil analisis yang dilakukan, pada algoritma KNN dihasilkan 76% dan Decision Tree 72%. Dengan hasil pemodelan akurasi 72% dan 76%, keduanya memenuhi kriteria kesuksesan >70%. Namun, model KNN dengan akurasi 76% lebih baik dan lebih diinginkan karena memberikan prediksi yang lebih akurat.