Claim Missing Document
Check
Articles

Found 2 Documents
Search
Journal : Jurnal Sistem Komputer dan Informatika (JSON)

Sistem Pendukung Keputusan Penerimaan Dosen Tetap Menggunakan Metode MOORA dan MOSRA Mesran, Mesran; Aldisa, Rima Tamara; Rangkuti, Wanda Tofani Devi; Sari, Cindy Nanda
Jurnal Sistem Komputer dan Informatika (JSON) Vol. 5 No. 2 (2023): Desember 2023
Publisher : Universitas Budi Darma

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.30865/json.v5i2.7140

Abstract

Lecturers are the forerunners and places to gain knowledge for the nation's children, good lecturers will produce good students too, and good students will become successors to the progress of the nation to be even better, the large number of lecturers at Budi Darma University results in a density of lecturers, it is important to do acceptance of permanent lecturers to provide rewards to lecturers who have worked diligently and earnestly, each lecturer has their own quality but permanent lecturers are lecturers who have a safer position and are trusted by the campus, the importance of selecting permanent lecturers using a system decision support to prevent fraud in the election process. In this study, the MOORA (Multi-Objective Optimization on the Basis Of Ratio Analysis) and MOOSRA (Multi-objective Optimization on the basis of Simple Ratio Analysis) methods are used to assist the selection process in a logical, systemic manner and can produce a decision value on the ranking value. which are different from each formula or algorithm, but these values are equally real and fair without any cheating. In this study the authors also used the ROC (Rank Order Centroid) value to obtain an effective and correct weighting value to perform calculations on the criteria values that had been set by the campus or college of the Budi Darma Medan University. The results in this study based on the calculation of the MOORA method, the highest result was achieved by A1, which is worth 0.4742 and in the MOOSRA method, the highest alternative result was achieved by A1, which is worth 28.1366.
Penerapan Data Mining Untuk Clustering Kualitas Udara Rifqi, Ahmad; Aldisa, Rima Tamara
Jurnal Sistem Komputer dan Informatika (JSON) Vol. 5 No. 2 (2023): Desember 2023
Publisher : Universitas Budi Darma

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.30865/json.v5i2.7145

Abstract

Human health at this time is the key to the continuity of life. Human health is very necessary in the process of development of human life. Environmental health is related to the circumstances or conditions that exist in the surrounding area where you live, whether in a small environment or a large environment. Air quality is the condition of the surrounding air. Air quality is very important for human life because air is what helps humans to live by breathing. With the availability of good air quality, it will certainly be an important factor for an area, not only for health but also for other sectors that interact directly in open areas. The important role of air quality for humans means that more attention needs to be paid and special treatment is given to areas exposed to bad air. The above is a very important problem that must be resolved immediately, if the problem is not resolved immediately it will have an impact on health. The process of solving problems requires a way to resolve them. Where the process of measuring air quality can be seen based on certain conditions or criteria that occur in an area. Data mining is a method used to carry out the problem solving process by processing data. In the process carried out in data mining, there are various ways of solving it. One thing that can be used is clustering. In clustering itself there are various kinds of algorithms such as DBSCAN, K-Means and K-Medoids. In this research, the solution process will use the three algorithms K-Means, K-Medoids and DBSCAN. The purpose of using these three algorithms is to compare the results obtained. In the process carried out in completing data mining, clustering techniques are used using 3 (three) algorithms, namely K-Means, K-Medoids and DBSCAN. The results obtained were that the K-Means algorithm had the highest accuracy value obtained at K=4 with a value of 0.843, for the K-Medoids algorithm the highest value was obtained at K=5 with a value of 0.896 and for the DBSCAN algorithm the highest value was obtained at K=2 with a value of 0.885.