Claim Missing Document
Check
Articles

Found 8 Documents
Search
Journal : INCODING: Journal of Informatics and Computer Science Engineering

Penggunaan Algoritma Fuzzy C-Means untuk Optimalisasi Pengelompokan Data Cuaca dalam Prediksi Curah Hujan di Indonesia Hidayah, Safrina; Muliono, Rizki
INCODING: Journal of Informatics and Computer Science Engineering Vol 5, No 2 (2025): INCODING OKTOBER
Publisher : Mahesa Research Institute

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.34007/incoding.v5i2.839

Abstract

This study develops an information system to optimize rainfall data clustering in Indonesia using the Fuzzy C-Means method. Rainfall clustering aims to provide accurate information about climatic conditions by categorizing regions into three rainfall levels: high, medium, and low. The data used in this study were obtained from observations by the Meteorology, Climatology, and Geophysics Agency (BMKG) from 2011 to 2015 across various provinces. The Fuzzy C-Means method was selected due to its ability to handle uncertainty by assigning membership degrees to each cluster. The resulting clustering information is expected to assist the community and relevant sectors such as agriculture, fisheries, and regional planning in predicting rainfall and making informed decisions. The developed system can also be extended to process other weather data, including air quality and wind speed.
Penerapan Data Mining Menggunakan Algoritma K-Medoids Dalam Pengelompokan Nasabah Penerima Reward Pada PT Dotri Gadai Jaya Zebua, Meniati; Muliono, Rizki
INCODING: Journal of Informatics and Computer Science Engineering Vol 4, No 2 (2024): INCODING OKTOBER
Publisher : Mahesa Research Institute

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.34007/incoding.v4i2.709

Abstract

The increasing growth of the financial industry makes companies experience intense competitive pressure. PT Dotri Gadai Jaya (PT DGJ) is a private pawnshop company, facing challenges in maintaining and increasing customer loyalty amidst this competition. One of the strategies used by PT DGJ is to provide rewards to customers based on the number of pawn loan transactions. However, companies experience difficulties in grouping reward recipient customers efficiently and accurately. To overcome this problem, it is necessary to apply data mining using the K-Medoids algorithm. The main objective of this research is to apply the K-Medoids algorithm in grouping reward recipient customers at PT DGJ, knowing the grouping results, and evaluating the results using the Davies Bouldin Index (DBI). The results of the grouping of 1,085 customers are 314 customers who received a 30% reward, 540 customers with a 20% reward and 231 customers with a 10% reward. The cluster evaluation result using DBI is 0.368812, which means the cluster quality value is close to 0 or is quite small. So it can be said that the resulting cluster is quite good.
Analisis Klustering Menggunakan Algoritma DBSCAN untuk Deteksi Anomali dalam Data Transaksi Keuangan Alwi, Buchori; Muliono, Rizki
INCODING: Journal of Informatics and Computer Science Engineering Vol 5, No 1 (2025): INCODING APRIL
Publisher : Mahesa Research Institute

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.34007/incoding.v5i1.827

Abstract

Anomaly detection in financial transaction data is a crucial aspect due to the increasing use of e-money, which raises the risk of suspicious activities such as fraud and money laundering. This study applies the Density-Based Spatial Clustering of Applications with Noise (DBSCAN) algorithm to cluster transaction data and identify anomalies based on three main variables: transaction amount, transaction frequency, and final balance. The optimal parameters were determined by evaluating various combinations of epsilon (ε) and minPts values using the Davies-Bouldin Index (DBI) as a clustering quality indicator. The analysis results indicate that the optimal parameters are ε of 0.2727 and minPts of 6, with a DBI score of 1.1753. DBSCAN successfully formed six main clusters and detected 138 data points as noise, indicating potentially abnormal transactions. These findings demonstrate that DBSCAN can effectively distinguish between normal and suspicious data without requiring prior assumptions on the number of clusters, contributing to the development of more accurate and adaptive digital transaction anomaly detection systems.
Perancangan Sistem Informasi Manajemen Dalam Pengelolaan Data Kepegawain Di Kantor Dinas Perkebunan Provinsi Sumatera Utara Darkani, M. Farhan; Muliono, Rizki
INCODING: Journal of Informatics and Computer Science Engineering Vol 4, No 2 (2024): INCODING OKTOBER
Publisher : Mahesa Research Institute

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.34007/incoding.v4i2.731

Abstract

The use of this information system will provide convenience for the North Sumatra Plantation Service in managing employees, especially in terms of inputting employee data. In connection with the data input mechanism still using the classic method, namely using MS Excel, which the author considers to be less flexible. Therefore, through this Internship, the author hopes to be able to design a web application where employees can input their data more easily and flexibly. The creation of this system starts from data collection, system analysis, system design and implementation.
Deteksi Pola Kunjungan Pasien Berdasarkan Status Kesehatan Menggunakan Algoritma DBSCAN Razaq, Faisal; Muliono, Rizki
INCODING: Journal of Informatics and Computer Science Engineering Vol 5, No 2 (2025): INCODING OKTOBER
Publisher : Mahesa Research Institute

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.34007/incoding.v5i2.979

Abstract

This study identified eight visit clusters grouped into four service profiles: Acute (Clusters 1 5; 1,186/3,000 ≈ 39.5%; mean age 22.8 years; peaks on Saturday at 19:00 and Thursday at 08:00; predominant diagnoses: dengue fever, typhoid, acute respiratory infection, influenza, and gastroenteritis), Chronic (Clusters 3 4; 924/3,000 ≈ 30.8%; mean age 66–67 years; peaks on Thursday at 08:00 and Friday at 13:00; predominantly COPD, type 2 diabetes mellitus, heart failure, hypertension, and kidney failure), Routine Follow-up (Clusters 2 7; 590/3,000 ≈ 19.7%; mean age 41–42 years; peaks on Thursday at 11:00 and Friday at 15:00; including post-operative follow-up, annual check-ups, adult vaccination, cholesterol screening, and nutrition counseling), and Emergency (Clusters 0 6; 300/3,000 = 10%; mean age 44–46 years; peaks at 22:00 on Thursdays and Sundays; predominantly ischemic stroke, myocardial infarction, road-traffic injuries, appendicitis, and asthma exacerbations). The age–time–diagnosis patterns indicate a distinct segmentation of service needs: acute cases are concentrated among younger patients and peak on weekends and weekday mornings; chronic cases cluster among older adults with morning–midday weekday peaks.
Penerapan Algoritma K-Means Untuk Klasterisasi Pasien Berdasarkan Riwayat Kesehatan dan Jenis Layanan Kesehatan Putri, Riza Dwi; Muliono, Rizki
INCODING: Journal of Informatics and Computer Science Engineering Vol 5, No 2 (2025): INCODING OKTOBER
Publisher : Mahesa Research Institute

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.34007/incoding.v5i2.973

Abstract

The digital transformation in the healthcare sector has led to the generation of large and complex datasets, requiring appropriate analytical techniques to extract meaningful information. This study aims to implement the K-Means algorithm to cluster patients based on their health history and the types of healthcare services they use, in order to support data-driven decision-making in hospital management. The dataset consists of 1,459 patient records from Sapta Medika Hospital, covering attributes such as age, gender, chronic disease history (diabetes, hypertension, heart disease), visit frequency, medical costs, and healthcare service types including outpatient, inpatient, emergency (ER), and telemedicine. The research stages involved data preprocessing, transformation, categorical data encoding, numerical data normalization, and clustering using the K-Means algorithm. The optimal number of clusters was determined using the Elbow Method, which identified K = 3. The clustering results revealed three distinct patient groups: chronic patients with high treatment costs and frequent inpatient services, routine patients with stable conditions mostly using outpatient services, and general patients, usually younger with mild conditions. Principal Component Analysis (PCA) was used to visualize the cluster separation, while the clustering quality was evaluated with a Silhouette Score of 0.47. These results conclude that the K-Means algorithm is effective in producing meaningful and practical patient segmentation, which can be used to design more adaptive, efficient, and patient-centered healthcare service strategies.
Rancang Bangun Aplikasi Penerjemah Bahasa Indonesia Bahasa Nias Menggunakan Algoritma Levensthein Distance Laia, Aldi Irfan; Muliono, Rizki
INCODING: Journal of Informatics and Computer Science Engineering Vol 4, No 1 (2024): INCODING APRIL
Publisher : Mahesa Research Institute

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.34007/incoding.v4i1.469

Abstract

In this context, there is a need for the development of a translator application that can translate between Indonesian and Nias languages. The aim of this research is to design and build an Android-based translator application that can translate text from Indonesian to Nias and vice versa.In this study, the Levenshtein Distance algorithm was used as a method to perform the translation process. This algorithm uses the calculation of the distance between two strings to identify the similarities and differences between words in Indonesian and Nias. Thus, the Levenshtein Distance can be used to produce accurate and relevant translation results.The design and development of the translator app was done using the Android platform as the development base. The application is designed to provide an intuitive and easy-to-use user interface, thus allowing users to quickly and efficiently translate texts into Indonesian and Nias languages. During the development process, the translator application was tested to ensure good quality and performance. The test results showed that the translation application was capable of producing accurate translations in accordance with the context of the Indonesian language and the Nias language. In conclusion, the research successfully designed and built an Android-based Indonesian-Nias language translator application using the Levenshtein Distance algorithm. This application is expected to be a useful tool in facilitating communication between Indonesian-speaking users and the Nias language. This research also contributed to the development of Nias-based translator applications that were still limited in previous literature.
Analisis Persebaran Penyakit di Wilayah Menggunakan Algoritma K-Means Berbasis Data Kunjungan Fasilitas Kesehatan Suhaira, Zatin; Muliono, Rizki
INCODING: Journal of Informatics and Computer Science Engineering Vol 5, No 2 (2025): INCODING OKTOBER
Publisher : Mahesa Research Institute

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.34007/incoding.v5i2.983

Abstract

This study aims to analyze the distribution of diseases based on patient visit data to various healthcare facilities using the K-Means clustering method. The research data were obtained secondarily from the Kaggle platform, namely the ‘Healthcare Dataset’, which contains patient information, including healthcare facility attributes, medical conditions, and other related data. The determination of the optimal number of clusters was carried out using the Elbow Method, while the quality of clustering was evaluated with two internal metrics, namely the Silhouette Score and the Davies–Bouldin Index (DBI). The clustering results produced three main clusters with distinct characteristics. The first cluster was dominated by patients diagnosed with arthritis in the age group of 55–59 years with blood type O+. The second cluster showed a predominance of obesity in the age group of 35–39 years with blood type AB+, while the third cluster indicated cancer cases in the age group of 65–69 years with blood type O-. The evaluation resulted in a Silhouette Score of 0.5349 and a DBI of 0.5830, indicating that the clustering quality is fairly good, with compact and well-separated clusters. These findings not only highlight variations in disease distribution across healthcare facilities but also provide a foundation for mapping disease patterns and supporting strategic decision-making in public health..