Garuda - Garba Rujukan Digital

Implementasi K-Means Clustering Untuk Mengelompokan Data Sparepart Alat Berat Andira Rahmawati; Muhammad Fadlan; Anto Anto
Journal of Big Data Analytic and Artificial Intelligence Vol 7 No 2 (2024): JBIDAI Desember 2024
Publisher : STMIK PPKIA Tarakanita Rahmawati

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.71302/jbidai.v7i2.37

Data mining is a crucial process for extracting valuable information from existing data, which can then be used by companies for quick and accurate decision-making. One of the commonly used methods in data mining is the K-Means Clustering method. In this study, the author applied K-Means Clustering in the retail sector to address the challenges faced by PT. Patria Jaya Mandiri. The author designed an application that can cluster heavy equipment spare parts based on sales data, with the aim of helping the company identify which spare parts are most favored by consumers. This clustering is expected to simplify the process of determining optimal spare part stock, ultimately positively impacting the company’s revenue. The results of this study indicate that heavy equipment spare parts can be categorized into three groups: Most Popular, Popular, and Least Popular. Cluster 1 (Most Popular) consists of 3 data points, Cluster 2 (Popular) consists of 39 data points, and Cluster 3 (Least Popular) consists of 8 data points. This clustering result can serve as a guide for PT. Patria Jaya Mandiri in determining the optimal spare part inventory in the future.

Implementasi Algoritma K-Medoids Dalam Mengelompokkan Siswa Berdasarkan Keaktifan Dalam Proses Pembelajaran Noor Oktavia Ih’Diati; Anto Anto; Rosmini Rosmini
Journal of Big Data Analytic and Artificial Intelligence Vol 7 No 1 (2024): JBIDAI Juni 2024
Publisher : STMIK PPKIA Tarakanita Rahmawati

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.71302/jbidai.v7i1.38

Grouping students based on their level of engagement is an effective strategy to improve the quality of learning. SMP 9 Tarakan currently does not have a system that can group students based on their engagement in the learning process, which could assist in evaluating learning outcomes. In the initial stage of applying this method, the data collected came from the report card grades of 8th-grade students (Class VIII I) in the 2nd semester (Even Semester) of the 2022/2023 academic year. The characteristics used in the analysis include grades in Religion, Civic Education (PPKn), Mathematics, Science (IPA), Social Studies (IPS), Indonesian Language, English, Physical Education (Penjaskes), and Cultural Arts and Skills, with a total of 31 data points analyzed. The second step is to determine the number of clusters. The third step involves randomly selecting clusters with an initial medoid. The fourth step is to calculate the distance for each student using the Euclidean distance method, then mark the nearest distance and calculate the total distance. The fifth step is to calculate the total deviation (S) and use the Davies-Bouldin Index (DBI) to find the optimal value of k by conducting tests five times with k=3. Based on the calculation results, the analysis of student data grouping produced three clusters using Euclidean distance and Davies-Bouldin Index calculations. The results show that 3 students fall into the Highly Interested cluster, 4 students into the Interested cluster, and 24 students into the Less Interested cluster.

Implementasi Data Mining Menggunakan Algoritma Hash Based Terhadap Pola Pengeluaran Obat Pada Rumah Sakit Pertamina Tarakan Febriyan Febriyan; Rizky Juliansyah; Sinawati Sinawati; Anto Anto
Journal of Big Data Analytic and Artificial Intelligence Vol 7 No 1 (2024): JBIDAI Juni 2024
Publisher : STMIK PPKIA Tarakanita Rahmawati

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.71302/jbidai.v7i1.41

The role of information systems has expanded to other layers of life, especially in the health sector. Improving service and performance productivity in the health sector, including pharmacies. This study identifies drug dispensing patterns at the Pertamina Tarakan Hospital.This research was carried out in several stages. First, create tabular transaction data. Second, determine minimum support. Third, determine the memory address. Fourth, calculate the minimum confidence. Fifth, calculate confidence, and last, make a decision.This study focuses on Rs Pertamina Tarakan to identify common spending patterns. For example, in the previous 12 months, what drug combination was most commonly administered to patients. Knowing dispensing patterns allows Rs Pertamina Tarakan to arrange the layout of medicines, resulting in more optimal and excellent service in the medicine administration department. The results showed that in 2022, three pairs of drug combinations placed close together, namely Ranitidin, Urdahex Cap first pair, Urdahex Cap, Folamil genio kap, Propylathiorasil second pair, and Propylathiorasil, Ranitidin, and Urdahex Cap third pair.

Sistem Pakar Pendeteksi Kerusakan Mesin Genset Menggunakan Metode Case Based Reasoning Lestari Ningsih; Fitria; Anto Anto
Journal of Big Data Analytic and Artificial Intelligence Vol 6 No 2 (2023): JBIDAI Desember 2023
Publisher : STMIK PPKIA Tarakanita Rahmawati

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.71302/jbidai.v6i2.43

PT. Sabindo Raya Gemilang is an exporter of shrimp and fish in Indonesia. This company requires an enormous flow of electricity to support activities so that it is not bothered by blackouts from the PLN. A generator set or genset is a substitute for a voltage source. The problem is that genset damage requires reliable mechanical handling. Junior mechanics sometimes misanalyze damage, requiring the supervision of a senior mechanic. This study applies an expert system with the Case-Based Reasoning (CBR) method to detect genset damage. It meets the case researched for having old cases to compare to new ones through the Retrieve, Reuse, Revise, and Retain so that problems are solved easily and quickly. The validity test showed that 5 of 6 test data were equal to the expert's diagnosis. Expert system accuracy 83%. There was a system diagnosis different from experts because the system’s diagnosis was based on the symptoms weight calculation and made a diagnosis with the highest similarity value. Meanwhile, experts observed the genset damage directly.

Analisa Metode Association Rule Untuk Penjualan Skincare Menggunakan Algoritma Pincer Serach Nadia Indah Tarakanita; Yusni Amaliah; Anto Anto
Journal of Big Data Analytic and Artificial Intelligence Vol 6 No 2 (2023): JBIDAI Desember 2023
Publisher : STMIK PPKIA Tarakanita Rahmawati

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.71302/jbidai.v6i2.46

Rumah Masker Tarakan is a business entity in the field of selling cosmetic and skincare products. A common problem faced by Rumah Masker is the placement of products, which often leads to difficulties in finding similar products. Therefore, a method is needed to organize the layout of skincare products that are frequently purchased together, making it easier for customers to select the products they want. The Pincer Search Algorithm, also known as the Two-Way Search, uses two approaches: Top-Down and Bottom-Up. In its process, the primary direction of the Pincer Search is Bottom-Up. The Maximum Frequent Set is a collection of maximal itemsets that are classified as frequent. The purpose of the Maximum Frequent Set is to reduce (prune) the number of candidate Frequent Itemsets that need to be examined in the Bottom-Up process. Based on research conducted using the Pincer Search Algorithm with sales data from January to June 2022, involving a total of 148 data points, it was found that 72 transactions yielded a value of 0.028% for a candidate of 3 itemsets. This demonstrates that the best sales performance occurs with a combination of 3 itemsets.

Analisa Tingkat Kepuasan Pelanggan Menggunakan Metode Servqual Dan Lexicon Based Ahmad Syar; Nurhalizah Noor; Aida Indriani; Anto Anto
Journal of Big Data Analytic and Artificial Intelligence Vol 5 No 1 (2019): JBIDAI Juni 2019
Publisher : STMIK PPKIA Tarakanita Rahmawati

Show Abstract | Download Original | Original Source | Check in Google Scholar

Customer satisfaction can be achieved through delivering optimal service performance to customers. Excellent service is realized when a business can maintain and develop customer loyalty while continuously improving the quality of services provided. In the case of Trijaya Computer Store, there is currently no evaluation framework in place to assess service quality, making it difficult to gauge customer satisfaction. In this study, two methods are employed: the SERVQUAL method and the Lexicon-Based method. The SERVQUAL method is used to measure service quality and customer satisfaction based on the services provided, while the lexicon-based method identifies which variables within the service quality dimensions need improvement by analyzing the three highest negative opinion scores. The analysis using the SERVQUAL method reveals that the largest service quality gap occurs in the Tangibles dimension, with a gap score of -0.44, indicating that customer expectations in this dimension are significantly unmet, leading to dissatisfaction. In the lexicon-based sentiment analysis, the highest negative opinion scores are 21, 17, and 16, corresponding to variables V3, V6, and V2, respectively. These variables represent specific aspects of the service that need improvement at Trijaya Computer. The variable V6, which states that "Trijaya Computer employees can resolve issues in a timely manner," is particularly important, as it appears in the results of both methods, indicating that timely problem resolution is a key concern for customers and should be prioritized for improvement. By addressing these issues, Trijaya Computer can enhance its service quality and improve customer satisfaction.

Implementasi Moora Pada Penilaian K3 Pemerintah Kota Tarakan Romadan; Rusmin; Yusni Amaliah; Anto Anto
Journal of Big Data Analytic and Artificial Intelligence Vol 5 No 1 (2019): JBIDAI Juni 2019
Publisher : STMIK PPKIA Tarakanita Rahmawati

Show Abstract | Download Original | Original Source | Check in Google Scholar

The Tarakan City Government, specifically the Organization Division of the Regional Secretariat (Setda), has struggled to maintain clean, organized, and aesthetically pleasing office environments, leading to numerous public complaints. These issues have negatively impacted the government's image and service quality. To address this, a decision support system is needed to help the Regional Secretary identify which government offices meet the standards for being comfortable and welcoming, based on Occupational Health and Safety (OHS) criteria outlined in Perwali No. 14 of 2017. Currently, Setda lacks an integrated information system to assess which offices meet these standards and identify which criteria require improvement. The existing evaluation method, done through Microsoft Excel, is inefficient, making the decision-making process less effective. In response, this study proposes a system that can categorize offices based on OHS standards and highlight criteria for improvement using the MOORA method. The study evaluates 16 government offices as alternatives, with data collected from interviews and the 2017 OHS evaluation sheets, covering eight criteria (seven benefit and one cost criterion) and 24 sub-criteria. The MOORA method is applied to generate final scores that provide rankings, categories, and improvement criteria. The OHS categories are defined as Green (scores between 70 and 90, indicating a high level of OHS), Yellow (scores between 50 and 70, indicating a moderate level of OHS), and Red (scores between 30 and 50, indicating a low level of OHS). The results show that the system is effective, informative, and efficient in displaying the OHS status of the offices. Out of the 16 offices, 11 are classified in the Green category, while 5 fall into the Yellow category. The Green category scores range from 70.78 to 79.63, and the Yellow category scores range from 66.26 to 69.09. The study identifies the need for improvements, particularly in Waste Management and Policies/Innovations by the Heads of Offices. This MOORA-based decision support system enables the government to better evaluate and improve the OHS performance of its offices, contributing to enhanced service quality and government reputation.

Klasifikasi Penyakit Karies Gigi Menggunakan Algoritma Modified K-Nearest Neighbor Arnold Kalalo; Rosmini Rosmini; Anto Anto
Journal of Big Data Analytic and Artificial Intelligence Vol 7 No 2 (2024): JBIDAI Desember 2024
Publisher : STMIK PPKIA Tarakanita Rahmawati

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.71302/jbidai.v7i2.60

Dental caries, commonly known as tooth cavities, is a disease where bacteria damage the structure of tooth tissues such as enamel, dentin, and cementum. The primary cause of dental caries is the demineralization of tooth surfaces caused by organic acids from sugary foods. If dental caries is not promptly treated or checked from the beginning, the damage can worsen to the point where the tooth must be extracted. To facilitate identifying the severity of caries, a dental caries classification system was developed using the MKNN (Modified K-Nearest Neighbor) algorithm. The MKNN method is an enhancement of the KNN method, with the main differences being in the calculation of training data validity and the weight voting process. In this study, there are three different classes of dental caries and six symptoms or variables. The stages of the MKNN method used are: distance calculation using Euclidean distance, testing the validity of training data, determining k based on distance calculation, and weight voting calculation in KNN. The test results show that the k value, the number of training data, and the number of test data affect the classification results. The classification results from the test using 20 training data, 10 test data, and k=3 are as follows: 1 patient classified with superficial caries, 5 patients with media caries, and 3 patients with profunda caries. The diagnosis produced by the application is consistent with the expert (doctor) diagnosis.

Title

Found 8 Documents
Search

Abstract

Abstract

Abstract

Abstract

Abstract

Abstract

Abstract

Abstract

Title Search

Found 8 Documents Search

Abstract

Abstract

Abstract

Abstract

Abstract

Abstract

Abstract

Abstract

Title

Found 8 Documents
Search