Claim Missing Document
Check
Articles

Analysis of Naïve Bayes and K-Nearest Neighbors Algorithms for Classifying Fishermen Aid Eligibility Nasrullah, Muhammad; Surarso, Bayu; Nurhayati, Oky Dwi
Jurnal Penelitian Pendidikan IPA Vol 10 No 10 (2024): October
Publisher : Postgraduate, University of Mataram

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.29303/jppipa.v10i10.8818

Abstract

This article analyzes the use of data mining with Naïve Bayes and K-Nearest Neighbor (KNN) algorithms to build classification models and evaluate their performance in identifying fishermen eligible for aid. The study aims to compare the effectiveness of these algorithms in handling imbalanced datasets using the Synthetic Minority Over-sampling Technique (SMOTE). The research applies SMOTE to improve the balance of the dataset before classification. Without SMOTE, Naïve Bayes achieved an accuracy of 97.01%, precision of 94.16%, recall of 96.67%, and F1-score of 95.39%. KNN, on the other hand, reached an accuracy of 94.04%, precision of 94.53%, recall of 86.00%, and F1-score of 90.06%. After applying SMOTE, both algorithms improved: Naïve Bayes attained an accuracy of 98.33%, precision of 96.86%, recall of 100.00%, and F1-score of 98.49%, while KNN reached an accuracy of 96.90%, precision of 97.72%, recall of 96.19%, and F1-score of 96.94%. The results show that Naïve Bayes, with SMOTE, outperforms KNN in managing data imbalance and accurately classifying eligible fishermen for aid.
Multi-criteria Group Decision Making Model to Selecting the Most Appropriate Performance of Contract Employees Using the Weighted Aggregated Sum Product Assessment and Borda Methods Puspita, Yuanita Candra; Surarso, Bayu; Suseno, Jatmiko Endro
Jurnal Penelitian Pendidikan IPA Vol 10 No 12 (2024): December
Publisher : Postgraduate, University of Mataram

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.29303/jppipa.v10i12.9566

Abstract

This research aims to evaluate the performance of contract employees using the Weighted Aggregated Sum Product Assessment and Borda methods. The Weighted Aggregated Sum Product Assessment approach uses mathematical calculations, thereby producing precise and accurate results. The Borda technique is useful because it efficiently aggregates diverse rankings of alternatives through voting groups that combine many preferences. The choice of these two methodologies has the ability to produce more efficient solutions to complex problems by guiding or making decisions more quickly in resolving problems. The determining criteria for assessing the performance of contract employees are punctuality or attendance, skills, knowledge, communication, decision making, discipline, adaptation to the work environment, work safety and initiative.  The results of the calculation process from the ranking combination of the weighted aggregate sum product assessment and the borda method were obtained by placing alternative 5 as an employee who is worthy of being a permanent employee and getting the highest score of 0.288. All calculations using this method can be a guide in making decisions at PT SMART Tbk Surabaya, because the accuracy of contract employee performance evaluations has a significant influence on achieving company goals in the permanent employee recruitment process.
Customer Segmentation Based on Recency, Frequency, Monetary Analysis Using K-Means Algorithms in Apple Ecosystem Setiawan, Edwin; Surarso, Bayu; Nugraheni, Dinar Mutiara Kusumo
Jurnal Penelitian Pendidikan IPA Vol 11 No 2 (2025): February
Publisher : Postgraduate, University of Mataram

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.29303/jppipa.v11i2.10011

Abstract

One of the companies in Semarang engaged in gadget sales services has an Apple Ecosystem information system for selling products from an exclusive brand, Apple. Inside there are sales transactions and also service devices iPad, Macbook Air, Macbook Pro, AirPods, Mac, and Apple Accsessories. This research uses purchase transaction data from Apple Ecosystem customers for the period 2023. The use of RFM (Recency, Frequency, Monetary) analysis helps in determining the attributes used for customer segmentation. To determine the optimal number of clusters from the RFM dataset, the Elbow method is applied. The dataset generated from RFM is grouped using the K-Means algorithm, the quality of the algorithm will be compared in cluster formation using the Silhouette Coefficient method. All procedures will be loaded into the Customer Segmentation App (RFM Clustering) web application. Customer segmentation from RFM datasets that have been clustered produces 3 optimal clusters, namely Cluster 2 is High Spenders with 326 customers, Cluster 0 is VIP Customers, Cluster 1 is Frequent Buyers. Cluster validation of k-means using the silhouette coefficient produces a value of 0.3524.
Hybrid heuristic model and Fuzzy C-Means for stock forecasting using Type 2 Fuzzy Time Series Satriani, Rineka Brylian Akbar; Farikhin, Farikhin; Surarso, Bayu
Interdisciplinary Social Studies Vol. 4 No. 1 (2024): Regular Issue: October-December 2024
Publisher : International Journal Labs

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.55324/iss.v4i1.742

Abstract

Forecasting is important in investment because of the inconsistent stock price pattern that requires in-depth analysis. This study proposes using a combination of heuristic and Fuzzy C-Means (FCM) models on Fuzzy Time Series Type 2. This study aims to obtain accurate forecasting results by using more data from the time series. The results show that the proposed model provides accurate forecasting. The FCM model is used to group data into clusters and form intervals. Heuristics also optimizes the performance of Fuzzy Logical Relationships Group (FLRG) by using up and down trends. Type 2 FTS is an extension of  Type 1 that uses union and intersection operators to refine fuzzy relations. The results show that the modification by combining FCM and heuristics in Type 2  FTS for stock forecasting provides excellent results with a MAPE value of 2,87%.
Elementary School Accreditation Assessment Using Fuzzy Tsukamoto and SMARTER Method Rahmawati, Nurhita; Nurhayati, Oky Dwi; Surarso, Bayu
Scientific Journal of Informatics Vol. 12 No. 4: November 2025
Publisher : Universitas Negeri Semarang

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.15294/sji.v12i4.30729

Abstract

Purpose: The primary objective of this study is to develop and validate an Elementary School Accreditation Evaluation Model that is both measurable and fair. The proposed model integrates the Fuzzy Tsukamoto method to calculate and consistently generate the final score of each alternative, and the SMARTER method to produce a prioritized ranking that serves as a practical guide for schools in their efforts to improve and strengthen quality. Methods: This study integrates the Fuzzy Tsukamoto method to process numerical data through a rule-based inference mechanism. Simultaneously, the SMARTER method is employed to systematically assign weights to each criterion and sub-criterion using the Rank Order Centroid (ROC) approach. The evaluation is carried out on 16 alternatives based on four main criteria. The research data are derived from the IASP 2020 instrument issued by BAN-S/M, which serves as the official accreditation standard for schools and madrasahs in Indonesia. Result: The developed structured assessment model proved effective. Through ROC weighting, Criterion K1 was identified as the main determining factor (0.611). System validation using Fuzzy Logic showed a high level of consistency (87.5% agreement) with the manual assessor's decisions, confirming the model's accuracy in replicating assessments based on data triangulation. The SMARTER ranking provides targeted recommendations, placing Alternatives A13, A2, A7, and A8 as standards to be maintained, while pointing to A3 as the priority for immediate improvement. Novelty: This study offers a novel approach by integrating the Fuzzy Tsukamoto and SMARTER methods within the context of primary school accreditation a combination that has been rarely explored in previous research. The proposed model not only generates evaluation scores but also produces a ranking system that can serve as a reference for school evaluation.
Unpacking Customer Churn Management through Information Systems Syibli, Mohammad; Gernowo, Rahmat; Surarso, Bayu; Setiawan, Aldi; Andi Setiabudi, Nur
Jurnal Sisfokom (Sistem Informasi dan Komputer) Vol. 15 No. 01 (2026): JANUARY
Publisher : ISB Atma Luhur

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.32736/sisfokom.v15i01.2552

Abstract

The high churn rate in the telecommunications industry remains a persistent challenge affecting customer retention, revenue stability, and long-term competitiveness. This study aims to identify, classify, and synthesize the current state of customer churn management (CCM) research through a systematic lens. Despite growing interest in churn prediction, most studies focus narrowly on improving model accuracy, overlooking the organizational, strategic, and adaptive aspects essential for effective churn management. To address this gap, this paper conducts a systematic literature review (SLR) covering publications from 2020 to 2025, analyzed through the Input–Process–Output (IPO) framework. Twenty high-impact papers were coded across various industries to examine data inputs, modelling processes, output insights, and feedback or retraining mechanisms. The findings reveal a strong bias toward predictive modelling and performance metrics, with limited attention to adaptive retraining, business impact assessment, and decision-support feedback loops. This imbalance highlights the need for a more holistic approach to churn management that integrates analytical and managerial dimensions. The study concludes by proposing a synthesized reference model and future research agenda to guide the development of adaptive, information-systems-based churn management frameworks for dynamic business environments.
Co-Authors A. Nafis Haikal Adi Wibowo Adi Wibowo Agus Subagio Ahmad Abdul Chamid Ahmad Aviv Mahmudi Aldi Setiawan, Aldi Alfajri, Willy Bima Ali Bardadi Anak Agung Gede Sugianthara Andi Setiabudi, Nur Antariksa, Muhammad Deagama Surya Arief Hidayat Aris Puji Widodo Aris Sugiharto Aslam Fatkhudin Aulia, Lathifatul Badieah Assegaf Bambang Irawanto Beta Noranita Budi Warsito Budi Warsito Budi Warsito Che Pee, Ahmad Naim Dedy Kurniadi Dinar Mutiara Kusumo Nugraheni Dwi Putri Handayani Dwiyanasari, Desty Edwin Setiawan Eko Adi Sarwoko Eko Sediyono Etna Vianita Fajar Nugraha Fra Siskus Dian Arianto Ghufron Ghufron Harjito - Henny Indriyawati Imam Tahyudin Indah Jumawanti Irfan Santiko I’tishom Al Khoiry Jatmiko Endro Suseno Jumawanti, Indah Jumawanti, Indah Juwanda, Farikhin Khoerunnisa, Selvi Fitria Khusnah, Miftakhul Laily Rahmania, Laily Lili Rusdiana, Lili LM Fajar Israwan, LM Fajar Lucia Ratnasari Masruroh, Fitriana Maunah, Uun Migunani Migunani Muhammad Haris Qamaruzzaman Muhammad Nasrullah Muhammad Sam'an Mustafid Mustafid Mustaqim Mustaqim Mustaqim Mustaqim, Mustaqim Nugraheni, Dinar Oky Dwi Nurhayati Pukky Tetralian Bantining Ngastiti Puspita, Yuanita Candra Putri, Aina Latifa Riyana Putri, Nitami Lestari Putut Sriwasito Rachmat Gernowo Ragil Saputra Ragil Saputra Rahmat Gernowo Rahmawati, Nurhita Ratri Wulandari Rezki Kurniati, Rezki Robertus Heri Sulistyo Utomo Saputra, Ragil Satriani, Rineka Brylian Akbar Siti Alfiatur Rohmaniah St. Budi Waluya Sugiyamto Sugiyamto, Sugiyamto Sulastri Daruni Sulistiyo, Budi Suryono Suryono Suryono Suryono Suryono, Suryono Susi Hendartie Susilo Hariyanto sutimin sutimin Sutrisno, Sutrisno Sutrisno, Sutrisno Syibli, Mohammad T Indriastuti . Titi Udjiani SRRM Tri Retnaningsih Soeprobowati Uswatun Khasanah Vianita, Etna Wahyul Amien Syafei Wicaksono, Mahad Zainal Arifin Hasibuan