Claim Missing Document
Check
Articles

Integrating Bayesian Optimization into Ensemble Logistic Regression for Explainable AI-Based Customer Behavior Analysis Jeffry, Jeffry; Azis, Azminuddin I. S.; Kandakon, Elisabeth Tri Juliana
Sinkron : jurnal dan penelitian teknik informatika Vol. 9 No. 4 (2025): Articles Research October 2025
Publisher : Politeknik Ganesha Medan

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.33395/sinkron.v9i4.15219

Abstract

Understanding customer behavior is a strategic factor in business decision-making, particularly within the automotive sector, where competition is intense and product variety is diverse. While previous studies often rely on limited demographic variables, such as age and gender, this research advances the field by integrating ensemble logistic regression with Bayesian Optimization for hyperparameter tuning and SHAP-based interpretability. The proposed model incorporates additional features beyond demographics, including vehicle category, product type, vehicle year, dealer branch, and transaction source, to enhance predictive accuracy. The methodology involves data preprocessing through encoding and cleaning, class balancing using SMOTE combined with undersampling, and stratified train-test splitting (80:20). Baseline Logistic Regression achieved an accuracy of 80%, ROC AUC of 0.89, precision of 0.47/0.96, recall of 0.84/0.79, and F1-scores of 0.59/0.89. By applying ensemble logistic regression with Bayesian Optimization, performance improved to 84% accuracy, ROC AUC of 0.92, precision of 0.51/0.98, recall of 0.83/0.84, and F1-scores of 0.63/0.92. SHAP analysis confirmed that the additional features significantly contribute to prediction outcomes. The novelty of this study lies in combining Ensemble Logistic Regression with Bayesian Optimization and SHAP explainability in the automotive domain, offering not only improved accuracy but also interpretability and fairness for business decision-making, providing actionable insights for targeted marketing strategies and product management. Future studies may incorporate broader behavioral and transactional variables to capture more nuanced customer decision patterns..
A Deep Learning Approach to Respiratory Disease Classification Using Lung Sound Visualization for Telemedicine Applications Wahyudi, Andi Enal; Batau, Radus; Aziz, Firman; Jeffry, Jeffry
Journal of System and Computer Engineering Vol 6 No 4 (2025): JSCE: October 2025
Publisher : Universitas Pancasakti

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.61628/jsce.v6i4.2144

Abstract

This study presents the development of an intelligent system for the classification of respiratory diseases using lung sound visualizations and deep learning. A hybrid Convolutional Neural Network and Bidirectional Long Short-Term Memory (CNN–BiLSTM) model was designed to classify four conditions: asthma, bronchitis, tuberculosis, and normal (healthy). Lung sound recordings were converted into time-frequency representations (e.g., mel-spectrograms), enabling spatial-temporal feature extraction. The system achieved an overall classification accuracy of 99.5%, with F1-scores above 0.93 for all classes. The confusion matrix revealed minimal misclassifications, primarily between asthma and bronchitis. These results suggest that the proposed model can effectively support real-time, non-invasive respiratory screening, particularly in telemedicine environments. Future work includes clinical validation, integration of patient metadata, and adoption of transformer-based models to further enhance diagnostic performance.
Enhancing Human Activity Recognition with Attention-Based Stacked Sparse Autoencoders Batau, Radus; Sari, Sri Kurniyan; Aziz, Firman; Jeffry, Jeffry
Journal of System and Computer Engineering Vol 6 No 4 (2025): JSCE: October 2025
Publisher : Universitas Pancasakti

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.61628/jsce.v6i4.2148

Abstract

This study presents the development of an intelligent system for the classification of respiratory diseases using lung sound visualizations and deep learning. A hybrid Convolutional Neural Network and Bidirectional Long Short-Term Memory (CNN–BiLSTM) model was designed to classify four conditions: asthma, bronchitis, tuberculosis, and normal (healthy). Lung sound recordings were converted into time-frequency representations (e.g., mel-spectrograms), enabling spatial-temporal feature extraction. The system achieved an overall classification accuracy of 99.5%, with F1-scores above 0.93 for all classes. The confusion matrix revealed minimal misclassifications, primarily between asthma and bronchitis. These results suggest that the proposed model can effectively support real-time, non-invasive respiratory screening, particularly in telemedicine environments. Future work includes clinical validation, integration of patient metadata, and adoption of transformer-based models to further enhance diagnostic performance.
Implementing Leadership Style to Improve Organizational Performance Mediated by Organizational Identification Sentoso, Antony; Jeffry, Jeffry; Mon, Muhammad Donal
Almana : Jurnal Manajemen dan Bisnis Vol 8 No 3 (2024): December
Publisher : Bandung: Prodi Manajemen FE Universitas Langlangbuana

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.36555/almana.v8i3.2599

Abstract

Before the progress of the times and technology that was increasingly developing and rapidly as well as intense business competition, experts highlighted the important role of performance in organizations to improve the sustainability of life based on the promotion of leadership styles. This research aims to provide additional information on companies operating in the distributor sector in implementing leadership styles to improve organizational performance. This research uses quantitative methods. The population in this study was 261 workers in the distributor industry in Batam City. All research has met the criteria for validity and reliability. The results of the research state that Innovative leadership significantly affects both organizational performance and organizational identification, but not when organizational identification acts as a mediating factor. Second, organizational performance and identification are not significantly impacted by transformational leadership. The same holds if organizational identity acts as a mediator. It was discovered that organizational identity significantly affects organizational performance and that strategic leadership styles do not significantly affect organizational performance. Thus, it can be said that while leadership style has a significant impact on organizational performance, not all leadership philosophies are appropriate for usage in a given setting.
Sistem Manajemen Penjadwalan Pengajaran Dosen berbasis SMS Gateway jeffry, jeffry; Velayaty, Ali Akbar; Aziz, Firman
Journal of System and Computer Engineering Vol 4 No 2 (2023): JSCE: Juli 2023
Publisher : Universitas Pancasakti

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.61628/jsce.v4i2.648

Abstract

To improve performance in teaching time, one of which is by being punctual in teaching, therefore a system is needed to remind lecturers when teaching time arrives. Along with the development of technology, almost everyone has a communication device called a cell phone, one of the functions that are often used is sending messages or SMS. SMS Gateway is a platform that can be used to send and receive SMS whose settings can be made using PHP with data storage tools in the form of MySQL. Reminder SMS and teaching schedule monitoring using the SMS Gateway is a system used to remind lecturers about class schedules via SMS that was developed using the PHP programming language.
Perancangan Sistem Monitoring Kualitas Udara Ruangan Berbasis Internet of Things (IoT) Iskandar, Imran; Rimalia, Watty; jeffry, Jeffry; Panggabean, Benny Leonard Enrico
Journal of System and Computer Engineering Vol 5 No 1 (2024): JSCE: Januari 2024
Publisher : Universitas Pancasakti

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

Abstract

The purpose of the research is to monitor air quality in the room by using MQ2, Microcontroller NodeMCU equipped with ESP8266 wireless module by using MQTT protocol for Data Communication sensor node to server. System testing uses a black box testing approach.The type of research used in this study is experimental research with the approach of PPDIOO methodology. This research is done by conducting trials where mechanical and electronic design for hardware components designed to build tools using this sensor can work according to the objectives and target desired.The results of this research show that when the sensor detects the existence 150 of smoke, gas, carbon monoxide (CO) then the value of the ADC sensor will give a warning notification in the form of an alarm generated from the tone/tone of the alarm will be active when the sensor is actively reading the value of ADC with a tolerance above 150 ppm
PENGARUH KETERLIBATAN KERJA DAN KECERDASAN EMOSIONAL TERHADAP KINERJA MELALUI KEPUASAN KERJA PEGAWAI PADA PT PLN (PERSERO) UNIT INDUK PEMANGUNAN SUMBAGUT Jeffry, Jeffry; Handayani, Susi
Jurnal Ilmiah Manajemen, Ekonomi, & Akuntansi (MEA) Vol 8 No 1 (2024): Edisi Januari - April 2024
Publisher : LPPM STIE Muhammadiah Bandung

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.31955/mea.v8i1.3863

Abstract

Fokus utama penelitian ini adalah untuk mengetahui korelasi antara kepuasan kerja, kecerdasan emosional, dan keterlibatan kerja, serta bagaimana faktor-faktor tersebut secara kolektif berdampak terhadap kinerja karyawan di Unit Pengembangan Utama PT PLN (Persero) Sumbagut. Penelitian ini menggunakan metodologi kuantitatif dengan melakukan survEQ skala Likert kepada karyawa. Data yang terkumpul kemudian dianalisis menggunakan teknik PLS-SEM di SmartPLS. Prosedur ini meliputi pelaksanaan pengujian hipotesis, analisis inner dan outer model. Berdasarkan temuan penelitian ini, terlihat adanya hubungan yang signifikan antara keterlibatan kerja, kecerdasan emosional, kepuasan kerja, dan kinerja karyawan. Selain itu, penelitian ini menekankan pengaruh tidak langsung keterlibatan kerja dan kecerdasan emosional terhadap kinerja karyawan yang dimediasi oleh kepuasan kerja. Analisis ini menekankan pentingnya keterlibatan kerja, kecerdasan emosional, kepuasan kerja di lingkungan Unit Pengembangan Utama PT PLN (Persero) Sumbagut.
ANALISIS FAKTOR KUALITAS PELAYANAN ASURANSI KESEHATAN MENGGUNAKAN METODE AHP jeffry, jeffry
Advances in Computer System Innovation Journal Vol. 2 No. 1: April 2024, ACSI Journal
Publisher : Unit Publikasi Ilmiah Perkumpulan Intelektual Madani Indonesia

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.51577/acsijournal.v2i1.507

Abstract

Sistem Pendukung Keputusan (SPK) dikembangkan seiring dengan perkembangan teknologi informasi sebagai sarana bantu dalam pengambilan keputusan yang cepat dan efisien. Penelitian ini bertujuan mengimplementasikan SPK dalam analisis kualitas pelayanan asuransi kesehatan dengan menggunakan metode Analytical Hierarchy Process (AHP). SPK yang dibangun berbasis web, memanfaatkan Unified Modeling Language (UML) untuk pemodelan sistem. Hasil dari analisis ini menunjukkan faktor-faktor yang mempengaruhi kepuasan pelanggan dengan nilai kepuasan terhadap obat sebesar 0.22, faktor sosialisasi sebesar 0.22, faktor ketersediaan tenaga medis sebesar 0.20, faktor fasilitas sebesar 0.19, dan faktor pelayanan sebesar 0.18. SPK yang dihasilkan diharapkan dapat membantu perusahaan asuransi kesehatan dalam meningkatkan kualitas pelayanan berdasarkan prioritas yang telah diidentifikasi
APLIKASI HOUSEKEEPING HOTEL BERBASIS WEB PADA NOVOTEL MAKASSAR GRAND SHAYLA MENGGUNAKAN METODE WATERFALL Putra Yuzi Bachmid, Fadel Muhammad; Usman, Syahrul; Syam, Rahmat Fuadi; Jeffry, Jeffry
Advances in Computer System Innovation Journal Vol. 2 No. 1: April 2024, ACSI Journal
Publisher : Unit Publikasi Ilmiah Perkumpulan Intelektual Madani Indonesia

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.51577/acsijournal.v2i1.516

Abstract

Industri perhotelan merupakan salah satu industri yang berkembang pesat dalam dunia bisnis saat ini, maka oleh karena itu suatu hotel yang ingin bersaing ketat dengan hotel lainnya harus memiliki fasilitas, kemudahan, dan struktur manajemen yang lengkap. Namun menurut hasil wawancara dan observasi di Hotel Novotel Makassar Grand Shayla, mereka masih melakukan pengoperasian status kamar secara manual terutama operasional pembersihan kamar dengan informasi yang terdapat celah, sehingga dalam berkomunikasi pertukaran informasi selalu menggunakan WhatsApp. Penulis berinisiatif melakukan penelitian serta penerapan uji coba dengan membuat aplikasi berbasis web yang diharapkan dapat memberikan pengalaman kepada bagian housekeeping, serta supervisor housekeeping dalam melakukan pemantauan room status untuk setiap kamar.Aplikasi ini berbasis web application juga untuk memantau keadaan room status. Aplikasi web ini juga memiliki beberapa fitur diantaranya adalah dapat mengubah status kamar ke dalam beberapa status sesuai keadaan status kamar secara aktual, diharapkan dapat mengubah status pada pukul 2 pagi dengan metode sesuai dengan operasional yang telah berjalan metode penelitian dengan metode waterfall, karena metode penelitian paling tepat digunakan dengan memanfaatkan observasi dan wawancara.
Model Convolutional Neural Network yang Efektif dan Efisien untuk Segmentasi Semantik Awan Cumulonimbus Azminuddin I. S. Azis; jeffry, jeffry; Firman Aziz; Andi Taufiqurrahman Akbar
Advances in Computer System Innovation Journal Vol. 3 No. 1: April 2025, ACSI Journal
Publisher : Unit Publikasi Ilmiah Perkumpulan Intelektual Madani Indonesia

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.51577/acsijournal.v3i1.807

Abstract

Awan Cumulonimbus (CB) merupakan jenis awan yang dapat mengakibatkan petir, badai, tornado, hujan lebat, turbulensi penerbangan, dan cuaca ekstrim lainnya. Oleh karenanya, prediksi/deteksi keberadaan awan CB yang akurat dan real time akan mendukung kelancaran dan keselamatan banyak aktivitas manusia. Citra infrared (IR) pada satelit Himawari-8 di Badan Meteorologi, Klimatologi, dan Geofisika (BMKG) memiliki informasi mengenai pertumbuhan awan CB. Berbagai studi terkait telah membuktikan bahwa metode yang paling populer dan handal dalam bidang computer vision pada objek awan adalah Convolutional Neural Network (CNN). Untuk menemukan/memperoleh model CNN yang paling efektif dan efisien dalam menangani segmentasi semantik awan CB pada citra IR Himawari-8, maka berbagai pendekatan untuk CNN diuji coba, diantaranya arsitektur jaringan untuk CNN, optimalisasi pelatihan CNN berbasis Gradient Discent Optimizer (GDO), Weighted Class (WC) untuk mereduksi masalah imbalanced class, dan Data Augmentation (DA) untuk memperkaya keragaman data dan mencegah overfitting. Hasil studi menunjukkan bahwa model CNN yang paling efektif adalah dengan arsitektur jaringan U-NET, GDO menggunakan Adaptive Moment Estimation (Adam), dan WC dengan 99,56% global akurasi pengujian, 97,12% rata-rata akurasi pengujian, 94,42% rata-rata IoU, 94,48% akurasi prediksi pada class CB, 99,60% akurasi validasi, 99,61% akurasi pelatihan, 0,1071 loss validasi, dan 0.1072 loss pelatihan. Sedangkan model CNN yang paling efisien adalah dengan arsitektur jaringan Dilated, GDO menggunakan Root Mean Square Propagation (RMSProp), dan WC dengan 24 detik waktu proses/pemodelan, lebih cepat 20 detik namun dengan efektivitas yang tidak jauh berbeda daripada model CNN yang paling efektif.