Article Per Year (5 Year)
p-Index From 2020 - 2025
2.949
P-Index
This Author published in this journals
All Journal Jurnal Simetris Jurnal Teknologi dan Sistem Komputer JURNAL ENGGANO Jurnal DISPROTEK JURTEKSI Abdimas Universal Jurnal Teknologi Dan Sistem Informasi Bisnis SKANIKA: Sistem Komputer dan Teknik Informatika Bulletin of Computer Science Research Journal of Dinda : Data Science, Information Technology, and Data Analytics As-Sidanah : Jurnal Pengabdian Masyarakat Jurnal Minfo Polgan (JMP) Jurnal Rekognisi Akuntansi (JRA) J-ABDIPAMAS (Jurnal Pengabdian Kepada Masyarakat)
Mulyo, Harminto
Universitas Islam Nahdlatul Ulama Jepara
Religion Humanities Civil Engineering, Building, Construction & Architecture Computer Science & IT Control & Systems Engineering Decision Sciences, Operations Research & Management Economics, Econometrics & Finance Education Electrical & Electronics Engineering Engineering Industrial & Manufacturing Engineering Languange, Linguistic, Communication & Media Law, Crime, Criminology & Criminal Justice Library & Information Science Mathematics Social Sciences Other

Published : 20 Documents Claim Missing Document
Claim Missing Document
Check
Articles

Found 20 Documents
Search

 1   2 
Implementasi Algoritma K-Nearest Neighbor untuk Klasifikasi Jurusan pada Peserta Didik Baru Widiastuti, Nur Aeni; Azhar, Maulana; Mulyo, Harminto
Simetris: Jurnal Teknik Mesin, Elektro dan Ilmu Komputer Vol 14, No 2 (2023): JURNAL SIMETRIS VOLUME 14 NO 2 TAHUN 2023
Publisher : Fakultas Teknik Universitas Muria Kudus

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.24176/simet.v14i2.10092

Abstract

Majoring students is a process of placing students into certain majors in accordance with their interests and academic abilities in an effort to make it easier for students in the learning process. Madrasah Aliyah Darul Hikmah Menganti is a school equivalent to SMA, which has two majors, namely science and social studies. The difficulty of classifying the majors of new students is an obstacle for the school. Because the criteria assessment process is carried out one by one. From these problems, the K-Nearest Neighbor (K-NN) method was applied to classify majors in order to simplify and minimize errors in the process of determining new student majors. The data initially amounted to 638 records and 31 attributes. After preprocessing, the data used amounted to 635 records with 12 attributes, namely name, gender, major interest, school origin, children to, number of siblings, math scores, English grades, science grades, Indonesian language scores, test scores, and major recommendations. After testing using K-Fold Cross Validation and Confusion Matrix for evaluation and validation of results by calculating the Euclidean Distance distance, the best k value (optimal) k=3 which produces accuracy: 97.11%, precision: 96.82%, recall: 98.33%, and AUC: 0.951.
Short-Term Cryptocurrency Price Prediction Using Bi-LSTM Method with Interactive Web Andriansyach, Dimas Jordy; Sarwido, Sarwido; Mulyo, Harminto
Jurnal Teknologi Dan Sistem Informasi Bisnis Vol 6 No 4 (2024): Oktober 2024
Publisher : Prodi Sistem Informasi Universitas Dharma Andalas

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.47233/jteksis.v6i4.1645

Abstract

Short-term Bitcoin price prediction is a crucial aspect of transaction decision-making, especially for investors. In this study, a Bidirectional Long Short-Term Memory (Bi-LSTM) model was developed for short-term Bitcoin price prediction. The Bidirectional LSTM is designed to capture temporal context in both directions, allowing the model to process information from past and future time steps simultaneously. The model was validated using real-world data, including Bitcoin stock price datasets. The results show that the model achieved high accuracy, with a Root Mean Square Error (RMSE) of 56.90 on the training data and 157.35 on the test data, along with a Mean Absolute Error (MAE) of 366.40 and 486.63, respectively. The Bidirectional Least Square Memory model accurately predicted Bitcoin prices over a specific time period. This application integrates the model into a web application, enabling users to obtain real-time Bitcoin price predictions through a user-friendly interface.
Metode Single Linkage pada Agglomerative Hierarchical Clustering dalam Penentuan Tingkat Kepuasan Mahasiswa terhadap Layanan Akademik Maori, Nadia Annisa; Mulyo, Harminto
Simetris: Jurnal Teknik Mesin, Elektro dan Ilmu Komputer Vol 15, No 2 (2024): JURNAL SIMETRIS VOLUME 15 NO 2 TAHUN 2024
Publisher : Fakultas Teknik Universitas Muria Kudus

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.24176/simet.v15i2.12788

Abstract

Dalam era pendidikan modern, penting bagi institusi akademik untuk memahami dan meningkatkan kepuasan mahasiswa terhadap layanan akademik yang diberikan. Tingkat kepuasan ini berperan penting dalam menilai kualitas pendidikan, pengalaman belajar, serta reputasi dan daya saing institusi. Dengan kemajuan teknologi dan peningkatan pengumpulan data, analisis data menjadi krusial untuk memahami persepsi dan kebutuhan mahasiswa. Penelitian ini menggunakan teknik Agglomerative Hierarchical Clustering (AHC) dengan metode Single Linkage untuk mengelompokkan data survei kepuasan mahasiswa terhadap layanan akademik di Program Studi Teknik Informatika UNISNU Jepara. Metode ini dipilih karena tidak memerlukan penentuan jumlah klaster sebelumnya dan cocok untuk data dengan struktur yang tidak teratur.Hasil penelitian menunjukkan bahwa AHC dengan Single Linkage efektif dalam mengidentifikasi dua kelompok mahasiswa berdasarkan tingkat kepuasan mereka, yaitu puas dan tidak puas. Evaluasi menggunakan Silhouette Coefficient menunjukkan nilai tertinggi 0.80 untuk dua klaster mengindikasikan bahwa pengelompokan ini cukup baik. Visualisasi dendrogram memberikan wawasan tambahan tentang struktur klaster dan hubungan antar data. Penelitian ini memberikan kontribusi penting untuk pemahaman tentang kepuasan mahasiswa dan dasar untuk pengembangan strategi peningkatan kualitas layanan akademik di masa depan. Metode AHC dengan pendekatan Single Linkage terbukti efisien dalam mengelompokkan data berdasarkan jarak terdekat antar objek dalam klaster, meskipun sensitif terhadap outlier dan efek chaining.
OPTIMALISASI FP-GROWTH DENGAN TEKNIK PRUNING UNTUK ANALISIS POLA PEMINJAMAN BUKU UPT PERPUSTAKAAN UNISNU JEPARA Manzis, Akhmad Hossam; Kusumodestoni, R. Hadapiningradja; Mulyo, Harminto
SKANIKA: Sistem Komputer dan Teknik Informatika Vol 8 No 1 (2025): Jurnal SKANIKA Januari 2025
Publisher : Universitas Budi Luhur

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.36080/skanika.v8i1.3333

Abstract

In the digital age, libraries continue to adapt to provide access to information and knowledge, support education and research, and preserve cultural heritage. The increasing demand for reading materials along with the development of information technology and digital literacy has led to a surge in the amount of data stored on various information provision platforms including libraries. With the sheer volume of data, the challenges of management, analysis and storage have become increasingly complex. Lack of understanding of readers' preferences, inefficiency in book procurement, and difficulty in determining book layout are problems that arise in library management. Therefore, analyzing circulation data is very important, for example using FP- Growth to find patterns of book borrowing. Items that do not meet the criteria but are still included in the calculation process cause the results to be less significant, but pruning which removes items with low frequency of occurrence can improve the accuracy of the analysis. The results of the FP- Growth calculation reveal a relationship between management and economics books with a support of 27%, Confidence 54%, and Lift 956 which means that the two books have a large influence on each other's occurrence. While pruning the number of rules generated is getting smaller, from 26 to 8, but the rules have a strong relationship.
Optimalisasi Algoritma Naive Bayes Dengan Teknik Ensemble Dalam Analisis Sentimen Twitter Pantai Kartini Jepara Muhammad Arqom Anwar; Harminto Mulyo; Teguh Tamrin
Jurnal Minfo Polgan Vol. 13 No. 2 (2024): Artikel Penelitian
Publisher : Politeknik Ganesha Medan

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.33395/jmp.v13i2.14014

Abstract

Penelitian ini memanfaatkan Twitter untuk menganalisis opini publik tentang Pantai Kartini Jepara, dengan fokus pada optimisasi algoritma Naive Bayes dalam analisis sentimen. Penelitian ini mengidentifikasi bahwa akurasi Naive Bayes terbatas dalam menangani data besar dan kompleks. Tujuan utamanya adalah meningkatkan akurasi dan efisiensi analisis sentimen melalui optimisasi parameter dan teknik ensemble. Metode penelitian melibatkan pengumpulan data Twitter dari 2010–2023, preprocessing data, pelatihan model Naive Bayes, SVM, dan ensemble, serta evaluasi performa menggunakan akurasi, presisi, recall, dan F1-score. Model ensemble yang menggabungkan Naive Bayes dan SVM mencapai akurasi tertinggi sebesar 88,81%, meningkat dari 83,91% pada Naive Bayes dasar dan 86,01% pada SVM, menunjukkan perbaikan signifikan dalam analisis sentimen. Kombinasi algoritma Naive Bayes dengan teknik optimasi dan ensemble meningkatkan akurasi analisis sentimen. Penelitian selanjutnya disarankan untuk mengeksplorasi penerapan model ini pada data yang lebih besar atau platform media sosial lain.
Pengaruh Hyperparameter Tuning Gradient Boosting Terhadap Prediksi Pemilihan Program Studi Mahasiswa Baru Harminto Mulyo; Akhmad Khanif Zyen
Bulletin of Computer Science Research Vol. 5 No. 2 (2025): February 2025
Publisher : Forum Kerjasama Pendidikan Tinggi (FKPT)

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.47065/bulletincsr.v5i2.454

Abstract

This study aims to improve the accuracy of predicting new student major selection using the Gradient Boosting algorithm optimized through hyperparameter tuning. Gradient Boosting was chosen for its ability to handle complex and diverse data, which is crucial in the context of major prediction. The data used was sourced from the new student admissions database of Universitas Islam Nahdlatul Ulama Jepara for the 2013–2023 period, with preprocessing including data cleaning, imputation of missing values, and transformation of categorical features. The initial accuracy of the Gradient Boosting model with default configuration reached 99.01%, indicating that the dataset had relatively clear and structured patterns, enabling the baseline model to perform highly. However, to ensure generalization and avoid the risk of overfitting, hyperparameter tuning was performed using Randomized Search CV. The tuning results showed an increase in accuracy to 99.84% with optimal configurations including a learning rate of 0.1, 300 estimators, and a maximum tree depth of 4. Feature analysis also revealed that attributes such as "school_type," "school_origin," and "gender" significantly influenced the prediction outcomes. This study demonstrates that hyperparameter tuning can significantly enhance model performance, providing a more accurate and relevant predictive solution for the major selection process. Nevertheless, the study's limitation lies in the scope of the dataset, which originated from a single institution, suggesting the need for further exploration with more diverse data and advanced tuning methods like Bayesian Optimization. These findings provide valuable contributions to educational institutions in developing data-driven systems to support strategic decision-making.
Implementasi Algoritma K-Means untuk Klasterisasi Data Hasil Tangkapan Ikan di Karimunjawa Rivaldo, Muchammad Dwi; Wibowo, Gentur Wahyu Nyipto; Mulyo, Harminto
Jurnal Minfo Polgan Vol. 13 No. 1 (2024): Artikel Penelitian
Publisher : Politeknik Ganesha Medan

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.33395/jmp.v13i1.13928

Abstract

Penelitian ini berfokus pada implementasi algoritma K-Means untuk klasterisasi data hasil tangkapan ikan di Karimunjawa. Tujuan utama adalah untuk mengidentifikasi jenis ikan yang paling banyak ditangkap dan area penangkapan yang paling produktif. Data yang digunakan mencakup tanggal, nama nelayan, jumlah ikan (kg), jenis ikan, area penangkapan ikan, dan metode penangkapan ikan, yang dikumpulkan selama tahun 2020. Analisis klasterisasi menghasilkan tiga klaster utama: Klaster 0 dengan total 315,9 kg terdiri dari Cumi, Kakak Tua, Jinahak, Baronang, Panti, dan Tambak Jeron; Klaster 1 dengan total 856,9 kg terdiri dari Teri, Tengiri, Tambak Jeron, Udul, Panti, dan Pari; Klaster 2 dengan total 1383,2 kg terdiri dari Todak. Selain itu, area penangkapan yang produktif juga diklasterisasi menjadi tiga: Klaster 0 mencakup Karimunjawa Timur, Klaster 1 mencakup Karimunjawa Barat, dan Klaster 2 mencakup Karimunjawa Utara. Hasil evaluasi menggunakan metrik pengukuran menunjukkan bahwa Silhouette Score positif sebesar 0,48 mengindikasikan bahwa klaster yang dihasilkan cenderung terpisah dengan baik, meskipun masih ada ruang untuk perbaikan. Davies-Bouldin Index yang rendah sebesar 0,83 menunjukkan bahwa klaster yang dihasilkan cukup terpisah satu sama lain, meskipun tidak sempurna. Metode Elbow memberikan indikasi jumlah klaster optimal, membantu dalam pemilihan konfigurasi yang tepat untuk analisis klaster. Penelitian ini memberikan wawasan berharga tentang distribusi tangkapan ikan di Karimunjawa, yang dapat digunakan untuk meningkatkan strategi penangkapan dan manajemen perikanan. Implementasi dan analisis dilakukan menggunakan Google Colab dengan bahasa pemrograman Python.
Implementation of Random Forest Algorithm with RFE and SMOTE on Cardiotocography Dataset Nur Taqwimi, Muhammad Ahsani; Wahono, Buang Budi; Mulyo, Harminto
Indonesian Journal of Data Science, IoT, Machine Learning and Informatics Vol 5 No 2 (2025): August
Publisher : Research Group of Data Engineering, Faculty of Informatics

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.20895/dinda.v5i2.1818

Abstract

Having a healthy baby is a dream for mothers. However, the high rate of maternal and fetal mortality is still a serious problem, so more accurate fetal health monitoring is needed to prevent pregnancy complications. One of the devices used is Cardiotocography (CTG), which produces data on fetal conditions. The CTG dataset used in this study faces challenges in the form of class imbalance and a high number of features, which can reduce classification performance. This study aims to overcome these challenges by implementing the Random Forest algorithm combined with the Synthetic Minority Oversampling Technique (SMOTE) technique for class balancing and Recursive Feature Elimination (RFE) for feature selection. The dataset used is "Fetal Health Classification" from the Kaggle platform, which consists of 2,126 data with three classes: Normal, Suspect, and Pathological. The test results show that the RFE method is able to reduce the number of features from 22 to 18, while SMOTE increases the proportion of minority data. The model built produces good classification performance with an accuracy value of 95%, precision 93%, recall 89%, and F1-score 91%. The ROC-AUC value for the Normal class is 0.9881, Suspect 0.9789, and Pathological 0.9985. Although the model is able to predict the Normal and Pathological classes with high accuracy, the performance on the Suspect class still needs to be improved. Overall, the integration of Random Forest with SMOTE and RFE has proven effective in improving the accuracy of fetal health classification.
Heart Failure Classification Using a Hybrid Model Based on SVM and Random Forest Abdilllah, Muh Sajid; Mulyo, Harminto; Wibowo, Gentur Wahyu Nyipto
Indonesian Journal of Data Science, IoT, Machine Learning and Informatics Vol 5 No 2 (2025): August
Publisher : Research Group of Data Engineering, Faculty of Informatics

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.20895/dinda.v5i2.2001

Abstract

This study discusses the development of a model to classify heart failure disease by combining two algorithms in the field of data mining: Support Vector Machine (SVM) and Random Forest (RF). The dataset used is the Heart Failure Prediction Dataset, consisting of 918 patient records containing medical information such as blood pressure, cholesterol levels, and heart rate. The research process began with data cleaning, normalization using MinMaxScaler, and data balancing with the SMOTE technique to equalize the number of cases between heart failure patients and non-patients. The data was then split into training and testing sets. Each model (SVM and RF) was tested individually and also combined into a hybrid model. Validation was performed using 5-Fold Cross Validation to ensure consistent results. The results show that SVM performed better in terms of precision for detecting heart failure after applying SMOTE, while RF remained stable even without data balancing. The hybrid model combining both algorithms achieved the best performance, with an accuracy of 91.20%, precision of 90.85%, recall of 92.44%, and an AUC score of 0.961. These results indicate that the hybrid model can detect heart failure more accurately and in a more balanced manner. With its high and consistent performance, this model is suitable for use as a decision support system in the medical field, particularly for early detection of heart failure.
OPTIMIZATION OF SUPPORT VECTOR MACHINE WITH SMOTE AND BAYESIAN METHOD FOR HEART FAILURE CLASSIFICATION Doni Agung Prasetyo; Harminto Mulyo; Nadia Annisa Maori
JURTEKSI (jurnal Teknologi dan Sistem Informasi) Vol. 11 No. 3 (2025): Juni 2025
Publisher : Lembaga Penelitian dan Pengabdian Kepada Masyarakat (LPPM) STMIK Royal Kisaran

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.33330/jurteksi.v11i3.4057

Abstract

Abstract: This study applies an integrated approach to optimize heart failure classification. The main objective is to address the challenge of class imbalance in medical datasets and to improve the accuracy, sensitivity, and generalization of the classification model. The urgency of this issue is emphasized by statistics showing that cardiovascular diseases cause approximately 17.9 million deaths worldwide each year. Using a quantitative experimental approach, this study analyzes the "Heart Failure Prediction Dataset" from Kaggle, which consists of 918 records. The data were processed through normalization and encoding, followed by the application of SMOTE on the training set to balance class distribution. This step successfully increased model accuracy from 88.41% to 90.22% and minority class recall from 0.82 to 0.88. Furthermore, Bayesian Optimization was employed to refine the hyperparameters of SVM, resulting in a final model with an accuracy of 89.13% that demonstrated better generalization. This integrated approach significantly enhances the stability, sensitivity, and generalization of the model, making it a reliable tool for clinical decision support systems in predicting heart failure. Keywords: bayesian optimization; heart failure; machine learning; SMOTE; SVM. Abstrak: Penelitian ini menerapkan pendekatan terintegrasi untuk mengoptimalkan klasifikasi gagal jantung. Tujuan utama studi ini adalah untuk mengatasi tantangan ketidakseimbangan kelas dalam dataset medis dan meningkatkan akurasi, sensitivitas, serta generalisasi model klasifikasi. Urgensi ini ditegaskan oleh statistik yang menunjukkan bahwa penyakit kardiovaskular menyebabkan sekitar 17,9 juta kematian setiap tahun secara global. Menggunakan pendekatan eksperimental kuantitatif, penelitian ini menganalisis "Heart Failure Prediction Dataset" dari Kaggle, yang terdiri dari 918 catatan. Data diproses dengan normalisasi dan encoding, lalu SMOTE diterapkan pada data pelatihan untuk menyeimbangkan distribusi kelas. Langkah ini berhasil meningkatkan akurasi dari 88,41% menjadi 90,22% dan recall kelas minoritas dari 0,82 menjadi 0,88. Selanjutnya, Bayesian Optimization menyempurnakan hyperparameter SVM, menghasilkan model akhir dengan akurasi 89,13% yang menunjukkan generalisasi lebih baik. Pendekatan terintegrasi ini secara signifikan meningkatkan stabilitas, sensitivitas, dan generalisasi model. Hasil penelitian ini menjadikannya alat yang andal untuk sistem pendukung keputusan klinis dalam prediksi gagal jantung. Kata kunci: bayesian optimization; gagal jantung; machine learning; SMOTE; SVM
Co-Authors Abdilllah, Muh Sajid Akhmad Khanif Zyen Ali Sofwan An-Naas Shahifatun Na’iema Andriansyach, Dimas Jordy Anita Ameliya Arif Mustofa Aufa Maulana Hidayat Azhar, Maulana Buang Budi Wahono, Buang Budi Doni Agung Prasetyo Dwi Noviyani Fira Nadliratul Afrida Gunawan Mohammad Hayu Dian Yulistianti Irma Nur Af’idah Kartika Widiastuti Khalimatus Sa'diyah Manzis, Akhmad Hossam Maori, Nadia Annisa Mochamad Handre Fidiyanto Muhammad Arqom Anwar Muhammad Misbahul Munir Muhammad Verdy Fahrudin Nadia Annisa Maori Nadia Annisa Maori Nur Aeni Widiastuti Nur Aeni Widiastuti Nur Taqwimi, Muhammad Ahsani Nusrotus Sa’idah Octavianti Paramitha R Hadapiningradja Kusumodestoni Rahmawan Arniyanto Rivaldo, Muchammad Dwi Rohmatun Nikmah Santi Andriyani Sarwido Sarwido, Sarwido Teguh Tamrin Wibowo, Gentur Wahyu Nyipto Zubaedah Zubaedah