Article Per Year (5 Year)
p-Index From 2021 - 2026
1.686
P-Index
This Author published in this journals
All Journal TEKNIK INFORMATIKA Jurnal RESTI (Rekayasa Sistem dan Teknologi Informasi) Jurnal Eksplora Informatika JURNAL ILMIAH INFORMATIKA KOMIK (Konferensi Nasional Teknologi Informasi dan Komputer) Jurnal FASILKOM (teknologi inFormASi dan ILmu KOMputer) Jurnal Pengabdian Masyarakat Sains dan Teknologi Journal of Computing and Informatics Research J-Icon : Jurnal Komputer dan Informatika Jurnal Manajemen, Pendidikan dan Ilmu Komputer (JMENDIKKOM)
Dedy Hartama
Unknown Affiliation
Agriculture, Biological Sciences & Forestry Computer Science & IT Control & Systems Engineering Decision Sciences, Operations Research & Management Education Electrical & Electronics Engineering Engineering Social Sciences Other

Published : 15 Documents Claim Missing Document
Claim Missing Document
Check
Articles

Found 15 Documents
Search

 1   2 
Sentiment Classification in Imbalanced Data: Trade-Offs Between Metrics and Real-World Relevance Indra Swanto Ritonga; Wanayumini; Dedy Hartama
JURNAL TEKNIK INFORMATIKA Vol. 18 No. 2: JURNAL TEKNIK INFORMATIKA
Publisher : Department of Informatics, Universitas Islam Negeri Syarif Hidayatullah

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.15408/jti.v18i2.46652

Abstract

Sentiment analysis plays a crucial role in assessing public perception, particularly in healthcare services like BPJS Kesehatan, Indonesia’s national health insurance program. However, sentiment classification faces a challenge due to class imbalance, where negative feedback dominates positive responses. This study investigates whether sentiment classification should prioritize traditional evaluation or maintain real-world data representation by preserving the original sentiment distribution. Two feature extraction methods, Term Frequency-Inverse Document Frequency (TF-IDF) and Bag of Words (BoW), were evaluated using Naïve Bayes, Support Vector Machine (SVM), and Logistic Regression with varying maximum feature counts (100–300) to examine the impact of feature dimensionality. Model performance was evaluated using traditional metrics, while sentiment distribution fidelity was assessed by comparing predicted proportions with the dataset. Results show TF-IDF achieves higher precision and recall but fails to capture positive sentiments, leading to a skewed representation of real-world trends, while BoW offers a more balanced distribution with slightly lower accuracy. Paired t-tests and Wilcoxon signed-rank tests confirmed differences in accuracy and recall are significant, but not in precision and sentiment distribution. These findings highlight a trade-off between performance and sentiment diversity, vital in healthcare services and other fields with imbalanced datasets, emphasizing the need to align evaluation metrics with real-world objectives. Future research should investigate advanced models, such as deep learning and transformer-based approaches, to enhance both accuracy and fairness when analyzing imbalanced data.
Analisis Model Backpropagation Dalam Meramalkan Tingkat Penjualan Saldo “Link Aja” Dwi Findi Auliasari; Gita Febrianti; Agus Perdana Windarto; Dedy Hartama
Journal of Computing and Informatics Research Vol 2 No 1 (2022): November 2022
Publisher : Forum Kerjasama Pendidikan Tinggi (FKPT)

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.47065/comforch.v2i1.382

Abstract

Analysis of a prediction (forecasting) is very important in a study, so that research becomes more precise and directed (Wanto and Windarto, 2017). As is the case in predicting the level of Link Aja's balance sales. This research is expected to be useful for an agency as one of the study materials in business development. A system to predict the level of sales of Link Aja balance at PT. Wahana Putra Yudha. Artificial Neural Network is a method that is able to perform a mathematical process to predict the level of sales of Link Aja Balance at PT. Wahana Putra Yudha. By using the backpropagation method, the previous data processing process is carried out which will be used as input to predict the sales level of Link Aja Balance. The data were taken from January 2021 to April 2022. January 2021 to August 2021 were used as training data, while September 2021 to April 2022 were used as test data. The training architecture model used to predict the sales level of Link Aja's Balance is: 4-2-1; 4-25-1; 4-50.1; 4-75-1; and 4-100-1. The best architecture is 4-50-1, the percentage result is 75% in each test
Klasifikasi Peminatan Topik Keilmuan Dalam Penyelesaian Studi Menggunakan Algoritma Naive Bayes Waldi Setiawan; Dedy Hartama; Muhammad Ridwan Lubis; Ihsan Syajidan; Agus Perdana Windarto
Journal of Computing and Informatics Research Vol 3 No 2 (2024): March 2024
Publisher : Forum Kerjasama Pendidikan Tinggi (FKPT)

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.47065/comforch.v3i2.1200

Abstract

Academic expertise is a subject of study taught at the university level to assist students in completing their thesis writing, thereby enabling them to successfully complete their graduate studies. The chosen academic specialization aligns with the vision and mission of each program and can have a positive impact on the university. Students' chosen fields of expertise in completing their studies may either align or not align with the program's vision and mission. The variables used in this research are GPA, MKRV1, MKRV2, and Academic Expertise. The aim of this research is to determine how many students select an academic topic that aligns with the program's vision and mission, particularly in this case, the Computer Science program, as they complete their studies. The Naïve Bayes algorithm is employed in this research, yielding an accuracy rate of 98.11%. This research can provide valuable insights for STIKOM Tunas Bangsa Pematang Siantar to understand the extent to which students from other programs choose academic expertise that aligns with the vision and mission of each program.
Mengelompokkan Daerah Rawan Kecelakaan Di Sumatera Utara dengan Algoritma Clustering Dedy Hartama; Sapriyaldi, Muhammad
JURNAL FASILKOM Vol. 13 No. 3 (2023): Jurnal FASILKOM (teknologi inFormASi dan ILmu KOMputer)
Publisher : Unversitas Muhammadiyah Riau

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.37859/jf.v13i3.6137

Abstract

The large population has a very large need for motorized vehicles, both 2-wheeled and 4-wheeled, which most people consider to be a primary need, not a secondary need. The large number of vehicle users causes traffic congestion so that the number of accidents increases which can result in many fatalities, minor injuries and serious injuries. The aim of this research is to group accident-prone areas in North Sumatra using the clustering method. The data source used in the research is from BPS on the topic of traffic accidents in the North Sumatra region from 2015-2022. The method used to solve this problem is K-Means Data Mining. The results obtained from this search are 3 clusters with a DBI value of 0.384, cluster 1 contains 1 region, cluster 2 contains 16 regions, and cluster 3 contains 11 regions. Carrying out this research can provide knowledge input for further research regarding the development of the k-means clustering method and help the police, especially the traffic accident handling units in each region, in predicting accidents more easily and tracing possible causes. accidents in the area.
Analisis Sosial dan Ekonomi Kerajinan Bordir Sebagai Warisan Budaya Pematangsiantar Menggunakan Bahasa Python Abdi Rahim Damanik; Surya Darma; Dedy Hartama
Jurnal Manajemen, Pendidikan Dan Ilmu Komputer Vol. 1 No. 1 (2024): JMENDIKKOM Volume 1 No 1 Januari 2024
Publisher : Yayasan Darus Soleh Parung

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.65309/00pgpd35

Abstract

Kerajinan bordir di Pematangsiantar merupakan bagian integral dari warisan budaya lokal yang kaya. Penelitian ini bertujuan untuk melakukan analisis sosial dan ekonomi terhadap kerajinan bordir sebagai warisan budaya di Pematangsiantar. Metode penelitian yang digunakan adalah kombinasi antara studi literatur, observasi lapangan, dan wawancara mendalam dengan para pengrajin bordir serta pemangku kepentingan terkait. Analisis sosial mencakup identifikasi praktik budaya yang terkait dengan pembuatan bordir, peran perempuan dalam pengembangan kerajinan ini, serta cara bordir menjadi bagian dari identitas budaya masyarakat setempat. Sementara itu, analisis ekonomi mempertimbangkan nilai ekonomi kerajinan bordir dalam perekonomian lokal, dampaknya terhadap penghidupan pengrajin, serta tantangan dan peluang dalam memasarkan produk bordir secara lokal maupun global. Temuan dari penelitian ini menunjukkan bahwa kerajinan bordir di Pematangsiantar tidak hanya memainkan peran penting dalam melestarikan warisan budaya, tetapi juga memiliki potensi ekonomi yang signifikan. Namun, tantangan seperti pemenuhan pasar yang berkelanjutan dan pendidikan generasi muda dalam mempertahankan tradisi bordir perlu ditangani secara serius untuk menjaga keberlanjutan kerajinan ini sebagai warisan budaya yang hidup dan berdaya saing.
Co-Authors Abdi Rahim Damanik Achmad Noerkhaerin Putra Agus Perdana Windarto Ahmad Riza Fuadi Ambariyanto Ambariyanto Cici Astria Dwi Findi Auliasari Eko Ahadi Fahmi Firzada Ginanti Riski Gita Febrianti Hanifah Urbach Sari Ihsan Syajidan Ika Okta Kirana Indra Gunawan Indra Swanto Ritonga Jonas Rayandi Saragih Khairunnissa Fanny Irnanda Kristin Daya Rohani Sianipar Muasir Pagan Muhammad Ridwan Lubis Muhammad Zarlis Nasib Marbun P.P.P.A.N.W Fikrul Ilmi R.H Zer Rizky Khairunnisa Sormin Sapriyaldi, Muhammad Septri Wanti Siahaan Sinta Maulina Dewi Solikhun Suranta Bill Fatric Ginting Surya Darma Waldi Setiawan Wanayumini Wanto, Anjar Wendi Robiansyah