Article Per Year (5 Year)
p-Index From 2021 - 2026
0.444
P-Index
This Author published in this journals
All Journal INTECOMS: Journal of Information Technology and Computer Science JSI (Jurnal sistem Informasi) Universitas Suryadarma
Hendrakusuma, Dannie Febrianto
Unknown Affiliation
Computer Science & IT

Published : 2 Documents Claim Missing Document
Claim Missing Document
Check
Articles

Found 2 Documents
Search

 1 
Perbandingan Kinerja Algoritma Klasifikasi Machine Learning Dalam Analisis Sentimen Ulasan Mobile Banking WONDR BY BNI Riza, Faizal; Hendrakusuma, Dannie Febrianto; Wibowo, Budi; Al Afghani, Dhian Yusuf
INTECOMS: Journal of Information Technology and Computer Science Vol. 8 No. 2 (2025): INTECOMS: Journal of Information Technology and Computer Science
Publisher : Institut Penelitian Matematika, Komputer, Keperawatan, Pendidikan dan Ekonomi (IPM2KPE)

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.31539/intecoms.v8i2.14826

Abstract

Penelitian ini menganalisis sentimen ulasan pengguna terhadap layanan dan fitur yang ditawarkan oleh WONDR by BNI dengan menerapkan berbagai algoritma klasifikasi Machine Learning. Data diperoleh dari ulasan pengguna di Google Play Store, yang kemudian melalui tahap pra-pemrosesan teks, mencakup pembersihan data dan normalisasi, sebelum digunakan dalam pelatihan model klasifikasi. Studi ini membandingkan kinerja Support Vector Machine (SVM), Light Gradient Boosting Machine (LGBM), XGBoost, Random Forest, CatBoost, dan Naïve Bayes dalam mengklasifikasikan sentimen pengguna. Hasil evaluasi menunjukkan bahwa SVM memiliki akurasi tertinggi sebesar 0,881, diikuti oleh LGBM dengan akurasi 0,877, yang unggul dalam efisiensi waktu eksekusi sebesar 11,80 detik. XGBoost memperoleh akurasi sebesar 0,872, diikuti oleh Random Forest dengan akurasi 0,866, sementara CatBoost dan Naïve Bayes mencatatkan akurasi masing-masing sebesar 0,85 dan 0,72. Meskipun SVM menunjukkan kinerja terbaik dalam klasifikasi sentimen, algoritma ini memiliki kelemahan dalam hal waktu eksekusi yang lebih lama, yaitu 57,5 detik pada mesin dengan spesifikasi RAM 2 GB dan 2 vCPU. Sebaliknya, LGBM dan XGBoost, meskipun memiliki akurasi yang sedikit lebih rendah, menunjukkan efisiensi komputasi 47,5 detik lebih gegas, sehingga lebih optimal untuk implementasi dalam sistem berskala besar
PERANCANGAN BAHAN PELATIHAN BERBASIS WEBSITE DI KEJURUAN TEKNOLOGI INFORMASI DAN KOMUNIKASI BALAI BESAR PENGEMBANGAN LATIHAN KERJA BEKASI Riza, Faizal; Hendrakusuma, Dannie Febrianto; Setiabudi, Rachmat; Birowo, Anindito Kusumo; Hidayat, Cahyono Koerniawan
JSI (Jurnal sistem Informasi) Universitas Suryadarma Vol 10 No 2 (2023): JSI (Jurnal sistem Informasi) Universitas Suryadarma
Publisher : Universitas Dirgantara Marsekal Suryadarma

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.35968/jsi.v10i2.1081

Abstract

 AbstractBalai Besar Pengembangan Latihan Kerja (BBPLK) Bekasi is a training place for government-owned skills or expertise under the auspices of the Ministry of Manpower headed by Echelon II government officials. In the Center for Job Training Development (BBPLK) Bekasi, there are various kinds of vocational training, one of which is the Information and Communication Technology Vocational School. In the Information and Communication Technology (ICT) Vocational School there are 5 (five) training programs, namely: Graphic Design, Motion Graphic Artist, Network Professional, Mobile Programming and IT Software For Business Solution. In this vocational school, they still use the manual method in processing data on training materials stored in the warehouse. The method used is by filling out the training material expenditure form, then the form is stored in their respective archive folders. This system has been running well but still has weaknesses, including if the admin is asked to find training material data quickly, the admin will have difficulty finding one by one archive that has been stored and takes a long time and taking training materials that are not in accordance with the training program and possibly there is a missing file. The research method uses data collection by means of observation, interviews and literature study. So with the design of a website-based training material inventory information system, it can make it easier to find data on the availability of training materials according to the training program and make the work process of checking training materials more effective and efficient. Keywords : training materials, system, information, inventory, website..
Co-Authors Al Afghani, Dhian Yusuf Birowo, Anindito Kusumo Faizal Riza Hidayat, Cahyono Koerniawan Riza, Faizal Setiabudi, Rachmat