Article Per Year (5 Year)
p-Index From 2021 - 2026
2.794
P-Index
This Author published in this journals
All Journal Inovator : Jurnal Manajemen Abdi Dosen : Jurnal Pengabdian Pada Masyarakat Jurnal Ilmiah Manajemen Kesatuan JURMA : Jurnal Program Mahasiswa Kreatif Moneter : Jurnal Keuangan dan Perbankan Indonesian Journal of Multidisciplinary on Social and Technology Jurnal Pengabdian Nusantara Manager : Jurnal Ilmu manajemen
Hanif Zaidan Sinaga
Universitas Ibn Khaldun Bogor
Religion Agriculture, Biological Sciences & Forestry Humanities Computer Science & IT Economics, Econometrics & Finance Education Engineering Languange, Linguistic, Communication & Media Law, Crime, Criminology & Criminal Justice Public Health Social Sciences

Published : 16 Documents Claim Missing Document
Claim Missing Document
Check
Articles

Found 16 Documents
Search

 1   2 
Analisis Sentimen Opini Twitter (X) terhadap Penggunaan ChatGPT dalam Pendidikan Tinggi dengan Perbandingan Algoritma Naïve Bayes, Support Vector Machine, dan K-Nearest Neighbors Salma Difa Aristawidya; Tabita Nurtirta Purwita Sari; Hanun Syahidah Ulfya; Adinda Aghnia Fatihin; Hanif Zaidan Sinaga
Indonesian Journal of Multidisciplinary on Social and Technology Vol. 4 No. 3 (2026): Juli - Oktober
Publisher : PT Ilmu Data Indonesia

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.69693/ijmst.v4i3.11803

Abstract

Penelitian ini bertujuan untuk menganalisis sentimen opini Twitter (X) terhadap penggunaan ChatGPT dalam pendidikan tinggi serta membandingkan kinerja algoritma Naïve Bayes, Support Vector Machine (SVM), dan K-Nearest Neighbors (KNN) dalam proses klasifikasi sentimen. Dataset yang digunakan diperoleh dari Kaggle dan terdiri atas 1.153 data opini yang telah memiliki label sentimen positif, netral, dan negatif. Tahapan penelitian meliputi preprocessing data, ekstraksi fitur menggunakan Term Frequency-Inverse Document Frequency (TF-IDF), seleksi fitur menggunakan Chi-Square, klasifikasi sentimen, serta evaluasi model menggunakan metode Stratified 10-Fold Cross Validation dengan metrik Accuracy, Precision, Recall, F1-Score, dan Confusion Matrix. Hasil penelitian menunjukkan bahwa distribusi sentimen didominasi oleh sentimen negatif sebesar 40%, diikuti sentimen positif sebesar 35%, dan sentimen netral sebesar 25%. Berdasarkan hasil perbandingan algoritma, Support Vector Machine (SVM) memperoleh performa terbaik dengan tingkat akurasi sebesar 61,75%, diikuti oleh Naïve Bayes sebesar 58,20%, dan K-Nearest Neighbors (KNN) sebesar 44,41%. Hasil tersebut menunjukkan bahwa SVM memiliki performa relatif terbaik dibandingkan Naïve Bayes dan KNN, meskipun nilai akurasi yang diperoleh menunjukkan bahwa klasifikasi opini Twitter (X) masih menghadapi tantangan akibat kompleksitas bahasa informal, ambiguitas sentimen, dan konteks penggunaan ChatGPT dalam pendidikan tinggi.
Komparasi Kinerja Metode Naïve Bayes Dan SVM Untuk Analisis Sentimen Opini Publik Di Youtube David Gadgetin Terhadap Iphone 17 Pro Febrina Rahmadianti Putri; Miftahul Jannah; Restiayu Sekar Utami; Shintawati Khoirunnisa; Hanif Zaidan Sinaga
Indonesian Journal of Multidisciplinary on Social and Technology Vol. 4 No. 3 (2026): Juli - Oktober
Publisher : PT Ilmu Data Indonesia

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.69693/ijmst.v4i3.11812

Abstract

Penelitian ini bertujuan untuk membandingkan kinerja algoritma Naïve Bayes dan Support Vector Machine (SVM) dalam analisis sentimen opini publik pada komentar YouTube terkait penggunaan iPhone 17 Pro di kanal David Gadgetin. Penelitian menggunakan pendekatan Natural Language Processing (NLP) dengan memanfaatkan 1.382 komentar yang diperoleh melalui proses web scraping menggunakan YouTube Data API v3. Dataset diproses melalui tahapan text preprocessing yang meliputi case folding, normalization, tokenization, filtering (stopword removal), dan stemming, kemudian direpresentasikan menggunakan pembobotan Term Frequency–Inverse Document Frequency (TF-IDF). Proses klasifikasi dilakukan menggunakan algoritma Naïve Bayes dan SVM berbasis kernel linear dengan skema pembagian data 80:10:10, yaitu 80% data latih, 10% data validasi, dan 10% data uji. Kinerja kedua model dievaluasi menggunakan metrik accuracy, precision, recall, F1-score, serta analisis confusion matrix. Hasil pengujian menunjukkan bahwa model SVM memperoleh akurasi sebesar 81,16% pada data validasi dan 84,17% pada data uji, sedangkan Naïve Bayes memperoleh akurasi sebesar 77i,54% pada data validasi dan 74,82% pada data uji. Pada kelas sentimen negatif, SVM memperoleh F1-score sebesar 0,86, sedangkan Naïve Bayes memperoleh 0,00. Berdasarkan hasil eksperimen pada dataset penelitian ini, SVM menunjukkan performa yang relatif lebih baik dibandingkan Naïve Bayes dalam klasifikasi sentimen komentar YouTube terhadap iPhone 17 Pro. Meskipun demikian, interpretasi terhadap hasil pada kelas minoritas tetap perlu dilakukan secara hati-hati karena jumlah data negatif relatif sedikit.
Determining Marketing Performance Product Innovation and Digital Marketing Asti Marlina; Lucky Hikmat Maulana; Hanif Zaidan Sinaga; Jihan Fadhila; Miranti
Jurnal Ilmiah Manajemen Kesatuan Vol. 13 No. 4 (2025): JIMKES Edisi Juli 2025
Publisher : LPPM Institut Bisnis dan Informatika Kesatuan

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.37641/jimkes.v13i4.3431

Abstract

In the digital era, product innovation and digital marketing are vital for enhancing the marketing performance of fashion Micro, Small, and Medium Enterprises in Bogor City, where competition is intensifying. This study aims to examine the influence of product innovation and digital marketing, particularly through social media, on marketing performance. A quantitative approach with descriptive and verification methods was employed, using a questionnaire survey distributed to 65 fashion Micro, Small, and Medium Enterprises in Bogor City, selected via purposive sampling. Data were analyzed using SPSS software through multiple regression, correlation, and statistical hypothesis testing. The findings reveal that product innovation significantly enhances marketing performance by meeting consumer demands, while digital marketing, especially social media strategies, boosts brand visibility and customer engagement. Both factors collectively account for 60% of marketing performance variance, with social media playing a critical role in promoting innovative products. This study concludes that fashion Micro, Small, and Medium Enterprises must integrate innovative product designs with active social media marketing to remain competitive. These insights offer practical guidance for Micro, Small, and Medium Enterprises to optimize marketing strategies and strengthen market presence in a dynamic environment.
Implementasi Klasifikasi Genre Film Pada Platform Imdb Berdasarkan Deskripsi Sinopsis Film Menggunakan Algoritma Naïve Bayes, SVM, Dan Logistic Regression Muhammad Ilyassa; Satrio Akbar; Ikhwanul Akmal; Fakhrullah Abrisam; Hanif Zaidan Sinaga
Indonesian Journal of Multidisciplinary on Social and Technology Vol. 4 No. 3 (2026): Juli - Oktober
Publisher : PT Ilmu Data Indonesia

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.69693/ijmst.v4i3.12210

Abstract

Pertumbuhan data film digital dan meningkatnya penggunaan platform IMDb menuntut metode otomatis yang mampu mengelompokkan genre berdasarkan informasi tekstual secara konsisten. Penelitian ini bertujuan mengimplementasikan Natural Language Processing dan membandingkan kinerja Multinomial Naïve Bayes, Support Vector Machine LinearSVC, SVM dengan pembobotan kelas seimbang, serta Logistic Regression untuk mengklasifikasikan genre primer film berdasarkan sinopsis. Dataset IMDb Top 1000 dari Kaggle diseleksi menjadi 811 film pada lima genre utama, yaitu Drama, Action, Comedy, Crime, dan Biography. Tahapan penelitian meliputi case folding, cleaning, tokenization, stopword removal, stemming, ekstraksi fitur Term Frequency-Inverse Document Frequency, pembagian data secara stratified 80:20, pelatihan model, evaluasi accuracy, precision, recall, F1-score, confusion matrix, dan validasi silang stratified lima lipatan. Baseline kelas mayoritas menghasilkan akurasi 35,6%. Logistic Regression memberikan hasil terbaik dengan akurasi pengujian 43,56% dan rata-rata validasi silang 44,63%. SVM Balanced mencapai akurasi 42,33%, LinearSVC 41,10%, dan Naïve Bayes 40,49%. Naïve Bayes menunjukkan bias kuat terhadap kelas Drama, sedangkan SVM dan Logistic Regression menghasilkan prediksi lebih seimbang. Kinerja yang masih terbatas terutama dipengaruhi ketidakseimbangan kelas, tumpang tindih naratif antargenre, sinopsis yang pendek, dan penyederhanaan label multigenre menjadi genre primer. Hasil penelitian menegaskan bahwa Logistic Regression merupakan model klasik paling efektif pada konfigurasi data dan fitur yang digunakan serta layak dijadikan baseline untuk penelitian lanjutan.
Young Consumers’ Intention to use Blu by BCA: An Extended TAM Widhi Ariyo Bimo; Hanif Zaidan Sinaga; Mohammad Jibriel Avessina; Ali Nurjali; Nurwinda Nurwinda
Moneter: Jurnal Keuangan dan Perbankan Vol. 14 No. 1 (2026): APRIL
Publisher : Universitas Ibn Khladun Bogor

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.32832/moneter.v14i1.3107

Abstract

This study examines factors influencing intention to use the blu by BCA digital banking application among young consumers in Bogor, Indonesia. An extended Technology Acceptance Model was tested by incorporating perceived usefulness, perceived ease of use, perceived risk, trust, and convenience. Data were collected through a cross-sectional online survey of 100 blu by BCA users selected using purposive sampling and analyzed using item-validity testing, Cronbach's alpha, and multiple linear regression. All reported predictor items met the stated validity and reliability criteria. Convenience had a significant positive effect on intention to use (B = 0.432; p = 0.001). Trust (B = 0.229; p = 0.051) and perceived ease of use (B = 0.485; p = 0.088) were positive but not significant at the 5% level. Perceived risk was negative but non-significant (B = -0.183; p = 0.489), while perceived usefulness was negligible (B = -0.006; p = 0.959). The findings indicate that practical accessibility and transaction efficiency receive the strongest empirical support in the reported model. Digital banks should prioritize frictionless transactions, reliable access, and transparent security communication while strengthening distinctive service value.
MSME Digitalization through E-Catalogs, Google Maps, and QRIS in Bojong Kulur Mohammad Jibriel Avessina; Hanif Zaidan Sinaga; Widhi Ariyo Bimo; Asti Marlina; Asifa Ramadhanti Putri; Dava Pratama; Muhtadin Muhtadin; Nida Ussyaripah Khoerul Akilah; Raka Andika Pratita; Siti Uswatun Hasanah
Jurnal Pengabdian Nusantara Vol. 4 No. 4 (2026)
Publisher : Konsorsium Nasional Pengelola Jurnal Pengabdian

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.32832/jpn.v4i4.263

Abstract

Micro, small, and medium enterprises (MSMEs) are central to local economic activity, yet village-based businesses often face limited digital promotion, non-cash payment readiness, and online business visibility. This community service program assisted MSMEs in Bojong Kulur Village through e-catalog development, Google Maps registration, Quick Response Code Indonesian Standard (QRIS) implementation, and business-branding media. A participatory approach involved Village-Owned Enterprise managers and MSME actors in observation, planning, implementation, evaluation, and follow-up. The program produced an e-catalog containing 15 featured products, activated QRIS for one MSME, registered one business location on Google Maps, and delivered a new branding banner. Google Maps recorded more than 20 location searches during the first two weeks, while several non-cash transactions were reported during the first week of QRIS use. The intervention strengthened initial business visibility, digital transaction readiness, and promotional capacity. However, long-term sustainability requires a designated digital administrator, regular content training, periodic catalog updates, and continued support from the Village-Owned Enterprise and village government.
Co-Authors Adinda Aghnia Fatihin Ahmad Fathan Mujadidi Haqqani Akbar Supia Dirja Alfiyyah Rima Ali Nurjali Amanda, Sifa Saskia Amelia, Anggi Ardhy, Muhammad Maliki Arsyad Abdullah Asifa Ramadhanti Putri Asti Marlina Asti Marlina Cika Jelita Dava Pratama Digna, Akira Alpha Ecin Kuraesin Fahtir Angger Prabowo Fakhiroh, Siti Zakiyatul Fakhrullah Abrisam Fara Diba Febrina Rahmadianti Putri Hana Aurora Aida Rahma Hanun Syahidah Ulfya Hoirunnisa, Siti Mega Sania Ikhwanul Akmal Ilyas Lutfi Abdat Jihan Fadhila Karmelia, Anita Lana Aulia Salsabila M. Akhdan Putra Hermawan Marlina, Asti Maulana, Akhmad Dzulfiqar Maulana, Lucky Hikmat Miftahul Jannah Miranti Mohammad Jibriel Avessina Muhammad Ilyassa Muhammad Iqbal Fauzi Muhtadin Muhtadin Nida Ussyaripah Khoerul Akilah Nuraida, Dhita Nurwinda Nurwinda Pinkan Prilia Mahmudah Pramono, Setio Pratama, Muhammad Akmal Putra Rahma, Alya Rahmawati, Amelia Raka Andika Pratita Ramadhani, Anisa Nur Restiayu Sekar Utami Rusydiana MS, Rizqy Salma Difa Aristawidya Satrio Akbar Sephiana, Della Shintawati Khoirunnisa Sinta Tiara, Euis Siti Uswatun Hasanah Tabita Nurtirta Purwita Sari Widhi Ariyo Bimo Yola Tirta Nurfajriya