Rosihan Rosihan
Current Organizations / Institutes
The organizations/institutes from the scholar's latest article
Rosihan Rosihan has a total of 10 co-authors. Here are the top co-authors by number of shared publications:
Ardi Salman
Universitas Khairum
1 shared publication- Catur Eri Gunawan
Universitas Islam Negeri Raden Fatah Palembang
1 shared publication Fatorahman A Saadu
Universitas Khairum
1 shared publication- Firman Tempola
Teknik Informatika Universitas Khairun
1 shared publication - Hairil Kurniadi Siradjuddin
Universitas Khairun Ternate
1 shared publication - Masdar S Hanafi
Universitas Khairun
1 shared publication
Latest Publications
- Open Access
- 10 April 2025
- Published by Komunitas Dosen Indonesia in bit-Tech
- Vol. 7 (3) , 995-1003
Analisis Sentimen Calon Kepala Daerah Maluku Utara dengan Metode CRISP-DM
Pemilihan kepala daerah merupakan salah satu wujud nyata dari demokrasi lokal yang memungkinkan masyarakat mengekspresikan aspirasi politiknya secara langsung. Namun, untuk memahami dinamika persepsi publik terhadap calon kepala daerah, dibutuhkan pendekatan analitis yang mampu menafsirkan opini masyarakat dalam skala luas dan waktu nyata. Penelitian ini bertujuan untuk menganalisis sentimen masyarakat terhadap calon kepala daerah di Provinsi Maluku Utara dengan memanfaatkan data komentar dari media sosial YouTube dan TikTok. Subjek penelitian berupa komentar publik yang dikumpulkan melalui teknik crawling dengan kata kunci tertentu terkait nama calon dan isu politik lokal. Penelitian ini menggunakan pendekatan CRISP-DM (Cross Industry Standard Process for Data Mining) yang terdiri dari enam tahapan: pemahaman bisnis, pemahaman data, persiapan data, pemodelan, evaluasi, dan deployment. Proses analisis mencakup preprocessing teks seperti tokenizing, case folding, stop word removal, dan stemming. Komentar diklasifikasikan secara manual ke dalam tiga kategori sentimen: positif, negatif, dan netral. Model klasifikasi dibangun menggunakan algoritma Neural Network melalui platform Orange Data Mining. Hasil penelitian menunjukkan bahwa kandidat 3 memperoleh sentimen positif tertinggi, kandidat 1 mendapat sentimen negatif terbanyak, dan kandidat 5 mendominasi pada sentimen netral. Model menunjukkan kinerja tinggi dengan akurasi mencapai 97,2%, presisi 98,5%, dan recall 88,8%. Temuan ini menunjukkan bahwa kombinasi metode CRISP-DM dan pembelajaran mesin dapat memberikan wawasan strategis yang bermanfaat bagi pengambil kebijakan dan tim kampanye dalam memahami persepsi publik secara lebih komprehensif dan real-time. Penelitian ini dapat menjadi referensi bagi pemangku kepentingan dalam merancang strategi komunikasi politik yang lebih tepat sasaran.
- 10 April 2025
- Published by Komunitas Dosen Indonesia in bit-Tech
- Vol. 7 (3) , 995-1003
Open AccessAnalisis Sentimen Calon Kepala Daerah Maluku Utara dengan Metode CRISP-DM
Pemilihan kepala daerah merupakan salah satu wujud nyata dari demokrasi lokal yang memungkinkan masyarakat mengekspresikan aspirasi politiknya secara langsung. Namun, untuk memahami dinamika persepsi publik terhadap calon kepala daerah, dibutuhkan pendekatan analitis yang mampu menafsirkan opini masyarakat dalam skala luas dan waktu nyata. Penelitian ini bertujuan untuk menganalisis sentimen masyarakat terhadap calon kepala daerah di Provinsi Maluku Utara dengan memanfaatkan data komentar dari media sosial YouTube dan TikTok. Subjek penelitian berupa komentar publik yang dikumpulkan melalui teknik crawling dengan kata kunci tertentu terkait nama calon dan isu politik lokal. Penelitian ini menggunakan pendekatan CRISP-DM (Cross Industry Standard Process for Data Mining) yang terdiri dari enam tahapan: pemahaman bisnis, pemahaman data, persiapan data, pemodelan, evaluasi, dan deployment. Proses analisis mencakup preprocessing teks seperti tokenizing, case folding, stop word removal, dan stemming. Komentar diklasifikasikan secara manual ke dalam tiga kategori sentimen: positif, negatif, dan netral. Model klasifikasi dibangun menggunakan algoritma Neural Network melalui platform Orange Data Mining. Hasil penelitian menunjukkan bahwa kandidat 3 memperoleh sentimen positif tertinggi, kandidat 1 mendapat sentimen negatif terbanyak, dan kandidat 5 mendominasi pada sentimen netral. Model menunjukkan kinerja tinggi dengan akurasi mencapai 97,2%, presisi 98,5%, dan recall 88,8%. Temuan ini menunjukkan bahwa kombinasi metode CRISP-DM dan pembelajaran mesin dapat memberikan wawasan strategis yang bermanfaat bagi pengambil kebijakan dan tim kampanye dalam memahami persepsi publik secara lebih komprehensif dan real-time. Penelitian ini dapat menjadi referensi bagi pemangku kepentingan dalam merancang strategi komunikasi politik yang lebih tepat sasaran.
- Open Access
- 24 January 2024
- Published by Universitas 45 Surabaya in Cakrawala: Jurnal Pengabdian Masyarakat Global
- Vol. 3 (2) , 151-161
Pelatihan Penggunaan Aplikasi Online sebagai Penunjang Model Pembelajaran Digital di Sekolah Dasar Swasta (SDS) Muhammadiyah Indonesiana Kota Tidore Kepulauan
Education is the cornerstone of a nation's development, and in facing the era of globalization and the Fourth Industrial Revolution, education in Indonesia must continue to transform. One crucial aspect is the implementation of information and communication technology (ICT) in the learning process. Elementary Schools (SD) as the initial stage of formal education are expected to utilize technological advancements, including the use of online applications, to enhance the effectiveness and quality of learning. The use of online applications as a support for digital learning in elementary schools is urgent due to several factors: Rapid Technological Development, Curriculum Demands, Limitations of Conventional Learning Resources, Importance of Digital Skills from an Early Age, and Parental Involvement in Learning.The Community Service Program of Khairun University at SDS Muhammadiyah Tidore involves the socialization and training in the use of online learning methods (e-learning) utilizing web tools such as Edpuzzle and Quizizz. This program is expected to provide sustainable solutions and contributions to the field of education by leveraging technological advancements. The implementation of this program aims to enhance the digital literacy of teachers and students at SDS Muhammadiyah Tidore, strengthen digital literacy, and create an innovative and interactive learning environment while maintaining ethical and professional values. Through this activity, educators are expected to adapt to technological developments in the field of education, reduce the digital divide between educators and students, and implement a more modern learning model. Continuous evaluation and the development of additional materials are expected to sustain the positive impact of this training in the long term.
- 24 January 2024
- Published by Universitas 45 Surabaya in Cakrawala: Jurnal Pengabdian Masyarakat Global
- Vol. 3 (2) , 151-161
Open AccessPelatihan Penggunaan Aplikasi Online sebagai Penunjang Model Pembelajaran Digital di Sekolah Dasar Swasta (SDS) Muhammadiyah Indonesiana Kota Tidore Kepulauan
Education is the cornerstone of a nation's development, and in facing the era of globalization and the Fourth Industrial Revolution, education in Indonesia must continue to transform. One crucial aspect is the implementation of information and communication technology (ICT) in the learning process. Elementary Schools (SD) as the initial stage of formal education are expected to utilize technological advancements, including the use of online applications, to enhance the effectiveness and quality of learning. The use of online applications as a support for digital learning in elementary schools is urgent due to several factors: Rapid Technological Development, Curriculum Demands, Limitations of Conventional Learning Resources, Importance of Digital Skills from an Early Age, and Parental Involvement in Learning.The Community Service Program of Khairun University at SDS Muhammadiyah Tidore involves the socialization and training in the use of online learning methods (e-learning) utilizing web tools such as Edpuzzle and Quizizz. This program is expected to provide sustainable solutions and contributions to the field of education by leveraging technological advancements. The implementation of this program aims to enhance the digital literacy of teachers and students at SDS Muhammadiyah Tidore, strengthen digital literacy, and create an innovative and interactive learning environment while maintaining ethical and professional values. Through this activity, educators are expected to adapt to technological developments in the field of education, reduce the digital divide between educators and students, and implement a more modern learning model. Continuous evaluation and the development of additional materials are expected to sustain the positive impact of this training in the long term.
- Open Access
- 1 January 2024
- Published by Engineering and Technology Publishing in Journal of Advances in Information Technology
- Vol. 15 (3) , 407-413
Improving System Accuracy by Modifying the Transfer Learning Architecture for Detecting Clove Maturity Levels
JAIT is a scientific open access journal which focuses on empirical research results and critical analysis of technology development, use, management and impacts in information technology.
- 1 January 2024
- Published by Engineering and Technology Publishing in Journal of Advances in Information Technology
- Vol. 15 (3) , 407-413
Open AccessImproving System Accuracy by Modifying the Transfer Learning Architecture for Detecting Clove Maturity Levels
JAIT is a scientific open access journal which focuses on empirical research results and critical analysis of technology development, use, management and impacts in information technology.