Claim Missing Document
Check
Articles

Found 3 Documents
Search

Strategi Dinamis Menggunakan MongoDB untuk Analisis Sentimen Terhadap Komentar YouTube Pilkada Gubernur Indonesia 2024 Asro Asro; Alwin Hartono Limaran; Agung Wicaksono; Patah Herwanto
KRESNA: Jurnal Riset dan Pengabdian Masyarakat Vol 4 No 2 (2024): Jurnal KRESNA November 2024
Publisher : DRPM Universitas Budi Luhur

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.36080/kresna.v4i2.149

Abstract

Penelitian ini mengkaji penggunaan MongoDB dalam melakukan analisis sentimen terhadap komentar YouTube terkait Pilkada Gubernur Indonesia 2024. Studi ini mengumpulkan 40,396 komentar melalui YouTube Data API, yang kemudian diproses menjadi 27,102 komentar relevan setelah melalui tahapan preprocessing yang menyeluruh, meliputi case folding, tokenisasi, normalisasi, dan stemming. Tiga algoritma analisis sentimen digunakan dalam penelitian ini: Naive Bayes, Logistic Regression, dan Support Vector Machine (SVM). Di antara ketiganya, Logistic Regression menunjukkan kinerja terbaik dengan tingkat akurasi mencapai 91.39%. Analisis ini memberikan wawasan berharga tentang sentimen publik, secara efektif menangkap sikap dan opini terhadap calon gubernur. Untuk meningkatkan interpretasi hasil, penelitian ini menggunakan teknik visualisasi data seperti Word Cloud dan grafik frekuensi. Representasi visual ini memberikan pemahaman yang lebih jelas tentang frekuensi dan konteks penggunaan kata-kata kunci, sehingga membantu dalam pengembangan strategi komunikasi politik yang efektif dan responsif. Penelitian ini berkontribusi dalam bidangnya dengan mendemonstrasikan potensi MongoDB dalam memproses dan menganalisis data media sosial skala besar untuk analisis sentimen politik. Temuan-temuan ini memiliki implikasi signifikan untuk memahami tren pemilih dan membentuk kampanye politik dalam konteks Pilkada Gubernur Indonesia 2024.
Dampak Penggunaan Perangkat Mobile dalam Mendukung Kegiatan Pembelajaran Mandiri Siswa Kelas IX di SMP PGRI Rancaekek Santi Puspa Oktaviane; Patah Herwanto
Uranus : Jurnal Ilmiah Teknik Elektro, Sains dan Informatika Vol. 2 No. 4 (2024): Desember: Jurnal Ilmiah Teknik Elektro, Sains dan Informatika
Publisher : Asosiasi Riset Teknik Elektro dan Informatika Indonesia

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.61132/uranus.v2i4.491

Abstract

The development of information and communication technology, especially mobile devices, has changed the learning landscape significantly. This research aims to examine the impact of using mobile devices in supporting the independent learning activities of class IX students at SMP PGRI Rancaekek. This research is motivated by the increasingly widespread ownership of mobile devices among students and their potential to increase motivation and learning effectiveness. The research method used is descriptive qualitative with data collection through observation, interviews and documentation. The research subjects were class IX students and subject teachers. Data analysis was carried out inductively to identify emerging themes related to the impact of mobile device use. The research results show that the use of mobile devices has a positive impact on students' independent learning activities. Mobile devices provide easier access to various learning resources, facilitate communication between students and teachers, and increase learning motivation. However, the research also found several challenges, such as dependency on devices, lack of monitoring of use, and gaps in access. The implication of this research is the importance of integrating the use of mobile devices in the learning process. Schools need to develop clear policies regarding the use of mobile devices, provide training for teachers, and ensure equitable access for all students. In addition, further research is needed to examine the long-term impact of mobile device use on student learning achievement.
Pengembangan Sistem Cermin Cerdas dengan Fitur Rekomendasi Gaya Rambut Berbasis Metode Machine Learning Wahyu Tri Admaja; Patah Herwanto
Reslaj: Religion Education Social Laa Roiba Journal Vol. 7 No. 7 (2025): RESLAJ: Religion Education Social Laa Roiba Journal
Publisher : Intitut Agama Islam Nasional Laa Roiba Bogor

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.47467/reslaj.v7i7.8115

Abstract

Entering the modern era, at least changing the lifestyle of every human being, especially changes in appearance. Appearance is now a need that is considered important for society, including for men. A good appearance will make someone more confident in carrying out daily activities.[1]⁠ In appearance, hair is an important thing in appearance. In this modern era, hairstyles are increasingly diverse models. So, many teenage men have difficulty determining the right hairstyle.[2]⁠ Smart mirror is an interactive technology innovation that combines facial recognition system and artificial intelligence to provide a more personalized user experience. This study aims to develop a smart mirror system that is able to recommend hairstyles according to the user's face shape using machine learning methods. The system is designed by utilizing a camera as input for facial images, a facial shape classification module based on facial landmarks, and a hairstyle recommendation module. This simulation application is supported by a wide variety of hair model options that users can try.[3]⁠ The classification process is carried out by measuring the ratio of facial dimensions to determine the category of face shape, such as oval, round, oblong, square, and others. After classification, the system automatically displays the most appropriate hairstyle suggestions through the face shape that has been detected web-based. The test results show that the system can recognize faces and can provide hairstyle recommendations with a fairly good level of accuracy in limited environments.