REGIANDO, REGIANDO (2025) PENERAPAN ALGORITMA RECURRENT NEURAL NETWORK (RNN) DAN LONG SHORT-TERM MEMORY (LSTM) UNTUK SISTEM REKOMENDASI KONTEN TIKTOK FAKULTAS ILMU KOMPUTER UNIVERSITAS MERCU BUANA. S1 thesis, Universitas Mercu Buana Jakarta.
|
Text (HAL COVER)
01 COVER.pdf Download (1MB) | Preview |
|
![[img]](https://repository.mercubuana.ac.id/style/images/fileicons/text.png) |
Text (BAB I)
02 BAB 1.pdf Restricted to Registered users only Download (157kB) |
|
|
Text (BAB II)
03 BAB 2.pdf Restricted to Registered users only Download (247kB) |
|
|
Text (BAB III)
04 BAB 3.pdf Restricted to Registered users only Download (243kB) |
|
|
Text (BAB IV)
05 BAB 4.pdf Restricted to Registered users only Download (649kB) |
|
|
Text (BAB V)
06 BAB 5.pdf Restricted to Registered users only Download (120kB) |
|
|
Text (DAFTAR PUSTAKA)
07 DAFTAR PUSTAKA.pdf Restricted to Registered users only Download (156kB) |
|
|
Text (LAMPIRAN)
08 LAMPIRAN.pdf Restricted to Registered users only Download (868kB) |
Abstract
In the digital era, social media platforms like TikTok have become essential for the younger generation, including university students, to share and acquire information. However, the abundance of available content often makes it challenging for users to find relevant and beneficial information. This study aims to analyze and predict video interaction levels based on parameters such as duration, upload time, and content category using the Recurrent Neural Network (RNN) algorithm. The dataset was processed using TF-IDF techniques for textual features from video titles and descriptions, followed by numerical feature normalization. The RNN model was developed using a bidirectional LSTM architecture equipped with dropout layers to prevent overfitting. Model evaluation was conducted using MAE, RMSE, and R² metrics, demonstrating good performance in predicting video interactions. The analysis results indicate that video duration and upload time significantly affect average interactions, with recommended optimal durations capable of enhancing user engagement. This research contributes to developing data-driven strategies to improve content interaction on digital video platforms. Keywords: Recurrent Neural Network (RNN), LSTM, Interaction Prediction, TF-IDF, Video Duration, Upload Time, Digital Media Dalam era digital saat ini, media sosial seperti TikTok menjadi platform penting bagi generasi muda, termasuk mahasiswa, untuk berbagi dan mendapatkan informasi. Namun, besarnya jumlah konten yang tersedia mengakibatkan kesulitan dalam menemukan informasi yang relevan dan bermanfaat bagi setiap pengguna. Penelitian ini bertujuan untuk menganalisis dan memprediksi tingkat interaksi video berdasarkan parameter durasi, waktu unggah, dan kategori konten menggunakan algoritma Recurrent Neural Network (RNN). Dataset diproses menggunakan teknik TF-IDF untuk fitur teks dari judul dan deskripsi video, diikuti dengan normalisasi fitur numerik. Model RNN dibangun menggunakan arsitektur LSTM bidirectional, yang dilengkapi dengan lapisan dropout untuk mencegah overfitting. Evaluasi model dilakukan menggunakan metrik MAE, RMSE, dan R², menunjukkan kinerja yang baik dalam prediksi interaksi video. Hasil analisis menunjukkan bahwa durasi video dan waktu unggah memiliki pengaruh signifikan terhadap rata-rata interaksi, dengan rekomendasi durasi optimal yang dapat meningkatkan keterlibatan pengguna. Penelitian ini memberikan kontribusi pada pengembangan strategi peningkatan interaksi konten berbasis data pada platform video digital. Kata Kunci: Recurrent Neural Network (RNN), LSTM, Prediksi Interaksi, TF-IDF, Durasi Video, Waktu Unggah, Media Digital.
Actions (login required)
 |
View Item |
