CHANDRA, DENI (2025) ANALISIS PERFORMA SISTEM REKOMENDASI LAPTOP BERBASIS HYBRID DENGAN ANALISIS SENTIMEN MENGGUNAKAN MODEL INDOBERT DAN GRADIENT BOOSTING. S1 thesis, Universitas Mercu Buana Jakarta.
|
Text (HAL COVER)
01 Cover.pdf Download (600kB) | 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 (228kB) |
|
|
Text (BAB II)
03 Bab 2.pdf Restricted to Registered users only Download (586kB) |
|
|
Text (BAB III)
04 Bab 3.pdf Restricted to Registered users only Download (275kB) |
|
|
Text (BAB IV)
05 Bab 4.pdf Restricted to Registered users only Download (499kB) |
|
|
Text (BAB V)
06 Bab 5.pdf Restricted to Registered users only Download (196kB) |
|
|
Text (DAFTAR PUSTAKA)
07 Daftar Pustaka.pdf Restricted to Registered users only Download (242kB) |
|
|
Text (LAMPIRAN)
08 Lampiran.pdf Restricted to Registered users only Download (1MB) |
Abstract
Laptops have become essential devices that support various aspects of life, from education to strategic sectors such as business, health and tourism. However, the laptop ownership rate in Indonesia is still relatively low compared to other countries in Southeast Asia. One of the contributing factors is the relatively high price and variety of specifications that often confuse consumers. To help people find a laptop that suits their needs and budget, this research proposes the implementation of a recommendation system based on Hybrid Recommendation. This system utilizes the Gradient Boosting Decision Tree (GBDT) algorithm from the LightGBM library and the indoBERT model for sentiment analysis of user comments on YouTube. The variables used include text_comment, product_id, user_id, and rating. By using this technology, it is expected to provide more personalized and accurate recommendations for users. The results show that this recommendation system is able to achieve Mean Absolute Error (MAE) of 0.3621 and Root Mean Square Error (RMSE) of 0.7115 while at NDCG of 0.4543 with text features while on features without text with MAE score of 0.5345 and RMSE 0.9122 while at NDCG of 0.3815. Keyword: Hybrid Recommendation, Gradient Boosting,, indoBERT, MAE, Laptop telah menjadi perangkat penting yang mendukung berbagai aspek kehidupan, mulai dari pendidikan hingga sektor strategis seperti bisnis, kesehatan, dan pariwisata. Namun, tingkat kepemilikan laptop di Indonesia masih tergolong rendah dibandingkan negara-negara lain di Asia Tenggara. Salah satu faktor penyebabnya adalah harga yang relatif mahal dan beragamnya spesifikasi yang sering membingungkan konsumen. Untuk membantu masyarakat menemukan laptop yang sesuai dengan kebutuhan dan anggaran, penelitian ini mengusulkan penerapan sistem rekomendasi berbasis Hybrid Recommendation. Sistem ini memanfaatkan algoritma Gradient Boosting Decision Tree (GBDT) dari library LightGBM dan model indoBERT untuk analisis sentimen komentar pengguna di YouTube.Variabel yang digunakan meliputi teks_komentar, produk_id, user_id, dan rating. Dengan menggunakan teknologi ini, diharapkan dapat memberikan rekomendasi yang lebih personal dan akurat bagi pengguna. Hasil penelitian menunjukkan bahwa sistem rekomendasi ini mampu mencapai Mean Absolute Error (MAE) sebesar 0.3621 dan Root Mean Square Error (RMSE) sebesar 0.7115 sedangkan pada NDCG sebesar 0.4543 dengan fitur teks sedangkan pada fitur tanpa teks dengan score MAE sebesar 0.5345 dan RMSE 0.9122 sedangkan pada NDCG sebesar 0.3815. Kata Kunci : Hybrid Recommendation, Gradient Boosting,, indoBERT, MAE, NDCG, RMSE
Actions (login required)
 |
View Item |
