Genetic Algorithm Optimization of Hybrid LSTM-AutoEncoder in Tourism Recommendation System - Dalam bentuk pengganti sidang - Artikel Jurnal

BAYU SURYA DHARMA SANJAYA

Informasi Umum

Kode

24.04.5390

Klasifikasi

005.7 - Data in Computer Systems

Jenis

Karya Ilmiah - Skripsi (S1) - Reference

Subjek

Data Science

Dilihat

88 kali

Informasi Lainnya

Abstraksi

The tourism industry has rapid growth and has become one of the world's leading economic industries in recent years due to advances in information technology, such as the internet and social media. However, the overwhelming amount of information often makes it difficult for travelers to decide on their preferred travel destination. To address these issues, this research proposes a tourism recommendation system that combines Content-Based Filtering and Hybrid LSTM-AE, which is optimized using Genetic Algorithm (GA). There is no research that has developed a recommendation system using a combination of these methods and optimized using GA. So that this research can contribute to providing personalized recommendations and higher accuracy. The dataset consists of 9,504 ratings collected from the Ministry of Tourism and Creative Economy, Twitter, and web sources. The system was able to achieve a rating prediction accuracy of 96.82% by applying SMOTE to handle data imbalance and implementing a GA approach to the Hybrid LSTM-AE model. Accuracy has increased by 18.7% from the baseline model without using SMOTE and optimization. These results underscore that a strong integration between natural language processing and genetically optimized deep learning provides more accurate recommendations.<br /> Keywords: Recommendation System; Content-Based Filtering; Auto Encoder; LSTM; Classification;