Article Per Year (5 Year)
p-Index From 2021 - 2026
0.23
P-Index
This Author published in this journals
All Journal Indonesian Journal of Business Intelligence (IJUBI)
Laga Kusuma Laga
Sekolah Tinggi Teknologi Ronggolawe
Computer Science & IT Decision Sciences, Operations Research & Management

Published : 1 Documents Claim Missing Document
Claim Missing Document
Check
Articles

Found 1 Documents
Search

 1 
MODEL RANDOM FOREST PREDIKSI KUNJUNGAN ULANG BERDASARKAN POLA PENGELUARAN Laga Kusuma Laga; Joko Handoyo
Indonesian Journal of Business Intelligence (IJUBI) Vol 9 No 1 (2026): Indonesian Journal of Business Intelligence (IJUBI)
Publisher : Universitas Alma Ata

Show Abstract | Download Original | Original Source | Check in Google Scholar

Abstract

Noyo Gimbal View Blora tourism has experienced a significant decline in visits, from 500,552 (2024) to a projected 300,000 (2025), with revenue dropping from IDR 3 billion to IDR 1.5 billion. This research aims to build a predictive Model to identify potential tourist disloyalty using the Random Forest algorithm based on spending patterns and service satisfaction. The research method utilizes 3,000 visit records (June 2023–December 2025) from ticket transactions and Google Maps reviews. The target variable is loyalty status derived from revisit history. Modeling employs Random Forest with Hyperparameter optimization through GridSearchCV and 5-fold cross-validation, along with Model interpretation using Feature Importance and SHAP. Results show the Model achieves 87.3% Accuracy, 83.5% Precision, 78.2% Recall, 80.8% F1-Score, and 0.91 AUC. The three most influential features are review sentiment score (0.26), spending variation (0.20), and number of facility complaints (0.17). Characteristics of at-risk tourists include negative sentiment, at least 2 complaints, rating ≤ 2 stars, spending < IDR 20,000, high spending variation, and high Recency days. In conclusion, service satisfaction factors are more dominant than spending patterns in determining tourist loyalty. This Model can be implemented as an early warning system to design intervention strategies for increasing tourist retention.
Co-Authors