Analisis Ulasan Google Maps Wisata Bahari Lamongan Berbasis SERVQUAL dengan Algoritma Classifier Chains

Analysis of Google Maps Reviews for Wisata Bahari Lamongan Based on SERVQUAL Using the Classifier Chains Algorithm

Ulasan Google Maps merupakan sumber informasi untuk mengevaluasi kualitas layanan destinasi wisata berdasarkan pengalaman pengunjung. Penelitian ini bertujuan menganalisis sentimen ulasan Wisata Bahari Lamongan (WBL), memetakan dimensi SERVQUAL, mengevaluasi klasifikasi multi-label menggunakan Classifier Chains, serta memberikan rekomendasi peningkatan layanan melalui Importance Performance Analysis (IPA) yang diperkuat Latent Dirichlet Allocation (LDA). Penelitian ini menggunakan metode Knowledge Discovery in Databases (KDD) terhadap 2.485 ulasan valid hasil preprocessing. Analisis sentimen dilakukan menggunakan model w11wo/indonesian-roberta-base-sentiment-classifier, pemetaan SERVQUAL menggunakan semantic similarity, dan klasifikasi multi-label menggunakan Classifier Chains berbasis Logistic Regression. Evaluasi menggunakan precision, recall, dan F1-score. Hasil penelitian menunjukkan Micro F1-score sebesar 0,76, Macro F1-score sebesar 0,66, dan Weighted F1-score sebesar 0,75. Analisis IPA menunjukkan bahwa pada sentimen negatif, dimensi Tangibles, Reliability, dan Responsiveness menjadi prioritas perbaikan (Kuadran I), yang diperkuat melalui analisis LDA. Hasil penelitian ini diharapkan menjadi dasar bagi pengelola WBL dalam menentukan prioritas peningkatan kualitas layanan.

Google Maps reviews are a source of information for evaluating the service quality of tourist destinations based on visitors' experiences. This study aims to analyze the sentiment of Wisata Bahari Lamongan (WBL) reviews, map the SERVQUAL dimensions, evaluate multi-label classification using the Classifier Chains algorithm, and provide service quality improvement recommendations through Importance Performance Analysis (IPA) strengthened by Latent Dirichlet Allocation (LDA). This study employed the Knowledge Discovery in Databases (KDD) method on 2,485 valid reviews obtained after the preprocessing stage. Sentiment analysis was conducted using the w11wo/indonesian-roberta-base-sentiment-classifier model, SERVQUAL mapping was performed using semantic similarity, and multi-label classification was carried out using Classifier Chains based on Logistic Regression. The evaluation used precision, recall, and F1-score metrics. The results showed a Micro F1-score of 0.76, a Macro F1-score of 0.66, and a Weighted F1-score of 0.75. The IPA analysis showed that, in the negative sentiment data, the Tangibles, Reliability, and Responsiveness dimensions became priorities for improvement (Quadrant I), which was strengthened through LDA analysis. The results of this study are expected to serve as a basis for the management of Wisata Bahari Lamongan in determining priorities for service quality improvement.