Analisis Sentimen dan Sistem Pendukung Keputusan Menginap di Hotel Menggunakan Metode CRISP-DM dan SAW

Abstract

Research combining sentiment analysis approaches with decision support systems in the hospitality industry is still very limited. This study aims to analyze hotel guest sentiment based on location, cleanliness, service, and value criteria using Cross Industry Standard Process for Data Mining (CRISP-DM) to identify five and ten famous words based on sentiment data and the sentiment classification model with the best performance. Furthermore, the results of identifying five and ten famous words based on sentiment data related to location, cleanliness, service, and value ratings were processed using the Simple Additive Weighting (SAW) method. The results of this study show that the performance of SVM algorithms using SMOTE operators has accuracy values (99.74%), precision values (100%), recall values (99.48%), f-measure values (99.74%), and AUC (100%). This shows that sentiment classification results using the SVM algorithm with SMOTE operators can be used as a sentiment data classification model with 30% training and 70% test data. This model is relevant to be used as a model for the classification process of 1680 review data obtained from the Tripadvisor website at Grand Zuri Malioboro Hotel Yogyakarta, Tentrem Hotel Yogyakarta, Grand Ambarukmo Hotel, El Royale Yogyakarta Malioboro, and Ceria Hotel @alun alun Yogyakarta. Specifically, aspects that concern guests relate to staff, breakfast, room, and service. In addition, the results of the implementation of the SAW method show that Grand Zuri Malioboro Hotel Yogyakarta ranks first with a total value of 0.907663917. Otherwise, Ceria Hotel @alun Yogyakarta Square ranks second with a total value of 0.897193878. Meanwhile, Hotel Tentrem Yogyakarta ranks third with a total value of 0.861265289. This shows that hotel guests who assess the criteria of location, cleanliness, service, and value by giving value weights as rated on the Tripadvisor website and provide importance weights with similar values, where location (0.20), cleanliness (0.25), service (0.30), value (0.25) then the alternative with the highest weight is Hotel Grand Zuri Malioboro Yogyakarta.

Perbandingan Metode ARAS dan EDAS dalam Menghasilkan Rekomendasi Layanan Akomodasi Hotel

“…Penelitian ini bertujuan menganalisis perbandingan metode EDAS dan ARAS dalam menghasilkan rekomendasi hotel terbaik berdasarkan kriteria pelayanan (services), sanitasi atau kebersihan (cleanliness), lokasi yang strategis (location), dan harga yang layak (value). Berdasarkan penerapan ROC, kriteria yang diprioritaskan ialah pelayanan dengan mempertimbangkan beberapa kajian yang menunjukkan bahwa performa bisnis layanan akomodasi dapat diukur dari kualitas pelayanan yang diberikan kepada konsumen [18], [19]. Selanjutnya, Juniardi et al menunjukkan bahwa sanitasi hotel memiliki pengaruh terhadap kepuasan tamu [20].…”

Analisis Multi-Objective Optimization on the Basis of Ratio Analysis (MOORA) Terhadap Data Top 10 Best Value Hotel

“…Penelitian ini merupakan pengembangan dari penelitian sebelumnya yang menggunakan metode Technique for Other Preference by Similarity for Ideal Solution (TOPSIS) dalam pemilihan hotel berdasarkan kriteria harga, jarak, tipe atau kelas, amenitas, dan rating layanan [13]. Selanjutnya, data ulasan tamu hotel diklasifikasi berdasarkan sentimen negatif dan positif menggunakan Cross-Industry Standard Process for Data Mining (CRISP-DM) untuk dianalisis lebih lanjut menggunakan metode Simple Additive Weighting (SAW) [14]. Dalam konteks penelitian ini, metode yang digunakan ialah MOORA melalui studi kasus top 10 best value hotel di Labuan Bajo, berdasarkan informasi yang tersedia di platform Tripadvisor.…”

Analisis Hasil Implementasi Multi-Attribute Utility Theory (MAUT) dalam Pengemasan Paket Wisata Tematik

“…Penelitian ini merupakan pengembangan dari beberapa kajian terdahulu yang berhubungan dengan sistem pendukung keputusan dan kepariwisataan. Singgalen menggunakan metode pendukung keputusan Simple Additive Weighting (SAW) dan pendekatan analisis sentimen melalui metode Cross Industry Standar Process for Data Mining (CRISP-DM) untuk pemilihan layanan akomodasi terbaik [12]. Selanjutnya, Singgalen menggunakan model pedukung keputusan Technique for Other Preference by Similarity to Ideal Solution (TOPSIS) untuk menentukan layanan akomodasi terbaik berdasarkan karakterisitk destinasi wisata Pulau Morotai [13].…”