Ambient Intelligence Serious Game Pemilihan Destinasi Wisata Menggunakan Recommendation Engine Decentralized MCRS

Arif, Yunifa Miftachul (2022) Ambient Intelligence Serious Game Pemilihan Destinasi Wisata Menggunakan Recommendation Engine Decentralized MCRS. Doctoral thesis, Institut Teknologi Sepuluh Nopember Surabaya.

Text 07111760010007-Disertation.pdf - Accepted Version Restricted to Repository staff only until 1 April 2024. Download (8MB) | Request a copy

Abstract

Perjalanan wisata mempunyai tiga fase utama, yaitu before trip, during trip dan after trip. Diantara ke tiga fase tersebut, before trip merupakan fase yang mempunyai pengaruh besar terhadap keberhasilan kegiatan perjalanan wisata. Salah satu aktifitas penting dalam fase ini adalah tentang pemilihan destinasi wisata. Ketepatan dalam pemilihan destinasi wisata menjadi hal yang sangat penting karena akan berpengaruh terhadap rute perjalanan, jadwal perjalanan dan budget yang dikeluarkan selama kegiatan wisata. Karakteristik destinasi wisata yang bervariasi, destinasi baru yang belum pernah dikunjungi, serta kondisi lingkungan yang fluktuatif menjadi permasalahan tersendiri dalam pemilihan destinasi wisata. Oleh karena itu pihak-pihak yang berkepentingan yaitu wisatawan, travel agent maupun tour guide tentu harus memiliki pengetahuan dan keahlian dalam melakukan pemilihan tersebut. Media yang diajukan untuk menunjang pembelajaran tentang pemilihan destinasi wisata tersebut pada penelitian ini adalah serious game. Pada penelitian ini serious game didukung dengan teknologi ambient intelligent, sehingga diharapkan player mendapatkan respon pengetahuan dari system secara adaptif yang menyenangkan dengan data kondisi lingkungan yang sebenarnya.
Dalam rangka mendukung respon yang dihasilkan ambient intelligent serious game, maka diperlukan pengembangan recommendation engine sebagai pembangkit rekomendasi pemilihan destinasi wisata. Selanjutnya untuk menjamin keakuratan terhadap rekomendasi yang dihasilkan, maka penelitian ini menggunakan destinations attributes (DA), personal characteristics (PC), dan espektasi player sebagai masukan sistem yang merupakan faktor utama yang berpengaruh terhadap pemilihan destinasi wisata. Data DA terdiri beberapa variabel kriteria berdasarkan framework 6AsTD yaitu Attractions, Acessibility, Amenities, Avaliable Packages, Activities, Ancillary services. PC terdiri dari delapan variabel kriteria yaitu gender, age, job, hobby, motivations, marital status, origin, people in a group, educations, dan repetition. Sedangkan yang termasuk kriteria espektasi player antara lain, , meliputi cuaca, jumlah pengunjung, kelengkapan spot wisata, harga tiket, dan fasilitas umum. Untuk menjamin distribusi data PC dan DA antar node, maka konsep jaringan data sharing yang diusulkan bersifat distributed, decentralized dan secure. Selain itu metode sistem rekomendasi yang diusulkan juga disesuaikan dengan banyaknya variabel kriteria yang digunakan. Rekomendasi yang dihasilkan selanjutnya divisualisasikan dalam skenario serious game. Untuk menjawab beberapa persoalan tersebut, melalui penelitian ini diusulkan ambient intelligent serious game pemilihan destinasi wisata yang menggunakan Multi criteria Recommendation System (MCRS) berbasis pendekatan known and unknown rating (KUR) sebagai metode untuk menghasilkan rekomendasi berdasarkan variasi kriteria masukan. Sedangkan konsep data sharing yang diusulkan adalah menggunakan jaringan blockchain.
Dalam fase pengujian, recommendation engine dapat menghasilkan rekomendasi pemilihan destinasi wisata baik melalui pendekatan known rating maupun unknown rating. Hasil pengujian pendekatan known rating menggunakan metode MCRS menunjukkan bahwa system dapat menghasilkan accuracy rekomendasi tertinggi sebesar 0.743 melalui perhitungan pearson correlation-based similarity dan perengkingan worst case. Sedangkan hasil pengujian pendekatan unknown rating berdasarkan metode klasifikasi artificial neural network menunjukkan nilai akurasi tertinggi sebesar 0.660. Pada penelitian ini ethereum blockchain dapat bekerja mendukung decentralized data sharing dengan rata-rata waktu transaksi data 20.44. Berikutnya serious game dapat memvisualisasikan dan memilihkan rekomendasi untuk player melalui automatic scenario control dengan nilai accuracy sebesar 0.78.
===================================================================================================
The tour has three main phases, namely before trip, during trip and after trip. Among the three phases, the before trip is the phase that has a major influence on the success of travel activities. One of the important activities in this phase is the selection of tourist destinations. Accuracy in the selection of tourist destinations is very important because it will affect the travel route, travel schedule and budget spent during tourist activities. The varied characteristics of tourist destinations, new destinations that have never been visited, and fluctuating environmental conditions are separate problems in the selection of tourist destinations. Therefore, interested parties, namely tourists, travel agents and tour guides, must have knowledge and expertise in making these choices. The media proposed to support learning about the selection of tourist destinations in this study is a serious game. In this study, serious games are supported by ambient intelligent technology, so that players are expected to get a response from the system knowledge in an adaptive manner that is pleasant with data on actual environmental conditions.
To support the response generated by the ambient intelligent serious game, it is necessary to develop a recommendation engine as a generator of recommendations for selecting tourist destinations. Furthermore, to ensure the accuracy of the recommendations produced, this study uses destinations attributes (DA), personal characteristics (PC), and player expectations as system inputs which are the main factors that influence the selection of tourist destinations. DA data consists of several criteria variables based on the 6AsTD framework, namely Attractions, Accessibility, Amenities, Available packages, Activities, Ancillary services. The PC consists of eight criteria variables: gender, age, job, hobby, motivations, marital status, origin, people in a group, educations, dan repetition. Meanwhile, the criteria for player expectations include, among others, weather, number of visitors, completeness of tourist spots, ticket prices, and public facilities. To ensure the distribution of PC and rating DA data between nodes, the concept of a distributed, decentralized, and secure data-sharing network is proposed. In addition, the proposed recommendation system method is also adjusted to the number of criteria variables used. The resulting recommendations are then visualized in a serious game scenario. To answer some of these problems, this research proposes an ambient intelligent serious game for selecting tourist destinations using the Multi-Criteria Recommendation System (MCRS) based on the known and unknown rating (KUR) approach to generate recommendations based on a variety of input criteria. While the concept of data sharing proposed is to use a blockchain network.
The recommendation engine can generate recommendations for selecting tourist destinations in the testing phase through known and unknown rating approaches. The results of testing the known rating approach using the MCRS method show that the system can produce the highest recommendation accuracy of 0.743 through calculating Pearson correlation-based similarity and worst-case ranking. While the results of testing the unknown rating approach based on the artificial neural network classification method showed the highest accuracy value of 0.660. In this study, the ethereum blockchain can support decentralized data sharing with an average data transaction time of 20.44. Next, serious games can visualize and choose recommendations for players through automatic scenario control with an accuracy value of 0.78.

Item Type:	Thesis (Doctoral)
Uncontrolled Keywords:	perjalanan wisata, rekomendasi, serious game, destination attributes, personal characteristics, recommendation engine, MCRS, blockchain
Subjects:	L Education > LB Theory and practice of education > LB1029.S53 Educational games. Simulation methods
Divisions:	Faculty of Intelligent Electrical and Informatics Technology (ELECTICS) > Electrical Engineering > 20001-(S3) PhD Thesis
Depositing User:	Yunifa Miftachul Arif
Date Deposited:	14 Feb 2022 06:31
Last Modified:	14 Feb 2022 06:31
URI:	http://repository.its.ac.id/id/eprint/93958

Actions (login required)

View Item