Penerapan Universal Sentence Encoder (Use) Untuk Location-Aware Chatbot Pada Aplikasi Layanan Wisata Kebun Binatang

Cikanindi, Vania (2022) Penerapan Universal Sentence Encoder (Use) Untuk Location-Aware Chatbot Pada Aplikasi Layanan Wisata Kebun Binatang. Other thesis, Institut Teknologi Sepuluh Nopember.

Text 05111740000193-Undergraduate_Thesis.pdf - Accepted Version Restricted to Repository staff only until 1 April 2024. Download (2MB) | Request a copy

Abstract

Chatbot merupakan program yang mampu melakukan percakapan dengan penggunanya baik melalui teks maupun suara sehingga menghasilkan informasi dan atau melakukan tugas tertentu. Kebun binatang merupakan salah satu objek wisata rekreasi edukasi yang terdampak pandemi Covid-19. Untuk menghindari penularan Covid-19, salah satu kebun binatang di Jawa Barat memberlakukan beberapa kebijakan, salah satunya adalah mengurangi jumlah pekerja. Dengan pengurangan pekerja tersebut, pelayanan pemberian informasi menjadi tidak optimal sehingga pengunjung kebun binatang tidak mampu mendapatkan pengalaman berwisata secara maksimal. Dalam penelitian ini, teknologi GPS dan Universal Sentence Encoder (USE) digunakan untuk membuat sebuah location-aware chatbot untuk aplikasi layanan wisata kebun binatang. Universal Sentence Encoder digunakan untuk membuat vektor embedding dari kalimat pertanyaan pengunjung dan data pada brainfile yang kemudian dicocokan menggunakan perhitungan cosine similarity. Teknologi GPS dan geofencing digunakan untuk mengetahui lokasi pengunjung sehingga memungkinkan chatbot untuk memberikan informasi terkait koleksi kebun binatang serta mengirimkan informasi mengenai fasilitas terdekat. Aplikasi ini diuji menggunakan ponsel Android untuk mengetahui fungsionalitas, kerelevanan jawaban, dan ketergunaan aplikasi. Hasil dari pengujian fungsionalitas sistem menunjukkan bahwa sistem dapat berjalan sesuai dengan rancangan sistem. Pengujian kerelevanan jawaban menunjukkan bahwa chatbot menghasilkan kerelevanan sebesar 78%, akurasi sebesar 80%, presisi 86.7%, serta recall 90.7%. Hasil dari pengujian System Usability Scale (SUS) menunjukkan bahwa aplikasi ini memiliki ketergunaan yang dapat diterima oleh pengguna.
=====================================================================================================
Chatbot is a program that able to do conversations with users either through text or voice to obtain information and or perform certain tasks. Zoo is one of the tourist recreational attractions that affected by the Covid-19 pandemic. To avoid the transmission of Covid-19, the zoo management has implemented several policies, one of which was to reduce the number of workers. With the reduction of workers, the service of providing information became not optimal so the zoo visitors are unable to get the full experience and the tourists' interest in visiting the zoo is reduced. In this study, GPS technology and Universal Sentence Encoder (USE) were used to create a location-based chatbot for the zoo tourism application. Universal Sentence Encoder were used to create embedding vectors from question sentence and data in the brainfile which is then matched by using cosine similarity calculations. GPS and geofencing technology are used to determine the location of the visitor so the chatbot could provides information related to the zoo collections and send information about nearby facilities. This application was tested using an Android phone to test the functionality, relevance of the answers, and user ratings of the usability of the application. The results of the system functionality test indicate that the system could run according to the system design. The answer relevance test shows that the chatbot produces 78% relevance of answer, 80% of accuracy, 86.7% of precision, and 90.7% of recall. The results of System Usability Scale (SUS) show that this application is acceptable.

Item Type:	Thesis (Other)
Uncontrolled Keywords:	Chatbot, Geofencing, Location-Aware Applications, Universal Sentence Encoder, Chatbot, Geofencing, Location-Aware Applications, Universal Sentence Encoder
Subjects:	Q Science > QA Mathematics > QA336 Artificial Intelligence T Technology > T Technology (General) > T58.6 Management information systems
Divisions:	Faculty of Intelligent Electrical and Informatics Technology (ELECTICS) > Informatics Engineering > 55201-(S1) Undergraduate Thesis
Depositing User:	Vania Cikanindi
Date Deposited:	16 Feb 2022 15:09
Last Modified:	31 Oct 2022 02:37
URI:	http://repository.its.ac.id/id/eprint/94474

Actions (login required)

View Item