Chatbot Telegram untuk Rekomendasi Pariwisata Daerah Semarang Menggunakan Framework Rasa

Abid Ridlo  Alvinnajmi; R. Soelistijadi; Saefurrahman Saefurrahman

Authors

Abid Ridlo Alvinnajmi Universitas Stikubank Semarang
R. Soelistijadi Universitas Stikubank Semarang
Saefurrahman Universitas Stikubank Semarang

Keywords:

Tourism, Chatbot, Machine Learning, RASA Open Source, Semarang Regional Tourism

Abstract

Tourism is a leading industry in various countries because it can improve the country's economy. One city that has potential for tourism is Semarang, because this city has unique culture and beautiful nature. However, there are too many sources of information regarding tourist attractions such as the city of Semarang, which makes tourists sometimes confused about determining tourist destinations that suit their wishes at that time. Therefore, we need a system that can provide recommendations for tourism spots in a destination city. This research aims to develop a Question Answering System or digital question and answer system using a chatbot (ChatterBot). Chatbots are used as information service providers that can make it easier for tourists who are looking for information about tourist attractions. chat bot-based information service systems can work 24 hours or throughout the day, reducing the intensity of direct physical contact with officers and saving operational costs. The chatbot implementation is built on a Machine Learning Framework using RASA Open Source with the Python programming language. Basic knowledge of the chatbot system is drilled based on the FAQ (Frequently Asking Questions) dataset with tourism research objects in the Semarang area. The evaluation results and system performance based on data testing obtained a model accuracy level of 0.91. Furthermore, the weighted average value in the ConfusionMatrix produces a precision of 0.97, recall of 0.94, and an F1 score of 0.95. Training and processing models locally.

References

Rif’ah, S. (2022). Optimalisasi Wisata Halal Di Pantura Lamongan Sebagai Upaya Pemulihan Ekonomi Di Era New Normal. Al-Musthofa: Journal of Sharia Economics, 5(2), 54-69..

Balai Pelestarian Cagar Budaya, “Perpanjangan Penutupan Candi Prambanan,” 30 Juni, 2021. https://bpcbdiy.kemdikbud.go.id/berita-perpanjanganpenutupan-candi-prambanan (accessed Mar. 02, 2022).

B. P. Statistik, “Perkembangan Pariwisata Dan Transportasi Nasional,” Jakarta Badan Pus. Stat., no. 04, pp. 1-20, 2021.

Wonderful Indonesia, Trend Pariwisata 2021. Kemenparekraf Baparekra, 2020.

Simanjuntak, M. B., Lustyantie, N., & Iskandar, I. (2022). Pembelajaran Berbasis Telegram Group dan Microsoft Team di Kelas Bahasa Inggris (Penilaian berbasis Persepsi Siswa). Jurnal Pendidikan Tambusai, 6(2), 11114-11119.

Lenardo, G. C., & Irawan, Y. (2020). Pemanfaatan Bot Telegram sebagai Media Informasi Akademik di STMIK Hang Tuah Pekanbaru. JTIM: Jurnal Teknologi Informasi dan Multimedia, 1(4), 351-357.

RASA, “Model Configuration,” 2019. https://rasa.com/docs/rasa/model-configuration/ (accessed Jan. 22, 2022).

M. Grandini, E. Bagli, and G. Visani, “Metrics for Multi-Class Classification: an Overview,” pp. 1–17, 2020, [Online]. Available: http://arxiv.org/abs/2008.05756.

M. Khalusova, “Machine Learning Model Evaluation Metrics Part 2: Multi-Class Classification,” 17 April, 2019. https://www.mariakhalusova.com/posts/2019-04-17-ml-modelevaluation-metrics-p2/ (accessed Mar. 02, 2022).