PENGEMBANGAN CHATBOT MENGGUNAKAN BERT UNTUK LAYANAN BANTUAN TERKAIT PRODUK APPLE

KURNIAWAN, ADITIYA and Yusliani, Novi and Rachmatullah, Muhammad Naufal (2025) PENGEMBANGAN CHATBOT MENGGUNAKAN BERT UNTUK LAYANAN BANTUAN TERKAIT PRODUK APPLE. Undergraduate thesis, Sriwijaya University.

	Image RAMA_55201_09021182126028_cover.pdf - Cover Image Available under License Creative Commons Public Domain Dedication. Download (431kB)
	Text RAMA_55201_09021182126028.pdf - Accepted Version Restricted to Repository staff only Available under License Creative Commons Public Domain Dedication. Download (2MB) | Request a copy
	Text RAMA_55201_09021182126028_TURNITIN.pdf - Accepted Version Restricted to Repository staff only Available under License Creative Commons Public Domain Dedication. Download (4MB) | Request a copy
	Text RAMA_55201_09021182126028_0008118205_0001129204_01_front_ref.pdf - Accepted Version Available under License Creative Commons Public Domain Dedication. Download (719kB)
	Text RAMA_55201_09021182126028_0008118205_0001129204_02.pdf - Accepted Version Restricted to Repository staff only Available under License Creative Commons Public Domain Dedication. Download (419kB) | Request a copy
	Text RAMA_55201_09021182126028_0008118205_0001129204_03.pdf - Accepted Version Restricted to Repository staff only Available under License Creative Commons Public Domain Dedication. Download (401kB) | Request a copy
	Text RAMA_55201_09021182126028_0008118205_0001129204_04.pdf - Accepted Version Restricted to Repository staff only Available under License Creative Commons Public Domain Dedication. Download (774kB) | Request a copy
	Text RAMA_55201_09021182126028_0008118205_0001129204_05.pdf - Accepted Version Restricted to Repository staff only Available under License Creative Commons Public Domain Dedication. Download (578kB) | Request a copy
	Text RAMA_55201_09021182126028_0008118205_0001129204_06.pdf - Accepted Version Restricted to Repository staff only Available under License Creative Commons Public Domain Dedication. Download (186kB) | Request a copy
	Text RAMA_55201_09021182126028_0008118205_0001129204_07_ref.pdf - Bibliography Restricted to Repository staff only Available under License Creative Commons Public Domain Dedication. Download (198kB) | Request a copy
	Text RAMA_55201_09021182126028_0008118205_0001129204_08_lamp.pdf - Accepted Version Restricted to Repository staff only Available under License Creative Commons Public Domain Dedication. Download (360kB) | Request a copy

Abstract

Chatbots have become a crucial component in enhancing the efficiency of customer support services, including within the Apple product ecosystem. This study focuses on the development of a chatbot based on the pre-trained BERT (Bidirectional Encoder Representations from Transformers) model to provide relevant and fast responses to users. The dataset consists of customer conversations related to Apple products, which are processed using synonym-based data augmentation techniques to enrich input variation. The BERT model undergoes fine-tuning across six different scenarios involving combinations of learning rate and batch size. The best result is achieved with a learning rate of 4e-5 and a batch size of 8, yielding a testing accuracy of 95.45%. This approach demonstrates that leveraging BERT enables the chatbot to understand sentence context effectively. Furthermore, the data augmentation technique contributes significantly to improving the model’s generalization capability. The resulting chatbot is not only effective within the Apple ecosystem but also holds potential for implementation in other digital customer service platforms that require automated, fast, and accurate responses.

Item Type:	Thesis (Undergraduate)
Uncontrolled Keywords:	Chatbot, BERT, Deep Learning, Apple
Subjects:	Q Science > QA Mathematics > QA75-76.95 Calculating machines > QA75 Electronic computers. Computer science T Technology > T Technology (General) > T1-995 Technology (General)
Divisions:	09-Faculty of Computer Science > 55201-Informatics (S1)
Depositing User:	Aditiya Kurniawan
Date Deposited:	21 May 2025 05:38
Last Modified:	21 May 2025 05:38
URI:	http://repository.unsri.ac.id/id/eprint/173423

Actions (login required)

View Item