ANALISIS SENTIMEN ULASAN E-COMMERCE SHOPEE TERKAIT FITUR COD DENGAN MENGGUNAKAN METODE SUPPORT VECTOR MECHINE DAN SELEKSI FITUR MUTUAL INFORMATION

NILA, SONIA and Alvi, Syahrini Utami (2025) ANALISIS SENTIMEN ULASAN E-COMMERCE SHOPEE TERKAIT FITUR COD DENGAN MENGGUNAKAN METODE SUPPORT VECTOR MECHINE DAN SELEKSI FITUR MUTUAL INFORMATION. Undergraduate thesis, Sriwijaya University.

	Image RAMA_55201_09021382126175_0022127804_Cover.jpg - Cover Image Available under License Creative Commons Public Domain Dedication. Download (1MB)
	Text RAMA_55201_09021382126175.pdf - Accepted Version Restricted to Repository staff only Available under License Creative Commons Public Domain Dedication. Download (2MB) | Request a copy
	Text RAMA_55201_09021382126175_TURNITIN.pdf - Accepted Version Restricted to Repository staff only Available under License Creative Commons Public Domain Dedication. Download (7MB) | Request a copy
	Text RAMA_55201_09021382126175_0022127804_01_front_ref.pdf - Accepted Version Available under License Creative Commons Public Domain Dedication. Download (834kB)
	Text RAMA_55201_09021382126175_0022127804_02.pdf - Accepted Version Restricted to Repository staff only Available under License Creative Commons Public Domain Dedication. Download (499kB) | Request a copy
	Text RAMA_55201_09021382126175_0022127804_03.pdf - Accepted Version Restricted to Repository staff only Available under License Creative Commons Public Domain Dedication. Download (397kB) | Request a copy
	Text RAMA_55201_09021382126175_0022127804_04.pdf - Accepted Version Restricted to Repository staff only Available under License Creative Commons Public Domain Dedication. Download (930kB) | Request a copy
	Text RAMA_55201_09021382126175_0022127804_05.pdf - Accepted Version Restricted to Repository staff only Available under License Creative Commons Public Domain Dedication. Download (366kB) | Request a copy
	Text RAMA_55201_09021382126175_0022127804-06_ref.pdf - Bibliography Restricted to Repository staff only Available under License Creative Commons Public Domain Dedication. Download (242kB) | Request a copy
	Text RAMA_55201_09021382126175_0022127804_07_lamp.pdf - Accepted Version Restricted to Repository staff only Available under License Creative Commons Public Domain Dedication. Download (246kB) | Request a copy

Abstract

The rapid development of e-commerce in Indonesia has driven significant changes in people's shopping patterns, with Shopee as one of the most popular platforms. One of the superior features offered is Cash on Delivery (COD), which provides a sense of security for users in making transactions. However, this feature also raises various responses that are reflected in user reviews of the COD feature.This study aims to analyze user review sentiments towards the COD feature on Shopee e-commerce. The method used is Support Vector Machine (SVM) with feature selection using Mutual Information and text representation using Sentence Transformers. The data used are 1887 user reviews taken from Kaggle. The results of the study showed that the SVM model without feature selection produced an accuracy of 63.33% and SVM using Mutual Information feature selection produced an accuracy of 53.33%. Testing was carried out by comparing the performance of the two models using the C regulation parameter, the test results showed that the value of C = 1 gave the best performance for both models. Based on this, it can be concluded that the SVM model without feature selection provides better performance than the model with Mutual Information. Implementation of Mutual Information as a feature selection technique that is considered relevant, but decreases model performance, possibly due to the loss of important features during the selection process. This indicates that feature selection needs to be done carefully, especially on small data. Keywords: Sentence Transformers, SVM, Mutual Information,COD

Item Type:	Thesis (Undergraduate)
Uncontrolled Keywords:	Sentence Transformers, Support Vector Mechine, Mutual Information, COD
Subjects:	T Technology > TJ Mechanical engineering and machinery > TJ1-1570 Mechanical engineering and machinery
Divisions:	09-Faculty of Computer Science > 55201-Informatics (S1)
Depositing User:	Nila Sonia
Date Deposited:	22 Jul 2025 08:53
Last Modified:	22 Jul 2025 08:53
URI:	http://repository.unsri.ac.id/id/eprint/179763

Actions (login required)

View Item