Analisis Aspek Dan Sentimen Ulasan Produk Terhadap Keputusan Pembelian Menggunakan Spect-Based Sentiment Analysis Dan Conjoint Analysis

Sulistio, Hansen Gunawan (2025) Analisis Aspek Dan Sentimen Ulasan Produk Terhadap Keputusan Pembelian Menggunakan Spect-Based Sentiment Analysis Dan Conjoint Analysis. Masters thesis, Institut Teknologi Sepuluh Nopember.

Text 6032231014-Master_Thesis.pdf - Accepted Version Restricted to Repository staff only Download (3MB) | Request a copy

Abstract

Penelitian ini menganalisis pengaruh aspek dan sentimen ulasan produk terhadap Keputusan pembelian konsumen pada platform marketplace Shopee dengan objek produk smartphone. Metode yang digunakan adalah Aspect-Based Sentiment Analysis (ABSA) berbasis Support Vector Machine (SVM) untuk mengidentifikasi aspek dan sentimen ulasan, yang selanjutnya diintegrasikan dengan Conjoint Analysis untuk mengukur preferensi konsumen. Hasil penelitian menunjukkan bahwa model ABSA memiliki kinerja sangat baik dengan akurasi klasifikasi aspek sebesar 97% dan sentimen sebesar 98%. Analisis Conjoint menunjukkan bahwa aspek kualitas memiliki tingkat kepentingan relatif tertinggi sebesar 27,97%, diikuti oleh harga (25,41%), pelayanan (23,66%), kemasan (17,38%), dan pengiriman (5,57%). Sentimen positif pada aspek kualitas memberikan kontribusi utilitas terbesar terhadap preferensi konsumen, sedangkan sentimen negatif pada aspek pelayanan menurunkan nilai utilitas secara signifikan. Temuan ini menegaskan bahwa kualitas dan harga merupakan faktor dominan dalam keputusan pembelian konsumen di marketplace.
==================================================================================================================================
This study analyzes the influence of product review aspects and sentiments on consumer purchasing decisions on the Shopee marketplace platform with smartphone products as the object. The method used is Aspect-Based Sentiment Analysis (ABSA) based on Support Vector Machine (SVM) to identify review aspects and sentiments, which are then integrated with Conjoint Analysis to measure consumer preferences. The results show that the ABSA model has excellent performance with an aspect classification accuracy of 97% and a sentiment classification accuracy of 98%. Conjoint analysis shows that the quality aspect has the highest relative importance level of 27.97%, followed by price (25.41%), service (23.66%), packaging (17.38%), and delivery (5.57%). Positive sentiment on the quality aspect contributes the largest utility to consumer preferences, while negative sentiment on the service aspect significantly decreases the utility value. These findings confirm that quality and price are dominant factors in consumer purchasing decisions in the marketplace.

Item Type:	Thesis (Masters)
Uncontrolled Keywords:	Aspect-Based Sentiment Analysis (ABSA), Conjoint Analysis, Sentimen Konsumen, Keputusan Pembelian, Pengaruh Aspek Terhadap Pembelian Aspect-Based Sentiment Analysis (ABSA), Conjoint Analysis, Consumer Sentiment, Repeat Purchase Decision, Aspect Influence on Purchase
Subjects:	T Technology > T Technology (General) > T57.5 Data Processing T Technology > T Technology (General) > T58.5 Information technology. IT--Auditing T Technology > T Technology (General) > T58.62 Decision support systems
Divisions:	Interdisciplinary School of Management and Technology (SIMT) > 61101-Master of Technology Management (MMT)
Depositing User:	Hansen Gunawan Sulistio
Date Deposited:	28 Jan 2026 01:48
Last Modified:	28 Jan 2026 01:48
URI:	http://repository.its.ac.id/id/eprint/130720

Actions (login required)

View Item