Skip to main content Skip to main navigation menu Skip to site footer

Disagregasi Preferensi Konsumen menggunakan Metode Bayesian Linear Model

Authors
  • Chrisentiana Talithapurwa Jasindatama Departemen Teknik Mesin dan Industri, Fakultas Teknik, Universitas Gadjah Mada, Jl. Grafika No.2, Senolowo, Sinduadi, Kec. Mlati, Kabupaten Sleman, Daerah Istimewa Yogyakarta 55281, Indonesia
  • Andi Sudiarso Departemen Teknik Mesin dan Industri, Fakultas Teknik, Universitas Gadjah Mada, Jl. Grafika No.2, Senolowo, Sinduadi, Kec. Mlati, Kabupaten Sleman, Daerah Istimewa Yogyakarta 55281, Indonesia
  • Shi-Woei Lin Departement of Industrial Management, National Taiwan University of Science and Technology, No. 43, Section 4, Keelung Rd, Da’an District, Taipei City, Taiwan
Issue       Vol 7 No 1 (2024): Talenta Conference Series: Energy and Engineering (EE)
Section       Articles
Section      

Copyright (c) 2024 Talenta Conference Series: Energy and Engineering (EE)

Creative Commons License

This work is licensed under a Creative Commons Attribution-NonCommercial-NoDerivatives 4.0 International License.
Galley      
DOI: https://doi.org/10.32734/ee.v7i1.2170
Keywords: Bayesian Linear Model Pengambilan Keputusan Disagregasi Preferensi Ulasan Konsumen Decision Making Preference Disaggregation Customer Review
Published 2024-10-22

Abstract

Pemahaman terhadap preferensi dan perilaku konsumen sangat penting dalam berbagai rangkaian proses bisnis. Ulasan produk  menjadi sumber informasi berharga untuk pengambilan keputusan dalam pemasaran, desain produk, dan segmentasi pasar. Pendekatan melalui multiple attribute decision-aiding (MADA) banyak diterapkan dalam konteks disagregasi preferensi. Khususnya dengan menggunakan regresi linear bayesian untuk mengektraksi bobot dari aspek ulasan pelanggan. Penelitian ini menerapkan regresi linear bayesian yang berfokus pada ekstraksi bobot aspek yang memengaruhi penilaian peringkat konsumen untuk pengambilan keputusan bisnis dengan mempertimbangkan data yang hilang atau bernilai nol (0). Metode substitusi rerata yang digunakan untuk menyelesaikan masalah data yang hilang berhasil mengungsssguli model dengan data hilang dengan performa untuk menjelaskan keputusan penilaian peringkat produk. Bobot yang didapatkan dapat digunakan oleh pelaku bisnis dalam pengambilan keputusan seperti memfokuskan peningkatan kualitas produk dari aspek produk berdasarkan hasil matriks performa-kepentingan.

Gaining insight into consumer preferences and behavior is crucial in multiple corporate operations. Product reviews are an invaluable resource for making decisions in marketing, product design, and market segmentation. Therefore, a robust analysis is required to effectively utilize the available information. The use of multiple attribute decision-aiding (MADA) is commonly employed in the field of preference disaggregation. Specifically, Bayesian linear regression has been employed in research to derive weights for individual aspects of customer reviews. This study examines the application of Bayesian linear regression to extract aspect weights that impact consumer ratings for corporate decision-making, taking into account missing data or zero values (0). The mean substitution strategy effectively outperforms the model with missing data in accurately explaining consumer product rating decisions. The derived weights can be utilized by business professionals to inform decision-making, specifically in prioritizing efforts to enhance product quality based on the outcomes of the performance-importance matrix.