Peramalan Penjualan Produk Mobil Honda Hr -V Berdasarkan Data Historis Dan Sentimen Opini Publik

Roshafara, Fauziah (2022) Peramalan Penjualan Produk Mobil Honda Hr -V Berdasarkan Data Historis Dan Sentimen Opini Publik. Masters thesis, Institut Teknologi Sepuluh Nopember Surabaya.

Text 6003202014-Master_Thesis.pdf - Accepted Version Restricted to Repository staff only until 1 November 2024. Download (4MB) | Request a copy

Abstract

Perusahaan menggunakan peramalan penjualan produk sebagai bahan pertimbangan untuk memperkirakan penjualandi masa depan sehinggadapat dibuat keputusan untukmelakukan strategi pemasaran. Semakin berkembangnya teknologi seperti saat ini, opini publik terhadap suatu produk mampu mempengaruhi keputusan pembelian oleh konsumen lain.Sehingga sentimen dari opinimasyarakat terhadap suatu produk menjadi penting sebagai bahan pertimbangan dalam melakukan peramalan jumlah penjualan. Penelitian ini dilakukan untuk mengetahui pengaruh sentimen opini publik terhadap hasil akurasi peramalan jumlahpenjualan produk mobil menggunakan metode support vector regression.Data historis berupa jumlah penjualan mobil honda HR-V diperoleh melalui situs resmi GAIKINDOdaribulan Januari 2015sampai Januari2022. Sedangkan data opini publikdiperoleh melalui twitter berupa tweet dengan kata kunci “beli mobil honda” dan “jual mobil honda”yang dikumpulkan sejak Januari 2015 sampai dengan Januari 2022. Sebelum melakukan pemodelan dilakukan preprocessing teks terlebih dahulu, kemudian menghitung polaritas skor sentimen twittermenggunakan vader sentiment.Input untuk data historis ditentukan melalui lagyang signifikan pada plot PACF. Hasil penelitian ini menunjukkan peramalan penjualan mobil honda HR-V dengan adanya tambahan data berupa sentimenopini publik sebagai prediktor menurunkan nilai MAPE sebesar 1,137 jika dibandingkan hanya menggunakan data historis saja
=====================================================================================================================================
Companies using product sales forecasting as a consideration for estimating future sales so the decisions can be made to carry out marketing strategies. With the development of technology in this era, public opinion on a product is able to influence purchasing decisions another consumer. Therefore the sentiment of public opinion on a product becomes important as a consideration in forecasting the number of sales. This study was conducted to determine the effect of public opinion sentiment on the accuracy offorecasting the number of car product sales using the support vector regression method. The number of sales of Honda HR-V cars as historical data was obtained through the official website of GAIKINDO from January 2015 to January 2022. While the public opinion data was obtained from twitter by tweets with the keywords "buy a honda cars" and "sell a honda cars" which were collected since January 2015 to January 2022. Before doing the modeling, preprocessing text is done first, then calculating the polarity of opinionsentiment score using vader sentiment. The input for historical data is determined by the significant lag on the PACF plot. The results of this study indicate that the Honda HR-V sales forecast with additional data of public opinion sentiment as apredictor can decrease the MAPE value of 1.137 compared to using only historical data

Item Type:	Thesis (Masters)
Additional Information:	RTSt 519.535 Ros p-1 2022
Uncontrolled Keywords:	Analisis Sentimen, Peramalan, Support Vector Regression(SVR), VADER; Sentiment Analysis, Forecasting, Support Vector Regression(SVR), VADER determine
Subjects:	Q Science > QA Mathematics > QA280 Box-Jenkins forecasting
Divisions:	Faculty of Science and Data Analytics (SCIENTICS) > Statistics > 49101-(S2) Master Thesis
Depositing User:	EKO BUDI RAHARJO
Date Deposited:	28 Mar 2023 03:46
Last Modified:	28 Mar 2023 03:46
URI:	http://repository.its.ac.id/id/eprint/97817

Actions (login required)

View Item