ANALISIS SENTIMEN TERKAIT LAYANAN GOFOOD DAN GRABFOOD PADA MEDIA SOSIAL TWITTER MENGGUNAKAN ALGORITMA SUPPORT VECTOR MACHINE (SVM)
Total View This Week0
Total View This Week0
Institusion
Universitas Pembangunan Nasional Veteran Jakarta
Author
Realdy Agsar Dwi Anggoro, .
Subject
Q Science (General)
Datestamp
2021-12-21 07:46:12
Abstract :
The popularity of twitter social media in Indonesia has led to the rapid dissemination of information through social media this virtual world. The information is in the form of an opinion on a service that can be analyzed for sentiment in order to understand the meaning and emotion contained in the sentence of opinion. Research aims to find out how the public opinion about gofood and grabfood services on social media twitter with the search keywords gofood and grabfood taken for 6 days on February 23 until with 27 February 2021 and 5 March 2021. The data will then be labeled as positive class and negative class and classified by the Support Vector Machine algorithm. Model evaluation results The first one has an accuracy of 80.18%, recall of 100%, and specificity of 14%. Because the data on this study is not balanced, then in the second model the undersampling method is applied to the training data for overcome this. The results of the evaluation of the second model have an accuracy of 79.26%, a recall of 86.23%, and a specificity of 56%.
The popularity of twitter social media in Indonesia has led to the rapid dissemination of information through social media this virtual world. The information is in the form of an opinion on a service that can be analyzed for sentiment in order to understand the meaning and emotion contained in the sentence of opinion. Research aims to find out how the public opinion about gofood and grabfood services on social media twitter with the search keywords gofood and grabfood taken for 6 days on February 23 until with 27 February 2021 and 5 March 2021. The data will then be labeled as positive class and negative class and classified by the Support Vector Machine algorithm. Model evaluation results The first one has an accuracy of 80.18%, recall of 100%, and specificity of 14%. Because the data on this study is not balanced, then in the second model the undersampling method is applied to the training data for overcome this. The results of the evaluation of the second model have an accuracy of 79.26%, a recall of 86.23%, and a specificity of 56%.
Download
book
BibTex
Latex, Jabref
cloud_download
Original Resource
url resource Institution
