Abstract:
Twitter is one of the social media that people choosen to express their opinion. Opinion
is a statement from the public based on their problem which distribute through social
media. The government has legitimate the Job Creation Law No. 11 2020 which reap the
criticism from society elements. The aims of this study are to find out the public perception
to validation of the Job Creation also, to determine the performance of the Support Vector
Machine to obtaining the optimal hyperplane that gives distance or separator between
the two classes and classifying Naïve Bayes in determining to classify simple
probabilities. Data used in this study was collected using twitter scraper, in the form of
data tweets with the keywords #omnibuslawand #UUCiptaKerja in the month of October
2020. The result of the evaluation that used confusion matrix showed that Support Vector
Machine method produced 82% accuracy level which is better than the Naive Bayes
Classifier which produced 80% accuracy. Society was not completely against the
legalization of the Job Creation Law, as there were a few perspective of the society that
agreed towards the clauses that was considered profitable for the society such as lay off
policy, economic growth and investation, work field creation and simplifying the process
of business licensing.
