Bulletin of Information Technology (BIT)
Vol 3 No 2: Juni 2022

A Implementation of Text Mining In Sentiment Analysis of Shopee Indonesia Using SVM

Kusdarnowo Hantoro (Universitas Bhayangkara Jaya)
Dwipa Handayani (Universitas Bahayangkara Jaya)
Siti Setiawati (Universitas Bahayangkara Jaya)



Article Info

Publish Date
30 Jun 2022

Abstract

Many online shopping users convey an assessment of a product through status comments. Comments indicate a non-standard form of expression. Nowadays, user comments are increasing rapidly, which makes data management difficult. Today's society has so many choices; most of them are used for preference as a final recommendation. The top item shows the preferences of the available items, recommended based on future predictability. The predictive rating of some items is used by the user to recommend the item to other users. Sentiment analysis can recommend items of choice for online shopping in various fields of application, especially in e-commerce. Sentiment analysis is used to identify opinions, ideas, or thoughts from online media. Shopee is one of the largest online business platforms. Classification Sentiment analysis towards Shopee based on the Support Vector Machine (SVM) was used in this study on the Shopee application using 990 training data and 110 test data. From the test data, 28 data entered the negative class and the remaining 82 data entered the positive class and resulted in an accuracy rate of 80.90%, meaning that from 110 assessments there were 89 assessments classified exactly in the sentiment class.

Copyrights © 2022






Journal Info

Abbrev

BIT

Publisher

Subject

Computer Science & IT

Description

Jurnal Bulletin of Information Technology (BIT) memuat tentang artikel hasil penelitian dan kajian konseptual bidang teknik informatika, ilmu komputer dan sistem informasi. Topik utama yang diterbitkan mencakup:berisi kajian ilmiah informatika tentang : Sistem Pendukung Keputusan Sistem Pakar Sistem ...