Claim Missing Document
Check
Articles

Found 1 Documents
Search
Journal : Journal of Embedded Systems, Security and Intelligent Systems

Sentiment Analysis on Tiktok Application Reviews Using Natural Language Processing Approach Abdul Majid; Dian Nugraha; Faisal Dharma Adhinata
Journal of Embedded Systems, Security and Intelligent Systems Vol. 4 No. 1 (2023): Vol 4, No 1 (2023): May 2023
Publisher : Program Studi Teknik Komputer

Show Abstract | Download Original | Original Source | Check in Google Scholar

Abstract

Technology today is very developed, there are so many media that can be used to communicate, these media are very easy to use by connecting to the internet network. Research on the sentiment of this analysis can still be relatively small and new. The rapid development of technology today makes it very easy for humans to communicate with one of the modern technologies, namely smartphones. The initial stage of this research begins with the review to be analyzed, then continues with the collection of review data. Conducted on reviews that have been collected with and without an NLP approach resulting in 2 datasets, with an NLP approach and datasets without an NLP approach. The first step is to identify the problem with the research object. It then looked for related literature studies from both journals and review proceedings used as many as 1000 reviews, which have been labeled by 5 correspondents and resulted in positive reviews and negative reviews. The review is used as a dataset, then pre-processed with an NLP approach. Classification using the NLP approach got an accuracy of 76.92%, a precision of 80.00% and a recall of 74.07%, while without NLP it only got an accuracy of 69.23%, a precision of 80.00% and a recall of 64.52% At the preprocessing stage, the stemming feature, and stopword removal features were applied to each review. Word normalizer to handle variations in writing words that have the same meaning to be counted as a single term Furthermore, a stopword removal process is carried out to remove the stopword from the review.