PERBANDINGAN ALGORITMA K-NEAREST NEIGHBOR,RANDOM FOREST, DAN LOGISTIC REGRESSIONTERHADAP KATEGORI REVIEW PADA MARKETPLACEXYZ

• Main Author: Aria Nanda Herdiawan, (Author)
• Format: Book
• Published: 2023-12-20.
• Subjects: Thesis; NonPeerReviewed
• Online Access: Link Metadata

Summary

With the rapid development of technology in the world, it can make doing everything easier. One of them is shopping online. One platform that can be used for online shopping is a marketplace. Many reviews are given by buyers when completing transactions in marketplace. So classification is needed to distinguish categories of buyer reviews, besides that we can know the difference performance of the K-nearest neighbor, random forest, and logistic regression classification algorithms. By using review data of 1000 data which is divided into product classes 618 data, delivery 214 data, and service 168 data. The data was preprocessed using case folding, cleansing, normalization, tokenization, filtering, and stemming stages. Then the word weights in the data are calculated using tf-idf. The data will be divided into train data and test data. After that, classification was carried out with K-nearest neighbor, random forest, and logistic regression. Which produces the highest accuracy of K-nearest neighbor 76%, random forest 81%, and logistic regression 80%.

Item Description

http://repository.upnvj.ac.id/28959/24/AWAL.pdf
http://repository.upnvj.ac.id/28959/1/ABSTRAK.pdf
http://repository.upnvj.ac.id/28959/3/BAB%201.pdf
http://repository.upnvj.ac.id/28959/4/BAB%202.pdf
http://repository.upnvj.ac.id/28959/5/BAB%203.pdf
http://repository.upnvj.ac.id/28959/6/BAB%204.pdf
http://repository.upnvj.ac.id/28959/7/BAB%205.pdf
http://repository.upnvj.ac.id/28959/8/DAFTAR%20PUSTAKA.pdf
http://repository.upnvj.ac.id/28959/9/RIWAYAT%20HIDUP.pdf
http://repository.upnvj.ac.id/28959/10/LAMPIRAN.pdf
http://repository.upnvj.ac.id/28959/11/HASIL%20PLAGIARISME.pdf
http://repository.upnvj.ac.id/28959/12/ARTIKEL%20KI.pdf