ANALISIS SENTIMEN PADA ULASAN PENGGUNA APLIKASI MAXIM MENGGUNAKAN METODE KLASIFIKASI RANDOM FOREST DAN EKSTRAKSI FITUR WORD2VEC

Rafid Ammar Adinegoro, . (2023) ANALISIS SENTIMEN PADA ULASAN PENGGUNA APLIKASI MAXIM MENGGUNAKAN METODE KLASIFIKASI RANDOM FOREST DAN EKSTRAKSI FITUR WORD2VEC. Skripsi thesis, Universitas Pembangunan Nasional Veteran Jakarta.

[img] Text
ABSTRAK.pdf
Download (31kB)
[img] Text
AWAL.pdf
Download (609kB)
[img] Text
BAB 1.pdf
Restricted to Repository UPNVJ Only
Download (77kB)
[img] Text
BAB 2.pdf
Restricted to Repository UPNVJ Only
Download (417kB)
[img] Text
BAB 3.pdf
Restricted to Repository UPNVJ Only
Download (374kB)
[img] Text
BAB 4.pdf
Restricted to Repository UPNVJ Only
Download (1MB)
[img] Text
BAB 5.pdf
Download (67kB)
[img] Text
DAFTAR PUSTAKA.pdf
Download (138kB)
[img] Text
RIWAYAT HIDUP.pdf
Restricted to Repository UPNVJ Only
Download (71kB)
[img] Text
LAMPIRAN.pdf
Restricted to Repository UPNVJ Only
Download (588kB)
[img] Text
HASIL PLAGIARISME.pdf
Restricted to Repository staff only
Download (298kB)
[img] Text
Artikel KI.pdf
Restricted to Repository staff only
Download (694kB)

Abstract

The industrial revolution 4.0 has caused businesses to start moving into the digital realm, one of which is online transportation. Several online transportation companies have started to appear by launching their business through applications, one of which is Maxim. Along with the increase in the number of application users, the demand for application service quality has also increased. One way to assess an application service is to look at user reviews. This research conducted sentiment analysis on 2 classes, positive and negative, in user reviews of the Maxim application. The classification algorithm used is Random Forest and applies Word2vec feature extraction. Word2vec feature extraction is used to make sentiment analysis more effective because it is able to recognize semantics between words. By testing different Word2vec parameter values, the best performance was obtained at dimension parameters 300, window 7, and epochs 10. The results of this study demonstrate that the Random Forest classification with Word2Vec feature extraction achieved a higher level of performance compared to not using Word2Vec feature extraction, resulting in an accuracy of 93,39%, precision of 95,85%, recall of 91,54%, and an F1-score of 93,65%. In comparison, the performance without using feature extraction only yielded an accuracy of 92,63%, precision of 95,79%, recall of 90,10%, and an F1-score of 92,86%.

Item Type: Thesis (Skripsi)
Additional Information: [No. Panggil : 1910511076] [Pembimbing 1 : Henki Bayu Seta] [Pembimbing 2 : Iin Ernawati] [Penguji 1 : Yuni Widiastiwi] [Penguji 2 : Zatin Niqotaini]
Uncontrolled Keywords: Sentiment Analysis, Maxim, Classification, Random Forest, Word2Vec
Subjects: Q Science > QA Mathematics > QA75 Electronic computers. Computer science
Divisions: Fakultas Ilmu Komputer > Program Studi Informatika (S1)
Depositing User: Rafid Ammar Adinegoro
Date Deposited: 18 Jan 2024 02:27
Last Modified: 16 Feb 2024 08:54
URI: http://repository.upnvj.ac.id/id/eprint/27621

Actions (login required)

View Item View Item