SENTIMENT ANALYSIS OPTIMIZATION OF GOJEK APPLICATION: COMPARING PERFORMANCE OF RANDOM FOREST METHOD AND SUPPORT VECTOR MACHINE, BASED ON INFORMATION GAIN AND KERNELS

ARIESTAMA, DONNY RAFAEL (2023) SENTIMENT ANALYSIS OPTIMIZATION OF GOJEK APPLICATION: COMPARING PERFORMANCE OF RANDOM FOREST METHOD AND SUPPORT VECTOR MACHINE, BASED ON INFORMATION GAIN AND KERNELS. Skripsi thesis, UNIVERSITAS KATOLIK SOEGIJAPRANATA.

[img] Text
19.K1.0036-RAFAEL DONNY ARIESTAMA_COVER_1.pdf
Download (445kB)
[img] Text
19.K1.0036-RAFAEL DONNY ARIESTAMA_BAB I_1.pdf
Restricted to Registered users only
Download (344kB)
[img] Text
19.K1.0036-RAFAEL DONNY ARIESTAMA_BAB II_1.pdf
Restricted to Registered users only
Download (350kB)
[img] Text
19.K1.0036-RAFAEL DONNY ARIESTAMA_BAB III_1.pdf
Restricted to Registered users only
Download (850kB)
[img] Text
19.K1.0036-RAFAEL DONNY ARIESTAMA_BAB IV_1.pdf
Restricted to Registered users only
Download (1MB)
[img] Text
19.K1.0036-RAFAEL DONNY ARIESTAMA_BAB V_1.pdf
Restricted to Registered users only
Download (275kB)
[img] Text
19.K1.0036-RAFAEL DONNY ARIESTAMA_DAPUS_1.pdf
Download (278kB)
[img] Text
19.K1.0036-RAFAEL DONNY ARIESTAMA_LAMPIRAN_1.pdf
Restricted to Registered users only
Download (611kB)

Abstract

In today's digital environment, where consumer-generated content plays an important role in shaping consumer choices, the Google Play Store encourages customers to actively participate in app reviews and ratings to drive downloads decisions. On the Google Play Store, users regularly browse other users' reviews and app scores before downloading an app. This user rating study utilizes the random forest method and SVM. The evaluation is carried out based on around 1000 user reviews collected from the Gojek Indonesia application review on the Google play store. This study consists of four main steps. The first step is the data collection by scraping from the Gojek reviews and comments on Google Play Store. The next step is preprocessing text where the text is cleared of words that have no meaning and punctuation. After the text dataset that has been successfully cleaned will be done sentiment analysis. Finally, a model evaluation will be performed. Sentiment analysis divides into two categories: positive and negative. The results of this study highlight the essential value of analyzing feedback from users, as it provides app developers with helpful insights, helping them to navigate future strategic choices.

Item Type: Thesis (Skripsi)
Subjects: 000 Computer Science, Information and General Works > 004 Data processing & computer science
Divisions: Faculty of Computer Science > Department of Informatics Engineering
Depositing User: Mr Yosua Norman Rumondor
Date Deposited: 16 Apr 2024 01:15
Last Modified: 16 Apr 2024 01:15
URI: http://repository.unika.ac.id/id/eprint/35160

Actions (login required)

View Item View Item
Unika Repository is powered by EPrints 3 which is developed by the School of Electronics and Computer Science at the University of Southampton. More information and software credits.