Kharismojati, Dary (2023) Topic Modeling Review Penggunaan Aplikasi Indodax Mobile Menggunakan Metode Latent Dirichlet Allocation. Undergraduate thesis, UPN Veteran Jawa Timur.
|
Text (Cover)
18082010057-cover.pdf Download (4MB) | Preview |
|
|
Text (BAB 1)
18082010057-bab1.pdf Download (2MB) | Preview |
|
![[img]](http://repository.upnjatim.ac.id/style/images/fileicons/text.png) |
Text (BAB 2)
18082010057-bab2.pdf Restricted to Registered users only until 5 June 2025. Download (2MB) |
|
|
Text (BAB 3)
18082010057-bab3.pdf Restricted to Registered users only until 5 June 2025. Download (2MB) |
|
|
Text (BAB 4)
18082010057-bab4.pdf Restricted to Registered users only until 5 June 2025. Download (10MB) |
|
|
Text (BAB 5)
18082010057-bab5.pdf Download (647kB) | Preview |
|
|
Text (Daftar Pustaka)
18082010057-daftarpustaka.pdf Download (815kB) | Preview |
Abstract
Community feedback is very important in the development of an application product or service. By knowing people's responses, developers can find out the advantages and disadvantages of the application, so that they can improve its quality and security. Therefore, this research aims to implement the Latent Dirichlet Allocation method on the Indodax application review from Google Playstore, so as to reveal public responses to cryptocurrency trading on the Indodax application and provide insight for future application development. With the Latent Dirichlet Allocation method, a topic modelling model can be built on Indodax application reviews that can reveal public responses regarding cryptocurrency trading on the Indodax application. The steps taken include scraping review data taken from March 2021 to June 2022, followed by the data understanding stage where 14,854 user reviews were found, data preprocessing, topic modeling model tuning to produce a better coherence score, visualization using pyLDAvis and wordcloud on a static website and analysis of topic modeling results. The results showed that there were five main topics revealed by the topic modelling model used, with the 5th topic having the highest coherence of 0.627. The final model uses the LdaMulticore method with 5 topics were visualized using pyLDAvis and wordcloud. The 1st topic addresses app features and appearance, the 2nd topic addresses transactions and fees, the 3rd topic addresses security and verification, the 4th topic addresses app features and appearance, and the 5th topic addresses app performance and customer service issues. With this model, it is possible to reveal people's responses regarding cryptocurrency trading on the Indodax application and is expected to be an initial reference to provide insights for future application development and overall business.
| Item Type: | Thesis (Undergraduate) | ||||||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| Contributors: |
|
||||||||||||
| Subjects: | Q Science > QA Mathematics > QA75 Electronic computers. Computer science | ||||||||||||
| Divisions: | Faculty of Computer Science > Departemen of Information Systems | ||||||||||||
| Depositing User: | Mr Dary Kharismojati | ||||||||||||
| Date Deposited: | 07 Jun 2023 01:27 | ||||||||||||
| Last Modified: | 07 Jun 2023 01:27 | ||||||||||||
| URI: | http://repository.upnjatim.ac.id/id/eprint/14606 |
Actions (login required)
 |
View Item |