IMPLEMENTASI JARINGAN SARAF KONVOLUSI TERHADAP ANALISIS SENTIMEN TENTANG KULIAH ONLINE PADA MASA COVID-19
dc.contributor.author | Pratama, Putra Agung | |
dc.contributor.supervisor | Adnan, Arisman | |
dc.date.accessioned | 2022-08-02T04:43:23Z | |
dc.date.available | 2022-08-02T04:43:23Z | |
dc.date.issued | 2022-03 | |
dc.description.abstract | This paper discusses online course that use the internet network to stay connected during the activity. This study aims to see the impact of online course based on someone's opinion. One of the appropriate methods for this research is sentiment analysis. For this reason, there are 7000 tweets is analyzed from media social twitter April 2020–April 2021 which convey opinions about online course. Sentiment analysis uses a convolution neural network (one directional convolution) which classifies data in the form of text documents. Convolutional neural network is trained using keras programming with 100 epoch. The convolutional neural network trains using 5600 tweets and predicts 1400 different tweets. The training results from the convolution neural network give a neutral sentiment as the most dominant sentiment with amount 76.5% accuracy level. | en_US |
dc.description.sponsorship | Fakultas Matematika dan Ilmu Pengetahuan Alam Universitas Riau | en_US |
dc.identifier.citation | Perpustakaan | en_US |
dc.identifier.issn | Elfitra | |
dc.identifier.uri | https://repository.unri.ac.id/handle/123456789/10630 | |
dc.language.iso | en | en_US |
dc.publisher | Elfitra | en_US |
dc.subject | Online course | en_US |
dc.subject | Online course | en_US |
dc.subject | sentiment analysis | en_US |
dc.subject | convolutional neural network | en_US |
dc.subject | en_US | |
dc.title | IMPLEMENTASI JARINGAN SARAF KONVOLUSI TERHADAP ANALISIS SENTIMEN TENTANG KULIAH ONLINE PADA MASA COVID-19 | en_US |
dc.type | Article | en_US |
Files
Original bundle
1 - 1 of 1
No Thumbnail Available
- Name:
- Putra Agung Pratama_compressed.pdf
- Size:
- 251.87 KB
- Format:
- Unknown data format
- Description:
- artikel
License bundle
1 - 1 of 1
No Thumbnail Available
- Name:
- license.txt
- Size:
- 1.71 KB
- Format:
- Item-specific license agreed upon to submission
- Description: