ANALISIS SENTIMEN PADA KOMENTAR INSTAGRAM DPR RI MENGGUNAKAN METODE SUPPORT VECTOR MACHINE

dc.contributor.authorArisma, Fauzan
dc.contributor.supervisorFatayat, Fatayat
dc.date.accessioned2021-08-26T03:35:17Z
dc.date.available2021-08-26T03:35:17Z
dc.date.issued2020-11
dc.description.abstractInstagram is one of the most used social media in Indonesia. Instagram is often used by young people and government agencies to upload their activities. One feature on Instagram for uploading is posting. In the posting feature, Instagram users can write their opinions on the post, including the DPR RI Instagram. Instagram users can give both positive and negative opinions on posts made by the DPR RI Instagram account so that it is easy to give aspirations. One way to get information through Instagram is to do sentiment analysis. The algorithm that will be used in this research is Support Vector Machine (SVM). The results of this study found that the average negative opinion each month was greater that is equal to 58.69% compared to positive opinions which only amounted to 41.31%. From the SVM classification results obtained the highest average accuracy obtained by using the Rational Basic Function (RBF) kernel which is equal to 65.87%, and the lowest average accuracy value obtained using the Polynomial kernel which is equal to 57.44%.en_US
dc.description.sponsorshipJurusan Ilmu Komputer Fakultas Matematika dan Ilmu Pengetahuan Alam Unversits Riauen_US
dc.identifier.otherwahyu sari yeni
dc.identifier.urihttps://repository.unri.ac.id/handle/123456789/10163
dc.language.isoenen_US
dc.subjectDPR RIen_US
dc.subjectRBFen_US
dc.subjectSentiment Analysisen_US
dc.subjectSupport Vector Machineen_US
dc.subjectPolynomialen_US
dc.titleANALISIS SENTIMEN PADA KOMENTAR INSTAGRAM DPR RI MENGGUNAKAN METODE SUPPORT VECTOR MACHINEen_US
dc.typeArticleen_US

Files

Original bundle
Now showing 1 - 1 of 1
No Thumbnail Available
Name:
JOM FAUZAN ARISMA(REVISI3)_compressed.pdf
Size:
211.07 KB
Format:
Unknown data format
Description:
artikel
License bundle
Now showing 1 - 1 of 1
No Thumbnail Available
Name:
license.txt
Size:
1.71 KB
Format:
Item-specific license agreed upon to submission
Description:

Collections