Analisis sentimen Twitter terhadap US election 2020 dengan metode naive bayes dan support vector machine (SVM) = Twitter sentiment analysis based on us election 2020 using naive bayes and support vector machine (svm)

Wiyono, Brigitta (2021) Analisis sentimen Twitter terhadap US election 2020 dengan metode naive bayes dan support vector machine (SVM) = Twitter sentiment analysis based on us election 2020 using naive bayes and support vector machine (svm). Bachelor thesis, Universitas Pelita Harapan.

[img] Text (Title)
Title (13).pdf
Restricted to Repository staff only
Available under License Creative Commons Attribution Non-commercial Share Alike.

Download (1MB)
[img]
Preview
Text (Abstract)
Abstract.pdf
Available under License Creative Commons Attribution Non-commercial Share Alike.

Download (471kB) | Preview
[img]
Preview
Text (ToC)
ToC.pdf
Available under License Creative Commons Attribution Non-commercial Share Alike.

Download (494kB) | Preview
[img]
Preview
Text (Chapter1)
Chapter1.pdf
Available under License Creative Commons Attribution Non-commercial Share Alike.

Download (479kB) | Preview
[img] Text (Chapter2)
Chapter2.pdf
Restricted to Registered users only
Available under License Creative Commons Attribution Non-commercial Share Alike.

Download (679kB)
[img] Text (Chapter3)
Chapter3.pdf
Restricted to Registered users only
Available under License Creative Commons Attribution Non-commercial Share Alike.

Download (1MB)
[img] Text (Chapter4)
Chapter4.pdf
Restricted to Registered users only
Available under License Creative Commons Attribution Non-commercial Share Alike.

Download (2MB)
[img] Text (Chapter5)
Chapter5.pdf
Restricted to Registered users only
Available under License Creative Commons Attribution Non-commercial Share Alike.

Download (459kB)
[img]
Preview
Text (Bibliography)
Bibliography.pdf
Available under License Creative Commons Attribution Non-commercial Share Alike.

Download (471kB) | Preview
[img] Text (Appendices)
Appendices.pdf
Restricted to Repository staff only
Available under License Creative Commons Attribution Non-commercial Share Alike.

Download (5MB)

Abstract

Penggunaan media sosial pada manusia saat ini beragam, salah satunya adalah penggunaan Twitter. Dengan media ini, manusia dapat dengan bebas berekspresi dan mengutarakan pendapatnya mengenai suatu kejadian yang sedang hangat diperbincangkan. Pendapat yang disampaikan ini tentu beragam, dan dari keberagaman ini dapat dilihat sentimen atau kecenderungan manusia terhadap suatu kejadian. Penelitian ini membahas mengenai analisis sentimen Twitter terhadap kedua kandidat US Election 2020 dengan metode Naive Bayes dan Support Vector Machine (SVM). Kedua metode ini merupakan metode supervised learning, yang digunakan karena data yang diperoleh merupakan data yang sudah berlabel dan pengolahannya dilakukan untuk menguji akurasi pada penggunakan model di data baru. Analisis ini dilakukan terhadap 6437 tweets yang diperoleh dari Twitter dan dikumpulkan selama 30 hari. Data tersebut diolah, hasilnya terbagi menjadi 15 variabel pada model linear dan 29 variabel pada model nonlinear. Hasil penelitian menunjukkan pada pengujian test data, dapat dilihat bahwa SVM lebih berkompetensi dibandingkan Naive Bayes, dengan selisih paling kecil 0.6% dan selisih terbesar 3.3%. / The use of social media varies, take Twitter for instance. With this platform, people have the freedom to share thoughts and opinions about some topics. There might be lots of comments, and it can be infered that this variety could gain some sentiments on some events. The purpose of this research is to find out the sentiments from Twitter about the two candidates of US Election 2020. The methods used in finding these components are Naive Bayes and Support Vector Machine (SVM). Both of this methods are supervised learning methods. The reason of using these methods are because the data has been labeled and used to test the accuracy of the model in the new data. The analysis was conducted on 6437 tweets from Twitter and was collected for 30 days. It was processed, then divided into 15 variables on linear model and 29 variables on nonlinear model. The results showed that the test data on SVM are more competent on predicting the label than Naive Bayes, with the smallest difference is around 0.6% and the biggest is about 3.3%.

Item Type: Thesis (Bachelor)
Creators:
CreatorsNIMEmail
Wiyono, BrigittaNIM01112170017monicawiyono20@gmail.com
Contributors:
ContributionContributorsNIDN/NIDKEmail
Thesis advisorSaputra, Kie Van IvankyNIDN0401038203kie.saputra@uph.edu
Thesis advisorKrisnadi, DionNIDN0316029002dion.krisnadi@uph.edu
Uncontrolled Keywords: analisis sentimen; SVM; naive bayes
Subjects: Q Science > QA Mathematics
Divisions: University Subject > Current > Faculty/School - UPH Karawaci > Faculty of Science and Technology > Mathematics
Current > Faculty/School - UPH Karawaci > Faculty of Science and Technology > Mathematics
Depositing User: Brigitta Monica Wiyono
Date Deposited: 05 Aug 2021 01:27
Last Modified: 05 Aug 2021 01:27
URI: http://repository.uph.edu/id/eprint/41178

Actions (login required)

View Item View Item