Prediksi video perfomance akun Youtube Buzzfeed menggunakan metode naive bayes

Leonie, Valerie (2020) Prediksi video perfomance akun Youtube Buzzfeed menggunakan metode naive bayes. Bachelor thesis, Universitas Pelita Harapan.

[img] Text (Title)
Title.pdf
Restricted to Registered users 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 (465kB) | Preview
[img] Text (ToC)
ToC.pdf
Restricted to Registered users only
Available under License Creative Commons Attribution Non-commercial Share Alike.

Download (940kB)
[img]
Preview
Text (Chapter1)
Chapter1.pdf
Available under License Creative Commons Attribution Non-commercial Share Alike.

Download (499kB) | Preview
[img] Text (Chapter2)
Chapter2.pdf
Restricted to Registered users only
Available under License Creative Commons Attribution Non-commercial Share Alike.

Download (869kB)
[img] Text (Chapter3)
Chapter3.pdf
Restricted to Registered users only
Available under License Creative Commons Attribution Non-commercial Share Alike.

Download (858kB)
[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 (550kB)
[img]
Preview
Text (Bibliography)
Bibliography.pdf
Available under License Creative Commons Attribution Non-commercial Share Alike.

Download (418kB) | Preview
[img] Text (Appendices)
Appendices.pdf
Restricted to Repository staff only
Available under License Creative Commons Attribution Non-commercial Share Alike.

Download (29MB)

Abstract

Selama beberapa tahun terakhir, perkembangan media sosial sangat pesat sehingga mencakup hampir seluruh aspek kehidupan manusia. Dengan berkembangnya media sosial ini, tentu saja hiburan dan teknologi ikut berkembang beriringan bersamanya. Orang-orang sekarang mencari kesenangan besosialisasi dan berbagi segala sesuatunya melalui internet, salah satunya melalui video dan disebarkan pada berbagai media sosial seperti Facebook, Twitter dan Youtube. Hal ini mendorong banyak orang sekarang menjadikan platform-platform tersebut sebagai media untuk mata pencaharian mereka, salah satunya adalah Buzzfeed. Oleh karena itu diperlukan penelitian mengenai performa video-video tersebut untuk mengetahui bagaimana perkembangan akun Youtube kita. Dari sekian banyak metode untuk memprediksi performa video-video tersebut, digunakanlah metode naive bayes. Metode ini digunakan karena merupakan metode yang mudah serta cepat, serta dapat bekerja dengan jumlah data yang sedikit. Pengumpulan data dilakukan dengan teknik dokumentasi, dengan mengumpulkan semua data-data yang ada pada akun Youtube Buzzfeed dan kemudian diolah sesuai keperluan penelitian. Total data yang dikumpulkan selama setahun yaitu pada tahun 2019 mencapai 772 data yang kemudian dikelola menggunakan software Rapid Miner menggunakan metode naive bayes. Kemudian dari penelitian ini dapat disimpulkan bahwa naive bayes memiliki akurasi total yang cukup baik yaitu 86.01%, dengan class recall untuk kelas rendah sebesar 91.21%, dan kelas sedang dan tinggi sebesar 0%, sedangkan class precision untuk prediksi kelas rendah sebesar 94.32% dan kelas sedang dan tinggi sebesar 0%. Hal ini dikarenakan minimnya variasi data dengan kelas sedang dan tinggi sehingga algoritma masih sulit untuk memprediksi dengan tepat kelas dengan nilai sedang dan tinggi tersebut. / During the past few years, the development of social media has grown so fast that it affects almost all aspects of human life. With the development of social media, entertainment and technology industries also develop with it. Nowadays, people are looking for fun by socializing and sharing every little things on the internet, by using video and share it to various social media such as Facebook, Twitter and Youtube. This encourages people to use these platforms as their source of money, one of which is Buzzfeed. Therefore, we need to research on the perfomance of the videos to know how the Youtube account is developing. Out of many others methods that is used to predict the perfomance of these videos, naive bayes is being used. This method is used because it’s an easy and fast method, and it works well with small amount of data. Data collection is carried out by documentation technique, by collecting all the existing data on Buzzfeed Youtube account and then use it according to the research needs. Total data collected during the year of 2019 is 772 data, that those data were then managed by using Rapid Miner software using the naive bayes method. Then after this research it can be concluded that naive bayes has a pretty good total accuracy of 86.01%, with class recall for low classes at 91.21%, and for medium and high classes at 0%, while class precision for low class prediction at 94.32% while for medium and high class is 0%. Due to the lack of variation of data on the medium and high classes, the algorithm is still unable to predict precisely on the medium and high value class.

Item Type: Thesis (Bachelor)
Creators:
CreatorsNIMEmail
Leonie, ValerieNIM00000022818vleonie.vl@gmail.com
Contributors:
ContributionContributorsNIDN/NIDKEmail
Thesis advisorPangaribuan, Jefri JuniferNIDN0130108901jefri.pangaribuan@uph.edu
Uncontrolled Keywords: analisa; data mining; Youtube; naive bayes
Subjects: Q Science > QA Mathematics > QA75 Electronic computers. Computer science
Divisions: University Subject > Current > Faculty/School - UPH Medan > School of Information Science and Technology > Information Systems
Current > Faculty/School - UPH Medan > School of Information Science and Technology > Information Systems
Depositing User: Users 9232 not found.
Date Deposited: 15 Aug 2020 04:57
Last Modified: 13 Jan 2022 08:35
URI: http://repository.uph.edu/id/eprint/10217

Actions (login required)

View Item View Item