DETEKSI UNSUR DEFAMASI PADA MEDIA SOSIAL TWITTER MENGGUNAKAN KLASIFIKASI SUPPORT VECTOR MACHINE

CANDRA FEBRIAN TRI PAMBUDI, . (2023) DETEKSI UNSUR DEFAMASI PADA MEDIA SOSIAL TWITTER MENGGUNAKAN KLASIFIKASI SUPPORT VECTOR MACHINE. Sarjana thesis, UNIVERSITAS NEGERI JAKARTA.

[img] Text
COVER.pdf

Download (1MB)
[img] Text
BAB 1.pdf

Download (1MB)
[img] Text
BAB 2.pdf
Restricted to Registered users only

Download (766kB) | Request a copy
[img] Text
BAB 3.pdf
Restricted to Registered users only

Download (2MB) | Request a copy
[img] Text
BAB 4.pdf
Restricted to Registered users only

Download (3MB) | Request a copy
[img] Text
BAB 5.pdf
Restricted to Registered users only

Download (338kB) | Request a copy
[img] Text
DAFTAR PUSTAKA.pdf

Download (288kB)
[img] Text
LAMPIRAN.pdf
Restricted to Registered users only

Download (1MB) | Request a copy

Abstract

Defamasi atau Pencemaran nama baik adalah perbuatan yang secara sengaja membuat nama baik atau reputasi orang tersebut menjadi tercemar atau buruk. Hal ini menimbulkan pembunuhan karakter atau character assassination yang merupakan pelanggaran HAM. Skripsi ini bertujuan untuk membangun model deteksi defamasi dengan klasifikasi SVM pada data Twitter dan SafeNet, menghitung performa SVM dan mengetahui pengaruh TF-IDF serta SVD pada ekstraksi fitur terhadap performa model. Berdasarkan SKB 3 Menteri poin g, tindakan yang dikatakan pencemaran nama baik yaitu tindakan yang mengandung unsur penuduhan. Sedangkan yang bukan termasuk defamasi, narasi yang mengandung penilaian, pendapat dan hasil evaluasi. Isi poin-poin tadi akan dijadikan parameter untuk proses labelling dengan 3 annotator. Dengan total 458 data yang di split menjadi 2 yaitu data testing dan data training menggunakan ratio 20:80. Klasifikasi dilakukan sebanyak 2 kali, pertama proses klasifikasi SVM dilakukan tanpa algoritma SVD dan hanya menggunakan TF-IDF pada fitur ekstraksinya. Hasilnya evaluasi yang didapat nilai akurasi 83%, nilai presisi 83%, recall 69%, spesifisitas 91%, nilai F-measure 75%. Proses Kedua, fitur ekstraksi menggunakan TF-IDF dan SVD dihasilkan akurasi 80%, nilai presisi 73%, recall 81%, spesifisitas 80%, nilai F-measure 77%. ***** Defamation is an act that intentionally makes the good name or reputation of the person to be polluted or bad. This leads to character assassination which is a violation of human rights. This skripsi aims to build a defamation detection model with SVM classification on Twitter and SafeNet data, calculate SVM performance and determine the effect of TF-IDF and SVD on feature extraction on model performance. Based on SKB 3 Minister point g, actions that are said to be defamation are actions that contain elements of accusation. Meanwhile, defamation does not include narratives that contain judgements, opinions and evaluation results. The contents of these points will be used as parameters for the labelling process with 3 annotators. With a total of 458 data split into 2, namely testing data and training data using a ratio of 20:80. Classification was carried out twice, first SVM classification process was carried out without the SVD algorithm and only used TF-IDF in feature extraction. The evaluation results obtained accuracy value 83%, precision value 83%, recall 69%, specificity 91%, F-measure value 75%. Second process, feature extraction using TF-IDF and SVD resulted in 80% accuracy, 73% precision value, 81% recall, 80% specificity, 77% F-measure value

Item Type: Thesis (Sarjana)
Additional Information: 1). Dr. Ria Arafiyah, M.Si. ; 2). Drs. Mulyono, M.Kom.
Subjects: Sains > Matematika > Ilmu Komputer
Divisions: FMIPA > S1 Ilmu Komputer
Depositing User: Users 20822 not found.
Date Deposited: 25 Sep 2023 23:54
Last Modified: 25 Sep 2023 23:54
URI: http://repository.unj.ac.id/id/eprint/42982

Actions (login required)

View Item View Item