Skripsi

IDENTIFIKASI PELECEHAN SEKSUAL PADA INSTAGRAM DENGAN METODE SUPPORT VECTOR MACHINE

---

Penilaian

---

The existence of social media makes many people upload their experiences, either in the form of writing or multimedia. The reactions from these uploads also vary, some respond well in the form of praise, suggestions and positive comments, some respond badly such as giving reproaches, insults and even harassing sentences. This study aims to identify whether a sentence on Instagram includes sexual harassment or not using the Support Vector Machine (SVM) classification method. The SVM method was chosen because this method is often used for classification and has good performance results. In its implementation, the data is pre-processed to make the data cleaner and more structured, then feature extraction is carried out using Term Frequency-Inverse Document Frequency (TF-IDF) and classification is carried out using the SVM method with a linear kernel. Testing this research using the K-Fold Cross Validation method, the data, in total 200 Instagram comments containing 100 harassment comments and 100 non-harassment comments, is divided into 90% training data and 10% test data with a K value of 10. After 10 tests, the SVM classification method with linear kernel produces an accuracy value of 71%, a precision value of 93%, a recall value of 65% and an f-1 value of 76% with an average computation time of 0.035 seconds.

---

Availability

Inventory Code	Barcode	Call Number	Location	Status
2307001485	T88069	T880692022	Central Library (REFERENS)	Available but not for loan - Not for Loan

Detail Information

Series Title

-

Call Number

T880692022

Publisher

Indralaya : Jurusan Teknik Informatika, Fakultas Ilmu Komputer, Universitas Sriwijaya., 2022

Collation

xiii, 64 hlm.; ilus.; 29 cm

Language

Indonesia

ISBN/ISSN

-

Classification

362.83 07

Content Type

-

Media Type

-

Carrier Type

-

Edition

-

Subject(s)

Jurusan Teknik Informatika

Specific Detail Info

-

Statement of Responsibility

ANUG

Other version/related

File Attachment

Comments

---

You must be logged in to post a comment