Text

SISTEM KLASIFIKASI JUDUL BERITA CLICKBAIT BERBAHASA INDONESIA MENGGUNAKAN NAIVE BAYES CLASSIFIER DAN SELEKSI FITUR CHI SQUARE

---

Penilaian

---

Clickbait refers to a type of web content ad designed to entice readers to click on a link. Not a few news that misuse clickbait by providing news content that does not match the title. Such news has the potential to harm readers so that clickbait news title classification system is needed to make it easier for readers to filter news. In this study, clickbait headlines were classified using the Naïve Bayes Classifier algorithm and chi square feature selection. Feature selection itself is done based on the division of feature ratios as much as 25%, 50%, 75%, and 100% of all features. There are two processes: using stopwords removal in the pre-process and no stopwords removal. Systems without stopwords removal result in average accuracy values of 80.73%, 80.80%, 79.15%, and 78.33% respectively with best accuracy achieved at a feature ratio of 50%. While with stopwords removal resulted in an average accuracy of 66.29%, 67.25%, 65.73%, and 67.71% with the best accuracy at a feature ratio of 100%. The test results showed that the selection of chi square features of each ratio affected the amount of data that could be classified and the computing time and stopwords removal had an effect on system performance results.

---

Availability

Inventory Code	Barcode	Call Number	Location	Status
2207000181	T62723	T627232022	Central Library (Referens)	Available but not for loan - Repaired

Detail Information

Series Title

-

Call Number

T627232022

Publisher

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

Collation

xxi, 136 hlm.; ilus.; 29 cm

Language

Indonesia

ISBN/ISSN

-

Classification

005.360 7

Content Type

-

Media Type

-

Carrier Type

-

Edition

-

Subject(s)

Jurusan Teknik Informatika
Program untuk Komputer Personal

Specific Detail Info

-

Statement of Responsibility

SEPTA

Other version/related

File Attachment

Comments

---

You must be logged in to post a comment