Klasifikasi Postingan Pengguna Facebook Untuk Deteksi Phising Menggunakan Naive Bayes

Abstrak

Phishing is a digital fraud that is commonly carried out by cybercriminals with the aim of taking user's personal information data by manipulating it. Facebook is a very popular social media platform in the world so it can be a wet place for phishing criminals. In this research, we built a Classification model to identify and prevent phishing attempts on Facebook posts. The dataset used in this study was obtained from Facebook user posts collected. Data processing is done by preprocessing the post text, including removing punctuation marks and words that are not important. The method used is Naïve Bayes to classify posts into phishing or not phishing categories. The Naïve Bayes method is used because of its ability to classify data with a good level of accuracy. This shows that the features selected in this study can be a strong indicator for detecting phishing on Facebook user posts. The results of the study show that Naïve Bayes can be an effective solution for phishing detection on Facebook user posts. In addition, the results of this research can provide valuable insight into the common characteristics of phishing posts on Facebook. With an accuracy value of 99.01%, it is hoped that this research can help increase awareness and security of Facebook users against phishing posts.