<![CDATA[Information and communication technology has now developed very rapidly, bringing significant changes to our daily lives. With the advancement of information and communication technology, access to information has become very easy and fast. However, this convenience also brings its own challenges, especially in terms of personal data security. As technology users, we are required to be wise and vigilant in safeguarding our personal data so that it is not misused by irresponsible parties. One example of cybercrime that often occurs is phishing emails. In this attack, the perpetrator uses a link containing a virus to encrypt the user's data or device, then asks for a ransom to restore access to the data. Phishing emails usually look like official emails from trusted sources, so recipients are often unaware of the dangers lurking. Therefore, to minimize the losses that can occur, we can also take advantage of technology so that we can automatically classify phishing emails. Therefore, this research will carry out the process of building a machine learning model which can automatically classify phishing emails. So that with the model built in this research, it is hoped that it can help in anticipating phishing emails. In this research, the construction of machine learning models will use data with a total of 18650 data which consists of 11322 non-phishing email data and 7328 phishing email data. The model that will be built in this research is a model using the Support Vector Machine and Random. Forest algorithms. In the model building process, to find the optimal parameters, the hyperparameter tuning process is carried out using CV gridsearch, so as to produce optimal parameters. After testing the model to classify phishing emails, the results show that using the Support Vector Machine algorithm produces a test accuracy of 97.27%, while using the Random Forest algorithm produces an accuracy of 96.51%.]]>
