The term “cloud computing” describes the phenomenon wherein resources such as data storage and processing power are made available on demand by personal computer systems, often without the client’s intervention. People and businesses alike often use email for the transfer of data. Credit reports and financial data are examples of the sensitive information that is often sent over the Internet. Con artists employ a technique called phishing to trick consumers into giving over vital information by making bogus emails seem official. A phished email may trick into divulging vital information via deceit. The email phishing attacks issue may happen at any point in the process, from sending to receiving. The attacker exploits your personal information when open as well as read an email to transmit spam. It has been a huge issue recently. This research uses different amounts of legitimate along with phishing data to detect fresh emails, categorise them using diverse features as well as algorithms, and draw conclusions. A fresh dataset is generated once the present methods are evaluated. This research used a hybrid strategy that combined deep learning (DL) and machine learning (ML) after creating a feature-extracted CSV file and a label file. Recognising a phished email is the focus of this experiment’s classification challenge. Several preexisting ML and DL methods are compared and put into practice. The results of the proposed hybrid algorithm demonstrate an improved as well as more accurate performance in identifying email phishing attacks.