• Deeksha Tiwari
• Astha Shukla
• Avika Gupta
• Akshita Dixit

Phishing emails frequently result in monetary losses, data breaches, and security flaws, they are becoming a bigger worry for people as well as companies. Phishing emails are usually designed to trick the recipient into disclosing private information, including financial information, login credentials, and personal identification numbers. Effectively delivering emails has come to be an enormous cybersecurity concern. to detect phishing emails, this article uses machine learning algorithms to find suspicious patterns and characters that are frequently present in false emails. Our method entails preprocessing email content, such as subject lines, body text, and metadata, with the goal to extract useful features that may be indicative of phishing. The paper also discusses the application of natural language processing (NLP) techniques for feature extraction, which aids in analysing the semantic content of emails for patterns like urgency, dubious links, or misleading language frequently employed in phishing attempts. By analysing the content's semantic meaning using Natural Language Processing (NLP) resources. The study builds prediction systems that can differentiate between malicious and legitimate emails using a variety of machine learning models, such as Random Forest classifiers, Decision Trees, and SVM Results from experiments shows that machine learning systems can detect phishing emails with excellent accuracy even when working with relatively little datasets. This illustrates how well these models work at spotting typical phishing traits including misleading wording, dubious URLs, and odd metadata patterns. The system can adjust to new phishing tactics and preserve its detection capabilities by regularly feeding the models new data. This method guarantees long-term dependability in addition to increasing precision. Furthermore, the suggested approach provides an effective and scalable solution that can be included into current email security systems.

Keywords— Cybersecurity, email security, phishing emails, phishing attacks, email detection, feature extraction, natural language processing (NLP), machine learning, Support Vector Machine (SVM), Decision Trees, Random Forest.

"InboxGuardian: A Synergistic Approach Combining BERT and GANs for Phishing Detection in email", International Journal of Emerging Technologies and Innovative Research (www.jetir.org), ISSN:2349-5162, Vol.12, Issue 7, page no.a55-a61, July-2025, Available :http://www.jetir.org/papers/JETIR2507006.pdf

"InboxGuardian: A Synergistic Approach Combining BERT and GANs for Phishing Detection in email", International Journal of Emerging Technologies and Innovative Research (www.jetir.org | UGC and issn Approved), ISSN:2349-5162, Vol.12, Issue 7, page no. ppa55-a61, July-2025, Available at : http://www.jetir.org/papers/JETIR2507006.pdf