Description:
The dataset "Nigerian_Fraud.csv" comprises records of email communications attributed to fraudulent activities, often referred to colloquially as "Nigerian scams." These phishing emails aim to lure recipients into financial fraud under various pretenses. The dataset includes the sender and receiver information, the date of the email, the subject line, the body content of the email, the presence of URLs within the email body, and a label indicating whether the email is fraudulent.
Attribute Description:
- `sender`: The email address and possibly the name of the email's sender.
- `receiver`: The recipient's initials or anonymized identifier, with some records missing.
- `date`: The date and time the email was sent, in various time zones.
- `subject`: The subject line of the email.
- `body`: The full content of the email, rich with details intended to persuade or deceive the recipient into a fraudulent scheme.
- `urls`: A binary indicator (0 or 1) signifying the absence or presence of URLs in the email body, which might lead to phishing websites.
- `label`: A binary indicator (1) confirming each email as fraudulent.
Use Case:
This dataset serves as a valuable resource for developing machine learning models to detect phishing and scam emails automatically. By analyzing patterns in the dataset, such as common phrases, the structure of the sender's email address, and the presence of URLs, researchers and cybersecurity experts can train models to recognize and flag similar fraudulent attempts in real-time, enhancing email security protocols and protecting users from potential financial scams. Additionally, it provides insights into the tactics used by fraudsters, contributing to digital literacy and cybersecurity awareness programs.