Phishing emails are a type of online scam where criminals send an email that appears to be from a legitimate company and ask you to provide sensitive information. This is usually done by including a link that will appear to take you to the company’s website to fill in your information – but the website is a clever fake and the information you provide goes straight to the person behind the scam.
DTonomy recently conducted research to learn how phishing techniques and themes have changed over the years, and how phishing emails are distributed over year, month and day of the week.
For this research, Unsupervised Machine Learning and the K-means Clustering technique were applied to identify the phishing themes.
Unsupervised Machine Learning uncovers previously unknown patterns in data and is best applied when there is no data on desired outcomes or for problems that the business has not seen before.
K-means Clustering is a popular unsupervised machine learning algorithm that groups similar data points together based on certain similarities to discover underlying patterns.
- The purpose of the study
- The analysis of the data set
- The process of cleaning the data
- Identification of a phishing theme using clustering
- Our conclusions
From the analysis we could find the overall theme of phishing attacks by applying Unsupervised Machine Learning techniques and also comparing the phishing attack themes between years. We could easily find the theme of the phishing attack without even reading the entire email using K-means clustering. This method also helped in analyzing how the theme of attack has changed over the years and how they are distributed over year, month, and day of the week.
For the detailed analysis paper, download the paper below
