Question

Deceitful emails are illusory measures that are taken up by attackers for personal gain in order to lure in innocent people. They are used to scam and defraud people. The emails usually involve offers that are too good to be true, and they are targeted towards naïve individuals. If you as a Cybersecurity expert are facing a phishing email scenario in your organisation Trident, how will you educate employees within your organisation. Illustrate and justify the use of machine learning to catch email fraud and spam to top management besides ensuring digital literacy in your organisation. Cite your sources.

Answer 1

Phishing emails are not only creating cyber attacks ,but also is the major contributor for data loss.

To overcome these scenario, one should need to enchance security to catch fruad and spam mails,.

Based on my learning , we can make use of machine learning models which helps in dealing with phishing mails.

In unsupervised learning one tries to uncover hidden regularities in data like spam messages. In e-mail filtering task some features could be the bag of words or the subject line, the input to e-mail classification task can be viewed as a two dimensional matrix, whose axes are the messages and the features.

E-mail classification tasks are divided into - First, Data collection and representation and second, e-mail feature selection and feature reduction attempt to reduce the size.

Naïve Bayes classifier -

Defination - Bayesian classifier is working on the dependent events and the probability of an event occurring in the future that can be detected from the previous occurring of the same event.

Working - : Naive tecnique can be used to classify spam e-mails; words probabilities play the main rule. If some words occur often in spam but not in ham, then this incoming e-mail is probably spam.Its very popular among email filtering software .Bayesian filter should be trained to work effectively. Every word has certain probability of occurring in spam or ham email in its database. If the total of words probabilities exceeds a certain limit, the filter will mark the e-mail to categories there are only two categories - spam or unspam ,

Almost all the probabilty related work comes under naive filter ,

Phrase -  if I[message] > threshold[message] is marked as spam else msg is marked as non-spam.

K-nearest neighbour classifier -

Defination :- The k-nearest neighbour (K-NN) classifier is considered an example-based classifier, meaning  that the training documents are used for comparison rather than any specific data.

Working : When a new document needs to be categorized, the k most similar documents (neighbours) are found and if a large enough proportion of them have been assigned to a certain category, then the new document is also assigned to this category, otherwise not .To decide whether a message is spam or ham, we look at the class of the messages that are closest to it.

Phrase - If l or more messages among the k nearest neighbours of x are spam, classify x as spam, otherwise classify it as genuine mail .

Similarly ,we can use Artifical Neural Network classifier , Support vector machines to filter/detect spam mails , depending on size and severity of organisationa and type of work,

For example - defense and cloud data servers are encrypted to next level technology were its hard to crack these systems.

Digital education for employees and pre-filteration of data at entry of organisation can be applied for better results.

Cite - MACHINE LEARNING METHODS FOR SPAM E-MAIL by W.A. Awad .