With the rising number of people using social networks after the pandemic of COVID-19, cybercriminals took the advantage of (i) the increased base of possible victims and (ii) the use of a trending topic as the pandemic COVID-19 to lure victims and attract their attention and put malicious content to infect the most possible number of people. Twitter platform forces an auto-shortening to any included URL within a 140-character message called “tweet” and this makes it easier for the attackers to include malicious URLs within Tweets. Here comes the need to adopt new approaches to resolve the problem or at least identify it to better understand it to find a suitable solution. One of the proven effective approaches is the adaption of machine learning (ML) concepts and applying different algorithms to detect, identify, and even block the propagation of malware. Hence, this study’s main objectives were to collect tweets from Twitter that are related to the topic of COVID-19 and extract features from these tweets and import them as independent variables for the machine learning models to be developed later, so they would identify imported tweets as to be malicious or not.