The COVID-19 has slowly spread all over the world. Hence, to get rid of this deadly virus, vaccination is very important, but for a country like India, vaccinating the whole country is not possible within a very short time. So vaccine prioritization should be done in a very effective way. For an instance, the elderly persons or people having health issues or frontline workers are to be given higher priority for vaccination than other masses. Age and job designation are not the only attributes that affect a person’s chance of getting infected. Covacdisor has been specifically developed for this purpose using machine learning (ML), where the infection risk factor is predicted. The prediction would help in the proper prioritization of vaccines. The predicted risk factor is dependent on 24 parameters. These parameters directly affect a person’s immunity. A dataset has been proposed with these 24 parameters. Support vector machine (SVM), K-nearest neighbour (KNN), logistics regression (LR), and random forest (RF) have been used for training the model on the proposed dataset and got the highest accuracy of 0.85 from RF. Random forest is applied on the backend of the Web Application which is acting as a user interface and predicts the risk group of the user. With this proposed technique, prediction of the urgency of a user to get vaccinated can be done, which would help in achieving herd immunity faster by prioritizing the vaccination of the vulnerable population.