E-commerce business has become a prominent entity of global retail as online transaction saves time and cost at the same time. COVID-19 pandemic and lockdown accelerated the growth of e-commerce. The new e-business companies are booming rapidly whereas insincerity in customer concern is noticeable. As a result, the purchasers are facing numerous problems while buying online. The main objective of this study is to predict preferences on online shopping of buyers and based on that analysis, the pattern can be observed. While doing the study, we used some popular Supervised machine learning algorithms such as Decision Tree (DT), K-Nearest Neighbor (KNN), Support Vector Machine (SVM), and Naıve Bayes (NB) algorithm. Amongst those, best accuracy was delivered by the Decision Tree algorithm. The output clearly demonstrates that, people are more likely to participate in online shopping if the obstacles could be alleviate which means, buyers are still not satisfied and confident about the online platform. Hopefully, the result of this study can be a great asset for improving the E-commerce sector of Bangladesh if it is optimized wisely.