In the modern world, businesses are very much affected by the reviews customers, especially, the frequent customers, put up online. Understanding the sentiment and performance of the restaurant makes it easier for better decision-making. Thus we propose a method in which the aspect-based opinion mining of the interpretive, linguistic customer rating is done using Natural Language Processing (NLP). In this paper, we use popular machine learning models; Decision Trees, Artificial Neural Networks (ANN), Logistic Regression Classification, K-Nearest Neighbors (KNN), Support Vector Machine (SVM), Random Forest, and Naive Bayes algorithms. We aim to compare the various models and find the most definitive and conclusive model with respect to several evaluation metrics. Data collection is done from various social media networking sites like Twitter (now X) or Instagram, and also from the commercial website of the target restaurant, if available. Based on feature selection this study also puts forward different kinds of aspect sentiment on restaurant survival prediction and helps them cater according to the customer needs.