The machine learning task of learning a function that translates an input to an output supported example input-output pairs is understood as supervised learning. It uses labelled training data and a set of training examples to infer a function. Each example in supervised learning is formed from an input object (usually a vector) and a desired output value (also called the supervisory signal). A supervised learning algorithm examines the training data and generates an inferred function which will be applied to fresh cases.
Unsupervised learning is a sort of machine learning method that is used to make conclusions from datasets containing unlabeled input data. Cluster analysis is the most frequent unsupervised learning method, which is used for exploratory data analysis to uncover hidden patterns or groupings in data.
Because the task you specified already has labelled datasets (with genuine reviews for each input), you can always treat it as a supervised learning issue and solve it with a machine learning model like SVM, Random Forest, or MLP.