PROBLEM TO BE SOLVED: To easily grasp intake of a subject.SOLUTION: A prediction system according to one embodiment comprises: an acquisition part； a determination part； an estimation part； and an output part. The acquisition part acquires a living body signal which is generated by deglutition movement by a subject. The determination part creates first conversion data based on the living body signal, and executes first machine learning in which the first conversion data is an explanatory variable, for determining whether or not the subject ingested an intake object. The estimation part creates second conversion data based on the living body signal when it is determined that the subject ingested, and executes second machine learning in which the second conversion data is an explanatory variable, for estimating an intake amount of the intake object of the subject. The output part outputs the estimated intake amount.
