MACHINE LEARNING DEVICE, LEARNED MODEL, DATA STRUCTURE, PERIODONTAL DISEASE INSPECTION METHOD, PERIODONTAL DISEASE INSPECTION SYSTEM AND PERIODONTAL DISEASE INSPECTION KIT

摘要

Provided is a machine learning device that learns a function for determining the probability of contracting periodontal disease or the state of periodontal disease, the machine learning device including a learning unit that learns the function for determining a determination value on the basis of a plurality of learning data items comprising oral cavity information detected from a specimen, and at least one determination value selected from a group consisting of periodontal disease contracting probabilities or periodontal disease states of the person who provided the specimen. The learning unit includes: a compensation calculation unit that uses the function to calculate, on the basis of the learning data items, a compensation in respect to the results of determining the periodontal disease contracting probability or periodontal disease state; a function updating unit that updates the function so as to increase the compensation calculated by the compensation calculation unit; and a convergence determination unit that causes the calculation by the compensation calculation unit and the updating by the function updating unit to repeat until a predetermined convergence condition is satisfied.