System, method, and program for predicting information

A method for predicting information using a system that includes

• a learning object storing section that stores objects to be learned,
• a learning result storing section that stores learning results, and
• a control section connected to an input section,
• the method causes the control section to perform:
• learning processing in which the control section
• computes a first feature vector and a second feature vector for each object to be learned stored in the learning object storing section,
• generates a row vector by dividing a principal component coefficient vector of the first feature vector and a principal component coefficient vector of the second feature vector by an index representing variation of the principal component coefficient vectors,
• by performing principal component analysis using the row vectors of all of the objects to be learned, generates a principal component basis vector based on the the principal component coefficient vector of the first feature vector and the principal component coefficient vector of the second feature vector, and
• stores the principal component basis vector in the learning result storing section, together with an average vector of the first feature vector and an average vector of the second feature vector, and
• prediction processing, in which the control section
• computes a principal component coefficient vector of a first feature vector of an object to be processed that is designated by the input section,
• computes a principal component coefficient vector of a second feature vector using the principal component basis vector stored in the learning result storing section, and
• using the principal component coefficient vector of the second feature vector, computes a second feature vector of the object to be processed.

A non-transitory computer readable storage medium storing a program for predicting information using a system that includes

• a learning object storing section that stores objects to be learned,
• a learning result storing section that stores learning results, and
• a control section connected to an input section,
• when the program is executed, the control section performs
• learning processing in which the control section
• computes a first feature vector and a second feature vector for each object to be learned stored in the learning object storing section,
• generates a row vector by dividing a principal component coefficient vector of the first feature vector and a principal component coefficient vector of the second feature vector by an index representing variation of the principal component coefficient vectors,
• by performing principal component analysis using the row vectors of all of the objects to be learned, generates a principal component basis vector based on the the principal component coefficient vector of the first feature vector and the principal component coefficient vector of the second feature vector, and
• stores the principal component basis vector in the learning result storing section, together with an average vector of the first feature vector and an average vector of the second feature vector, and
• prediction processing, in which the control section
• computes a principal component coefficient vector of a first feature vector of an object to be processed that is designated by the input section,
• computes a principal component coefficient vector of a second feature vector using the principal component basis vector stored in the learning result storing section, and
• using the principal component coefficient vector of the second feature vector, computes a second feature vector of the object to be processed.