Information processing method, system, and storage medium for setting a matrix used for learning

An information processing system comprising:

• at least one memory configured to store instructions;
• and
• at least one processor configured to execute the instructions to:
• extract, from each column of a target matrix, a subset of elements from among all elements of the column, wherein the target matrix is a matrix including an action status on an item in each of a plurality of setting statuses as an element of a column;
• calculate a degree of similarity between the items based on the subset of elements of the each column;
• set a weight matrix including a weight corresponding to an intersection of items as an element, wherein the weight matrix is a sparse matrix including a nonzero element based on the degree of similarity and the weight matrix is multiplied by the target matrix;
• learn a value of the nonzero element of the weight matrix so that a product of the target matrix and the weight matrix becomes substantially identical to the target matrix;
• and
• predict an item to recommend to a user from a newly acquired user-item matrix by using the weight matrix.

A non-transitory computer-readable storage medium in which a computer program is stored, the computer program comprising instructions for causing an information processing apparatus to execute:

• a process to extract, from each column of a target matrix, a subset of elements from among all elements of the column, wherein the target matrix is a matrix including an action status on an item in each of a plurality of setting statuses as an element of a column;
• a process to calculate a degree of similarity between the items based on the subset of elements of the each column;
• a process to set a weight matrix including a weight corresponding to an intersection of items as an element, wherein the weight matrix is a sparse matrix including a nonzero element based on the degree of similarity and the weight matrix is multiplied by the target matrix;
• a process to learn a value of the nonzero element of the weight matrix so that a product of the target matrix and the weight matrix becomes substantially identical to the target matrix;
• and
• a process to predict an item to recommend to a user from a newly acquired user-item matrix by using the weight matrix.