Transforming categorical features to numerical data

One or more non-transitory computer-readable storage media storing one or more instructions that, when executable by one or more computers, cause the one or more computers to perform operations comprising:

• generating, by a first party participating in a secure multi-party computation (MPC), a first table based on a set of data having first categorical features in a first order, wherein the first table comprises one or more columns indicating labels of the first categorical features in the first order, and one or more columns indicating counts of the first categorical features in the first order;
• receiving, from a second party participating in the secure MPC, a second table based on a secret sharing technique, wherein the second table comprises one or more columns indicating labels of a second categorical features in a second order, and one or more columns indicating counts of the second categorical features in the second order;
• combining, by the first party, the first table and the second table to a combined table comprising the first categorical features and the second categorical features in a combined order, wherein the combined table comprises one or more columns indicating labels of the first categorical features and the second categorical features in the combined order, and one or more columns indicating counts of the first categorical features and the second categorical features in the combined order;
• and
• calculating numerical values for each of the first categorical features and the second categorical features by utilizing the combined table.

A computer-implemented system, comprising:

• one or more computers;
• and
• one or more computer memory devices interoperably coupled with the one or more computers and having non-transitory computer-readable storage media storing one or more instructions that, when executed by the one or more computers, perform one or more operations comprising:
• generating, by a first party participating in a secure multi-party computation (MPC), a first table based on a set of data having first categorical features in a first order, wherein the first table comprises one or more columns indicating labels of the first categorical features in the first order, and one or more columns indicating counts of the first categorical features in the first order;
• receiving, from a second party participating in the secure MPC, a second table based on a secret sharing technique, wherein the second table comprises one or more columns indicating labels of a second categorical features in a second order, and one or more columns indicating counts of the second categorical features in the second order;
• combining, by the first party, the first table and the second table to a combined table comprising the first categorical features and the second categorical features in a combined order, wherein the combined table comprises one or more columns indicating labels of the first categorical features and the second categorical features in the combined order, and one or more columns indicating counts of the first categorical features and the second categorical features in the combined order;
• and
• calculating numerical values for each of the first categorical features and the second categorical features by utilizing the combined table.