Method, apparatus, electronic device and storage medium for predicting user attribute

An operating method of an electronic device comprising:

• obtaining demographic-related information of a user;
• identifying one or more features corresponding to the demographic-related information of the user and identifying one or more weights corresponding to the one or more features, wherein the one or more features comprise at least one of:
• a single feature type that corresponds to the demographic-related information, or a cross feature type that corresponds to at least two single feature types, the at least two single feature types including a first single feature type that is different from a second single feature type;
• and
• predicting a user attribute based on the one or more features and the one or more weights corresponding to the one or more features,
• wherein the identifying the one or more features comprises:
• performing an outer product operation on one or more input vectors corresponding to the one or more features;
• and
• determining a result of the outer product operation as one or more inputs of the one or more weights corresponding to the one or more features.

An electronic device comprising:

• a network communication interface;
• and
• a processor configured to:
• obtain demographic-related information of a user;
• identify a plurality of features corresponding to the demographic-related information of the user;
• identifying a plurality of weights corresponding to the plurality of features that have a single feature type that corresponds to the demographic-related information, and a cross feature type that represents interaction information between at least two different single feature types, the at least two different single feature types including a first single feature type that is different from a second single feature type;
• and
• predict a user attribute based on the plurality of features and the plurality of weights corresponding to the plurality of features,
• wherein a feature of the cross feature type is represented by a correlation that indicates a relationship between a first input of the first single feature type and a second input of the second single feature type.

A non-transitory computer readable storage medium, wherein the non-transitory computer readable storage medium stores a computer program, which when executed by a processor, causes the processor to:

• obtain demographic-related information of a user;
• identify a plurality of features corresponding to the demographic-related information of the user and identifying a plurality of weights corresponding to the plurality of features that have a single feature type that corresponds to the demographic-related information, and a cross feature type that represents interaction information between at least two different single feature types, the at least two different single feature types including a first single feature type that is different from a second single feature type;
• and
• predict a user attribute based on the plurality of features and the plurality of weights corresponding to the plurality of features,
• wherein a feature of the cross feature type is represented by a correlation that indicates a relationship between a first input of the first single feature type and a second input of the second single feature type.