Object recognition method and apparatus, electronic device and storage medium

An electronic device, comprising a memory, a processor, and a computer program stored on the memory that, when executed by the processor, causes the electronic device to perform an object recognition method including:

• obtaining relevant object data of a target object, wherein the target object is an account of a social networking platform and the relevant object data represents transactions between the account and other accounts on the social networking platform;
• predicting a first label of the target object by an object recognition model based on the relevant object data of the target object, the first label representing an object type among a plurality of object types and each object type representing a particular type of fraudulent transactions on the social networking platform, wherein the object recognition model is a neural network that has been pre-trained based on a training data set obtained from the social networking platform;
• obtaining a reference data set, the reference data set comprising relevant object data and second labels of a plurality of first sample objects with annotation labels, a first sample object of the plurality of first sample objects having an annotation label of the first sample object representing a real object type among the plurality of object types, and a second label of the first sample object representing a probability that the first sample object belongs to each of the plurality of object types, wherein the annotation label of the first sample object is a vector of n elements that has one element being 1 and (n−1) elements being 0 and the second label of the first sample object is also a vector of n elements and a sum of the elements in the second label being 1;
• determining first association relationships between the target object and the plurality of first sample objects according to the relevant object data of the target object and the relevant object data of the plurality of first sample objects, wherein a first association relationship of the first association relationships between the target object and a first sample object of the plurality of first sample objects indicates a probability that the target object and the first sample object perform same types of fraudulent transactions on the social networking platform;
• determining a second label of the target object according to the first label of the target object, the corresponding annotation labels, the corresponding second labels and the corresponding first association relationships of the plurality of first sample objects as a recognition result of the target object;
• and
• preventing the target object from performing predefined transactions with other accounts of the social networking platform using the second label of the target object.

A non-transitory computer-readable storage medium storing a computer program that, when executed by a processor of an electronic device, causes the electronic device to perform an object recognition method including:

• obtaining relevant object data of a target object, wherein the target object is an account of a social networking platform and the relevant object data represents transactions between the account and other accounts on the social networking platform;
• predicting a first label of the target object by an object recognition model based on the relevant object data of the target object, the first label representing an object type among a plurality of object types and each object type representing a particular type of fraudulent transactions on the social networking platform, wherein the object recognition model is a neural network that has been pre-trained based on a training data set obtained from the social networking platform;
• obtaining a reference data set, the reference data set comprising relevant object data and second labels of a plurality of first sample objects with annotation labels, a first sample object of the plurality of first sample objects having an annotation label of the first sample object representing a real object type among the plurality of object types, and a second label of the first sample object representing a probability that the first sample object belongs to each of the plurality of object types, wherein the annotation label of the first sample object is a vector of n elements that has one element being 1 and (n−1) elements being 0 and the second label of the first sample object is also a vector of n elements and a sum of the elements in the second label being 1;
• determining first association relationships between the target object and the plurality of first sample objects according to the relevant object data of the target object and the relevant object data of the plurality of first sample objects, wherein a first association relationship of the first association relationships between the target object and a first sample object of the plurality of first sample objects indicates a probability that the target object and the first sample object perform same types of fraudulent transactions on the social networking platform;
• determining a second label of the target object according to the first label of the target object, the corresponding annotation labels, the corresponding second labels and the corresponding first association relationships of the plurality of first sample objects as a recognition result of the target object;
• and
• preventing the target object from performing predefined transactions with other accounts of the social networking platform using the second label of the target object.