Training detection model using output of language model applied to event information

A computer program product comprising a non-transitory computer readable storage medium having instructions encoded thereon that, when executed by a processor, cause the processor to perform steps comprising:

• requesting a language model of a model serving system in communication with a computer system to determine, based on a prompt input into the language model, information about a refund event for a first order placed with the computer system by a user of the computer system;
• importing, from the model serving system and at a detection module of the computer system, the information about the refund event;
• accessing, via the detection module, a machine-learning model of the computer system, wherein the machine-learning model is trained to detect a fraudulent behavior associated with an order placed with the computer system;
• applying, via the detection module, the machine-learning model to the information about the refund event imported from the model serving system to generate a score associated with the refund event;
• determining, by the detection module and using the score, whether the refund event was due to a fraudulent behavior of the user;
• and
• performing at least one action associated with the computer system, based on the determination whether the refund event was due to the fraudulent behavior.

A computer system comprising:

• a processor;
• and
• a non-transitory computer-readable storage medium having instructions that, when executed by the processor, cause the computer system to perform steps comprising:
• requesting a language model of a model serving system in communication with the computer system to determine, based on a prompt input into the language model, information about a refund event for a first order placed with the computer system by a user of the computer system;
• importing, from the model serving system and at a detection module of the computer system, the information about the refund event;
• accessing, via the detection module, a machine-learning model of the computer system, wherein the machine-learning model is trained to detect a fraudulent behavior associated with an order placed with the computer system;
• applying, via the detection module, the machine-learning model to the information about the refund event imported from the model serving system to generate a score associated with the refund event;
• determining, by the detection module and using the score, whether the refund event was due to a fraudulent behavior of the user;
• and
• performing at least one action associated with the computer system, based on the determination whether the refund event was due to the fraudulent behavior.