Method and device for virtual resource allocation, modeling, and data prediction
Evaluation results of a plurality of users are received from a plurality of data providers. The evaluation results are obtained by the plurality of data providers evaluating the plurality of users based on evaluation models of the plurality of data providers. A plurality of training samples is constructed by using the evaluation results. Each training sample includes a respective subset of the evaluation results corresponding to a same user of the plurality of users. A label for each training sample is generated based on an actual service execution status of the same user. A model is trained based on the plurality of training samples and the plurality of labels, including setting a plurality of variable coefficients, each variable coefficient specifying a contribution level of a corresponding data provider. Virtual resources to each data provider are allocated based on the plurality of variable coefficients.
1. A computer-implemented method, comprising:
receiving, from a plurality of data providers, a plurality of user evaluation results of a plurality of users, generated by a plurality of user evaluation models, respectively, wherein each user evaluation model is trained on a corresponding training sample set by
generating, for a corresponding user data sample, a respective data feature vector comprising data feature values, wherein the data feature values correspond to data features of a plurality of dimensions that are extracted from the user data sample, and
constructing a target matrix based on the data feature vectors generated for the user data sample;
constructing a plurality of risk evaluation model training samples from the user evaluation results, wherein each risk evaluation model training sample of the plurality of risk evaluation model training samples comprises a respective subset of the user evaluation results corresponding to a first user of the plurality of users;
generating a label for each risk evaluation model training sample of the plurality of risk evaluation model training samples based on an actual service execution status of the first user to provide a plurality of labels;
training a risk evaluation model based on the plurality of risk evaluation model training samples and the plurality of labels, wherein training the risk evaluation model comprises setting a plurality of variable coefficients, each variable coefficient specifying a contribution level of a corresponding data provider of the plurality of data providers; and
allocating virtual resources to the plurality of data providers based on the plurality of variable coefficients.
2. The computer-implemented method of claim 1 , wherein the trained risk evaluation model is a linear model.
3. The computer-implemented method of claim 1 , further comprising:
receiving user evaluation results of a particular user that are uploaded by the plurality of data providers, and inputting the user evaluation results of the particular user to the trained risk evaluation model to obtain a final user evaluation result of the particular user.
4. The computer-implemented method of claim 1 , wherein a number of the virtual resources allocated to the plurality of data providers is directly proportional to the plurality of variable coefficients.
5. The computer-implemented method of claim 4 , wherein the virtual resources comprise a user data usage fund distributed to the plurality of data providers.
6. The computer-implemented method of claim 1 , wherein the plurality of user evaluation results comprise risk scores, and the label indicates whether the respective user is a risky user.
7. The computer-implemented method of claim 1 , wherein the trained user evaluation model comprises a machine learning model.
8. The computer-implemented method of claim 7 , wherein the machine learning model comprises a logistic regression model.
9. A non-transitory, computer-readable medium storing one or more instructions executable by a computer system to perform operations comprising:
receiving, from a plurality of data providers, a plurality of user evaluation results of a plurality of users, generated by a plurality of user evaluation models, respectively, wherein each user evaluation model is trained on a corresponding training sample set by
generating, for a corresponding user data sample, a respective data feature vector comprising data feature values, wherein the data feature values correspond to data features of a plurality of dimensions that are extracted from the user data sample, and
constructing a target matrix based on the data feature vectors generated for the user data sample;
constructing a plurality of risk evaluation model training samples from the user evaluation results, wherein each risk evaluation model training sample of the plurality of risk evaluation model training samples comprises a respective subset of the user evaluation results corresponding to a first user of the plurality of users;
generating a label for each risk evaluation model training sample of the plurality of risk evaluation model training samples based on an actual service execution status of the first user to provide a plurality of labels;
training a risk evaluation model based on the plurality of risk evaluation model training samples and the plurality of labels, wherein training the risk evaluation model comprises setting a plurality of variable coefficients, each variable coefficient specifying a contribution level of a corresponding data provider of the plurality of data providers; and
allocating virtual resources to the plurality of data providers based on the plurality of variable coefficients.
10. The non-transitory, computer-readable medium of claim 9 , wherein the trained risk evaluation model is a linear model.
11. The non-transitory, computer-readable medium of claim 9 , wherein the operations further comprise:
receiving user evaluation results of a particular user that are uploaded by the plurality of data providers, and inputting the user evaluation results of the particular user to the trained risk evaluation model to obtain a final user evaluation result of the particular user.
12. The non-transitory, computer-readable medium of claim 9 , wherein a number of the virtual resources allocated to the plurality of data providers is directly proportional to the plurality of variable coefficients.
13. The non-transitory, computer-readable medium of claim 12 , wherein the virtual resources comprise a user data usage fund distributed to the plurality of data providers.
14. The non-transitory, computer-readable medium of claim 9 , wherein the plurality of user evaluation results comprise risk scores, and the label indicates whether the respective user is a risky user.
15. The non-transitory, computer-readable medium of claim 9 , wherein the trained user evaluation model comprises a machine learning model.
16. The non-transitory, computer-readable medium of claim 15 , wherein the machine learning model comprises a logistic regression model.
17. 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 tangible, non-transitory, machine-readable media storing one or more instructions that, when executed by the one or more computers, perform operations comprising:
receiving, from a plurality of data providers, a plurality of user evaluation results of a plurality of users, generated by a plurality of user evaluation models, respectively, wherein each user evaluation model is trained on a corresponding training sample set by
generating, for a corresponding user data sample, a respective data feature vector comprising data feature values, wherein the data feature values correspond to data features of a plurality of dimensions that are extracted from the user data sample, and
constructing a target matrix based on the data feature vectors generated for the user data sample;
constructing a plurality of risk evaluation model training samples from the user evaluation results, wherein each risk evaluation model training sample of the plurality of risk evaluation model training samples comprises a respective subset of the user evaluation results corresponding to a first user of the plurality of users;
generating a label for each risk evaluation model training sample of the plurality of risk evaluation model training samples based on an actual service execution status of the first user to provide a plurality of labels;
training a risk evaluation model based on the plurality of risk evaluation model training samples and the plurality of labels, wherein training the risk evaluation model comprises setting a plurality of variable coefficients, each variable coefficient specifying a contribution level of a corresponding data provider of the plurality of data providers; and
allocating virtual resources to the plurality of data providers based on the plurality of variable coefficients.
18. The computer-implemented system of claim 17 , wherein the trained risk evaluation model is a linear model.
19. The computer-implemented system of claim 17 , wherein the operations further comprise:
receiving user evaluation results of a particular user that are uploaded by the plurality of data providers, and inputting the user evaluation results of the particular user to the trained risk evaluation model to obtain a final user evaluation result of the particular user.
20. The computer-implemented system of claim 17 , wherein a number of the virtual resources allocated to the plurality of data providers is directly proportional to the plurality of variable coefficients.
21. The computer-implemented system of claim 20 , wherein the virtual resources comprise a user data usage fund distributed to the plurality of data providers.
22. The computer-implemented system of claim 17 , wherein the plurality of user evaluation results comprise risk scores, and the label indicates whether the respective user is a risky user.
23. The computer-implemented system of claim 17 , wherein the trained user evaluation model comprises a machine learning model.
24. The computer-implemented system of claim 23 , wherein the machine learning model comprises a logistic regression model.