IPC분류정보
국가/구분 |
United States(US) Patent
등록
|
국제특허분류(IPC7판) |
|
출원번호 |
US-0886757
(2013-05-03)
|
등록번호 |
US-8706659
(2014-04-22)
|
발명자
/ 주소 |
- Mann, Gideon S.
- Breckenridge, Jordan M.
- Lin, Wei-Hao
|
출원인 / 주소 |
|
대리인 / 주소 |
|
인용정보 |
피인용 횟수 :
7 인용 특허 :
81 |
초록
▼
Methods, systems, and apparatus, including computer programs encoded on one or more computer storage devices, for training a predictive model. In one aspect, a method includes receiving over a network predictive modeling training data from a client computing system. The training data and multiple tr
Methods, systems, and apparatus, including computer programs encoded on one or more computer storage devices, for training a predictive model. In one aspect, a method includes receiving over a network predictive modeling training data from a client computing system. The training data and multiple training functions obtained from a repository of training functions are used to train multiple predictive models. A score is generated for each of the trained predictive models, where each score represents an estimation of the effectiveness of the respective trained predictive model. A first trained predictive model is selected from among the trained predictive models based on the generated scores. Access to the first trained predictive model is provided to the client computing system.
대표청구항
▼
1. A computer-implemented method comprising: receiving input data and a request for a predictive output from a client computing system;generating the predictive output using a first trained predictive model from a repository of trained predictive models and a second trained predictive model from the
1. A computer-implemented method comprising: receiving input data and a request for a predictive output from a client computing system;generating the predictive output using a first trained predictive model from a repository of trained predictive models and a second trained predictive model from the repository of trained predictive models;providing the predictive output to the client computing system; andapportioning a fee received from the client computing system to generate the predictive output to a first client computing system and to a second client computing system, wherein the first client computing system provided a first set of training data that was used to train the first trained predictive model and the second client computing system provided a second set of training data that was used to train the second trained predictive model. 2. The method of claim 1, further comprising: receiving a plurality of sets of predictive modeling training data from a first plurality of client computing systems;for each set of plurality of sets of predictive modeling training data: using the training data and a plurality of training functions to train a plurality of predictive models;generating a respective score for each of the plurality of trained predictive models, where each score represents an estimation of the effectiveness of the trained predictive model; andselecting a trained predictive model from among the plurality of trained predictive models based on the generated scores;wherein a plurality of trained predictive models are thereby generated and selected;storing the plurality of selected trained predictive models in the repository of trained predictive models. 3. The method of claim 2 wherein one of the training functions is maximum likelihood. 4. The method of claim 2, further comprising: providing a web page accessible by the client computing system that is configured to browse the repository of trained predictive models. 5. The method of claim 4 wherein the repository includes descriptions of types of input data received by each of the trained predictive models and types of predictive output generated by each of the trained predictive models. 6. The method of claim 1, further comprising: receiving data from the first client computing system indicating that permission is granted to the client computing system for the client computing system to access the first trained predictive model. 7. The method of claim 1 wherein the each trained predictive models is one of: a linear regression model, a logistic regression model, a regression tree model, multivariate adaptive regression spline model, a Naïve Bayes model, k-nearest neighbor model, a Support Vector Machine, or a perceptron model. 8. A system comprising: data processing apparatus programmed to perform operations comprising: receiving input data and a request for a predictive output from a client computing system;generating the predictive output using a first trained predictive model from a repository of trained predictive models and a second trained predictive model from the repository of trained predictive models;providing the predictive output to the client computing system; andapportioning a fee received from the client computing system to generate the predictive output to a first client computing system and to a second client computing system, wherein the first client computing system provided a first set of training data that was used to train the first trained predictive model and the second client computing system provided a second set of training data that was used to train the second trained predictive model. 9. The system of claim 8, wherein the operations further comprise: receiving a plurality of sets of predictive modeling training data from a first plurality of client computing systems;for each set of plurality of sets of predictive modeling training data: using the training data and a plurality of training functions to train a plurality of predictive models;generating a respective score for each of the plurality of trained predictive models, where each score represents an estimation of the effectiveness of the trained predictive model; andselecting a trained predictive model from among the plurality of trained predictive models based on the generated scores;wherein a plurality of trained predictive models are thereby generated and selected;storing the plurality of selected trained predictive models in the repository of trained predictive models. 10. The system of claim 9 wherein one of the training functions is maximum likelihood. 11. The system of claim 9, wherein the operations further comprise: providing a web page accessible by the client computing system that is configured to browse the repository of trained predictive models. 12. The system of claim 11 wherein the repository includes descriptions of types of input data received by each of the trained predictive models and types of predictive output generated by each of the trained predictive models. 13. The system of claim 8, wherein the operations further comprise: receiving data from the first client computing system indicating that permission is granted to the client computing system for the client computing system to access the first trained predictive model. 14. The system of claim 8 wherein the each trained predictive models is one of: a linear regression model, a logistic regression model, a regression tree model, multivariate adaptive regression spline model, a Naïve Bayes model, k-nearest neighbor model, a Support Vector Machine, or a perceptron model. 15. A computer-readable medium having instructions stored thereon that, when executed by data processing apparatus, cause the data processing apparatus to perform operations comprising: receiving input data and a request for a predictive output from a client computing system;generating the predictive output using a first trained predictive model from a repository of trained predictive models and a second trained predictive model from the repository of trained predictive models;providing the predictive output to the client computing system; andapportioning a fee received from the client computing system to generate the predictive output to a first client computing system and to a second client computing system, wherein the first client computing system provided a first set of training data that was used to train the first trained predictive model and the second client computing system provided a second set of training data that was used to train the second trained predictive model. 16. The computer-readable medium of claim 15, wherein the operations further comprise: receiving a plurality of sets of predictive modeling training data from a first plurality of client computing systems;for each set of plurality of sets of predictive modeling training data: using the training data and a plurality of training functions to train a plurality of predictive models;generating a respective score for each of the plurality of trained predictive models, where each score represents an estimation of the effectiveness of the trained predictive model; andselecting a trained predictive model from among the plurality of trained predictive models based on the generated scores;wherein a plurality of trained predictive models are thereby generated and selected;storing the plurality of selected trained predictive models in the repository of trained predictive models. 17. The computer-readable medium of claim 16 wherein one of the training functions is maximum likelihood. 18. The computer-readable medium of claim 16, wherein the operations further comprise: providing a web page accessible by the client computing system that is configured to browse the repository of trained predictive models. 19. The computer-readable medium of claim 18 wherein the repository includes descriptions of types of input data received by each of the trained predictive models and types of predictive output generated by each of the trained predictive models. 20. The computer-readable medium of claim 15, wherein the operations further comprise: receiving data from the first client computing system indicating that permission is granted to the client computing system for the client computing system to access the first trained predictive model. 21. The computer-readable medium of claim 15 wherein the each trained predictive models is one of: a linear regression model, a logistic regression model, a regression tree model, multivariate adaptive regression spline model, a Naïve Bayes model, k-nearest neighbor model, a Support Vector Machine, or a perceptron model.
※ AI-Helper는 부적절한 답변을 할 수 있습니다.