Feature and case importance and confidence for imputation in computer-based reasoning systems

1. A method comprising:

performing the following until there are no more cases in a computer-based reasoning model with missing fields for which imputation is desired:

determining which cases have fields to impute in the computer-based reasoning model;

determining conviction scores for the features that have data to impute in the computer-based reasoning model based on a certainty function associated with:

removing a feature from the cases in the computer-based reasoning model;

adding the feature back into the computer-based reasoning mode, wherein the certainty function is associated with a certainty that a particular set of data fits a model,

wherein the certainty function is associated with a certainty that a particular set of data fits a model;

determining for which one or more cases with missing fields to impute data for the missing fields based on the conviction scores, wherein determining for which one or more cases with missing fields to impute data for the missing fields based on the conviction scores comprises:

determining which particular feature of multiple features has a highest conviction score, and

determining to impute data for the missing fields for the one or more cases that are missing a value for the particular feature that has the highest conviction score;

for each of the determined one or more cases with the missing fields to impute:

determining imputed data for a missing field of the missing fields based on the case, and an imputation model, and the missing fields in the case;

modifying the case with the imputed data, wherein the modified case becomes part of the computer-based reasoning model in place of the original case to create an updated computer-based reasoning model,

wherein the method is performed by one or more computing devices.

2. The method of claim 1 , wherein determining for which one or more cases with missing fields to impute data comprises:

determining a case with the highest conviction score.

3. The method of claim 1 , wherein determining for which one or more cases with missing fields to impute data comprises:

determining two or more cases with the highest conviction score.

4. The method of claim 1 , wherein determining imputed data for the missing field comprises:

determining the imputed data based on a machine learning model for the computer-based reasoning model's data, wherein the machine learning model for the computer-based reasoning model's data has been trained using the data in the computer-based reasoning model.

5. The method of claim 4 , further comprising:

determining an update to the machine learning model based on the updated computer-based reasoning model.

6. One or more non-transitory storage media storing instructions which, when executed by one or more computing devices, cause performance of a method of:

performing the following until there are no more cases in a computer-based reasoning model with missing fields for which imputation is desired:

determining which cases have fields to impute in the computer-based reasoning model;

determining conviction scores for the features that have data to impute in the computer-based reasoning model based on a certainty function associated with:

removing a feature from the cases in the computer-based reasoning model;

adding the feature back into the computer-based reasoning mode, wherein the certainty function is associated with a certainty that a particular set of data fits a model,

wherein the certainty function is associated with a certainty that a particular set of data fits a model;

determining for which one or more cases with the missing fields to impute data for the missing fields based on the conviction scores, wherein determining for which one or more cases with missing fields to impute data for the missing fields based on the conviction scores comprises:

determining which particular feature of multiple features has a highest conviction score, and

determining to impute data for the missing fields for the one or more cases that are missing a value for the particular feature that has the highest conviction score;

for each of the determined one or more cases with missing fields to impute:

determining imputed data for a missing field of the missing fields based on the case, and an imputation model, and the missing fields in the case;

modifying the case with the imputed data, wherein the modified case becomes part of the computer-based reasoning model in place of the original case to create an updated computer-based reasoning model.

7. The one or more non-transitory storage media of claim 6 , wherein determining for which one or more cases with missing fields to impute data comprises:

determining a case with a highest conviction score among the one or more cases.

8. The one or more non-transitory storage media of claim 6 , wherein determining for which one or more cases with missing fields to impute data comprises:

determining two or more cases with a highest conviction among the one or more cases.

9. The one or more non-transitory storage media of claim 6 , wherein determining imputed data for the missing field comprises:

determining the imputed data based on a machine learning model for the computer-based reasoning model's data, wherein the machine learning model for the computer-based reasoning model's data has been trained using the data in the computer-based reasoning model.

10. The one or more non-transitory storage media of claim 9 , the method further comprising:

determining an update to the machine learning model based on the updated computer-based reasoning model.

11. A system comprising one or more computing devices, which one or more computing devices are configured to perform a method of:

performing the following until there are no more cases in a computer-based reasoning model with missing fields for which imputation is desired:

determining which cases have fields to impute in the computer-based reasoning model;

determining conviction scores for the features that have data to impute in the computer-based reasoning model based on a certainty function associated with:

removing a feature from the cases in the computer-based reasoning model;

adding the feature back into the computer-based reasoning mode, wherein the certainty function is associated with a certainty that a particular set of data fits a model,

wherein the certainty function is associated with a certainty that a particular set of data fits a model;

determining for which one or more cases with the missing fields to impute data for the missing fields based on the conviction scores, wherein determining for which one or more cases with missing fields to impute data for the missing fields based on the conviction scores comprises:

determining which particular feature of multiple features has a highest conviction score, and

determining to impute data for the missing fields for the one or more cases that are missing a value for the particular feature that has the highest conviction score;

for each of the determined one or more cases with missing fields to impute:

determining imputed data for a missing field of the missing fields based on the case, and an imputation model, and the missing fields in the case;

modifying the case with the imputed data, wherein the modified case becomes part of the computer-based reasoning model in place of the original case to create an updated computer-based reasoning model.

12. The system of claim 11 , wherein determining imputed data for the missing field comprises:

determining the imputed data based on a machine learning model for the computer-based reasoning model's data, wherein the machine learning model for the computer-based reasoning model's data has been trained using the data in the computer-based reasoning model.

13. The system of claim 11 , wherein determining for which one or more cases with missing fields to impute data comprises:

determining a case with a highest conviction score among the one or more cases.

14. The system of claim 13 , further comprising:

determining to impute data for the missing fields for the one or more cases with a highest conviction score among cases that are missing the particular feature that has the highest conviction score.