Automated dynamic data quality assessment
In general, embodiments of the present invention provide systems, methods and computer readable media for automated dynamic data quality assessment. One aspect of the subject matter described in this specification includes the actions of receiving a data quality job including a new data sample; and, if the new data sample is determined to be added to a reservoir of data samples, sending a quality verification request to an oracle; receiving a new data sample quality estimate from the oracle; and adding the new data sample and estimate to the reservoir. A second aspect of the subject matter includes the actions of receiving, from a predictive model, a judgment associated with a new data sample; analyzing the new data sample based in part on the judgment to determine whether to send a new data sample quality verification request to an oracle; and, if a new data sample quality estimate is received from the oracle, determining whether to add the new data sample and the judgment to the reservoir.
1. A computer-implemented method, comprising:
receiving a data quality job, the data quality job including configuration data and a new data sample having a particular data type, wherein the configuration data comprises an oracle identifier, the oracle identifier indicating a particular oracle to provide a verified quality measure for the new data sample, the particular oracle associated with an attribute of the new data sample;
determining, by a processor, whether to add the new data sample to a reservoir of data samples, the reservoir of data samples identified based at least in part on the particular data type, the determining based at least in part on whether the new data sample statistically belongs in the reservoir of data samples; and
in an instance in which the new data sample is to be added to the reservoir of data samples,
sending, to the particular oracle selected based on the oracle identifier, a quality verification request including the new data sample;
receiving a data quality estimate associated with the new data sample from the oracle in response to the quality verification request, wherein the data quality estimate comprises a quality score calculated based on one or more of a percentage of correctness of the data sample and a percentage of completeness of the data sample; and
adding the new data sample and the associated data quality estimate to the reservoir of data samples in response to receiving the data quality estimate.
2. The method of claim 1 , further comprising:
updating reservoir summary statistics.
3. The method of claim 1 , wherein updating the reservoir summary statistics comprises:
calculating an overall data quality estimate for the reservoir using data quality estimates respectively associated with each of the data samples; and
calculating a statistical variance for the data samples.
4. The method of claim 1 , wherein updating the reservoir summary statistics further comprises:
logging the updated reservoir summary statistics in persistent storage.
5. The method of claim 1 , further comprising:
receiving corpus summary statistics calculated for a corpus of previously collected data samples, wherein each of the previously collected data samples are respectively associated with the particular data type; and
generating an analysis comparing the updated reservoir summary statistics with the corpus summary statistics.
6. The method of claim 1 , wherein determining whether to add the new data sample to the reservoir is further based on a value of at least one attribute of the new data sample.
7. The method of claim 1 , wherein determining whether to add the new data sample to the reservoir is further based on a probabilistic sampling approach.
8. The method of claim 1 , wherein the oracle is a crowd, a flat file of previously received crowd data verification results, or a software system.
9. The method of claim 1 , wherein the new data sample is collected from a data stream.
10. The method of claim 1 , wherein the new data sample is a single data instance or a set of data instances collected from the data stream within a pre-defined time window.
11. The method of claim 1 , wherein the new data sample has been pre-processed by a data cleaning process.
12. A computer program product, stored on a non-transitory computer readable medium, comprising instructions that when executed on one or more computers cause the one or more computers to perform operations comprising:
receiving a data quality job, the data quality job including configuration data and a new data sample having a particular data type, wherein the configuration data comprises an oracle identifier, the oracle identifier indicating a particular oracle to provide a verified quality measure for the new data sample, the particular oracle associated with an attribute of the new data sample;
determining, by a processor, whether to add the new data sample to a reservoir of data samples, the reservoir of data samples identified based at least in part on the particular data type the determining based at least in part on whether the new data sample statistically belongs in the reservoir of data samples; and
in an instance in which the new data sample is to be added to the reservoir of data samples,
sending, to the particular oracle selected based on the oracle identifier, a quality verification request including the new data sample;
receiving a data quality estimate associated with the new data sample from the oracle in response to the quality verification request, wherein the data quality estimate comprises a quality score calculated based on one or more of a percentage of correctness of the data sample and a percentage of completeness of the data sample; and
adding the new data sample and the associated data quality estimate to the reservoir of data samples in response to receiving the data quality estimate.
13. A system, comprising:
one or more computers and one or more storage devices storing instructions that are operable, when executed by the one or more computers, to cause the one or more computers to perform operations comprising:
receiving a data quality job, the data quality job including configuration data and a new data sample having a particular data type, wherein the configuration data comprises an oracle identifier, the oracle identifier indicating a particular oracle to provide a verified quality measure for the new data sample, the particular oracle associated with an attribute of the new data sample;
determining, by a processor, whether to add the new data sample to a reservoir of data samples, the reservoir of data samples identified based at least in part on the particular data type, the determining based at least in part on whether the new data sample statistically belongs in the reservoir of data samples; and
in an instance in which the new data sample is to be added to the reservoir of data samples,
sending, to the particular oracle selected based on the oracle identifier, a quality verification request including the new data sample;
receiving a data quality estimate associated with the new data sample from the oracle in response to the quality verification request, wherein the data quality estimate comprises a quality score calculated based on one or more of a percentage of correctness of the data sample and a percentage of completeness of the data sample; and
adding the new data sample and the associated data quality estimate to the reservoir of data samples in response to receiving the data quality estimate.
14. The system of claim 13 , further comprising:
updating reservoir summary statistics.
15. The system of claim 14 , wherein updating the reservoir summary statistics comprises:
calculating an overall data quality estimate for the reservoir using data quality estimates respectively associated with each of the data samples; and
calculating a statistical variance for the data samples.
16. The system of claim 14 , wherein updating the reservoir summary statistics further comprises:
logging the updated reservoir summary statistics in persistent storage.
17. The system of claim 14 , further comprising:
receiving corpus summary statistics calculated for a corpus of previously collected data samples, wherein each of the previously collected data samples are respectively associated with the particular data type; and
generating an analysis comparing the updated reservoir summary statistics with the corpus summary statistics.
18. The system of claim 13 , wherein determining whether to add the new data sample to the reservoir is further based on a value of at least one attribute of the new data sample.
19. The system of claim 13 , wherein determining whether to add the new data sample to the reservoir is further based on a probabilistic sampling approach.
20. The system of claim 13 , wherein the oracle is a crowd, a flat file of previously received crowd data verification results, or a software system.
21. The system of claim 13 , wherein the new data sample is collected from a data stream.
22. The system of claim 21 , wherein the new data sample is a single data instance or a set of data instances collected from the data stream within a pre-defined time window.
23. The system of claim 13 , wherein the new data sample has been pre-processed by a data cleaning process.