Systems and methods for data correlation and artifact matching in identity management artificial intelligence systems
Systems and methods for embodiments of artificial intelligence systems for identity management are disclosed. Embodiments of the identity management systems disclosed herein may support the correlation of identities from authoritative source systems and accounts from non-authoritative source systems using artificial intelligence techniques.
1. An identity management system, comprising:
a processor;
a non-transitory, computer-readable storage medium, including computer instructions for:
obtaining identity management data associated with a plurality of source systems, the identity management data comprising data on a set of identity management artifacts, wherein the plurality of source systems include a non-authoritative data source and an authoritative data source and the identity management data comprises account data associated with accounts from the non-authoritative data source and identity data on identities associated with the authoritative data source;
determining a first set of identifiers associated with data of the non-authoritative data source;
determining a second set of identifiers associated with data of the authoritative data source;
forming a set of feature pairs specific to the non-authoritative data source and the authoritative data source wherein each feature pair of the set of feature pairs comprises a first identifier from the first set of identifiers and a second identifier from the second set of identifiers and the set of feature pairs are formed by correlating the first set of identifiers with the second set of identifiers;
generating feature values for each of the feature pairs for a set of account-identity pairs, where each account-identity pair comprises a first account of the accounts of the account data associated with the non-authoritative data source and a first identity of the identities of the identity data associated with the authoritative data source, and generating a feature value for a feature pair is based on a first value associated with the first identifier of the feature pair associated with the first account and a second value for the second identifier of the feature pair associated with the first identity; and
generating predictions for one or more account-identity pairs using a machine learning model (ML), wherein a prediction for an account-identity pair is based on the feature values associated with that account-identity pair, wherein when a prediction is over a threshold for the account-identity pair the account of the account-identify pair is associated with the identity of the account-identity pair.
2. The identity management system of claim 1 , wherein the instructions are further for:
selecting a training set of account-identity pairs associated with the non-authoritative data source and the authoritative data source; and
training the machine learning (ML) model specific to the non-authoritative data source and the authoritative data source based on the training set.
3. The identity management system of claim 2 , wherein the ML model is a Random Forest, an XGBoost model, a Deep Learning model, or a logistic regression model.
4. The identity management system of claim 2 , wherein the training set of account-identity pairs are selected based on feature values associated with the training set of account identity pairs.
5. The identity management system of claim 1 , wherein the first set of identifiers and second set of identifiers are column identifiers.
6. The identity management system of claim 5 , wherein the column identifiers are associated with a schema.
7. The identity management system of claim 1 , wherein the instructions are further for filtering the set of account-identity pairs to determine the one or more account-identity pairs.
8. A method, comprising:
obtaining identity management data associated with a plurality of source systems, the identity management data comprising data on a set of identity management artifacts, wherein the plurality of source systems include a non-authoritative data source and an authoritative data source and the identity management data comprises account data associated with accounts from the non-authoritative data source and identity data on identities associated with the authoritative data source;
determining a first set of identifiers associated with data of the non-authoritative data source;
determining a second set of identifiers associated with data of the authoritative data source;
forming a set of feature pairs specific to the non-authoritative data source and the authoritative data source wherein each feature pair of the set of feature pairs comprises a first identifier from the first set of identifiers and a second identifier from the second set of identifiers and the set of feature pairs are formed by correlating the first set of identifiers with the second set of identifiers;
generating feature values for each of the feature pairs for a set of account-identity pairs, where each account-identity pair comprises a first account of the accounts of the account data associated with the non-authoritative data source and a first identity of the identities of the identity data associated with the authoritative data source, and generating a feature value for a feature pair is based on a first value associated with the first identifier of the feature pair associated with the first account and a second value for the second identifier of the feature pair associated with the first identity; and
generating predictions for one or more account-identity pairs using a machine learning model (ML), wherein a prediction for an account-identity pair is based on the feature values associated with that account-identity pair, wherein when a prediction is over a threshold for the account-identity pair the account of the account-identify pair is associated with the identity of the account-identity pair.
9. The method of claim 8 , further comprising:
selecting a training set of account-identity pairs associated with the non-authoritative data source and the authoritative data source; and
training the machine learning (ML) model specific to the non-authoritative data source and the authoritative data source based on the training set.
10. The method of claim 9 , wherein the ML model is a Random Forest, an XGBoost model, a Deep Learning model, or a logistic regression model.
11. The method of claim 9 , wherein the training set of account-identity pairs are selected based on feature values associated with the training set of account identity pairs.
12. The method of claim 8 , wherein the first set of identifiers and second set of identifiers are column identifiers.
13. The method of claim 12 , wherein the column identifiers are associated with a schema.
14. The method of claim 8 , further comprising filtering the set of account-identity pairs to determine the one or more account-identity pairs.
15. A non-transitory computer readable medium, comprising instructions for:
obtaining identity management data associated with a plurality of source systems, the identity management data comprising data on a set of identity management artifacts, wherein the plurality of source systems include a non-authoritative data source and an authoritative data source and the identity management data comprises account data associated with accounts from the non-authoritative data source and identity data on identities associated with the authoritative data source;
determining a first set of identifiers associated with data of the non-authoritative data source;
determining a second set of identifiers associated with data of the authoritative data source;
forming a set of feature pairs specific to the non-authoritative data source and the authoritative data source wherein each feature pair of the set of feature pairs comprises a first identifier from the first set of identifiers and a second identifier from the second set of identifiers and the set of feature pairs are formed by correlating the first set of identifiers with the second set of identifiers;
generating feature values for each of the feature pairs for a set of account-identity pairs, where each account-identity pair comprises a first account of the accounts of the account data associated with the non-authoritative data source and a first identity of the identities of the identity data associated with the authoritative data source, and generating a feature value for a feature pair is based on a first value associated with the first identifier of the feature pair associated with the first account and a second value for the second identifier of the feature pair associated with the first identity; and
generating predictions for one or more account-identity pairs using a machine learning model (ML), wherein a prediction for an account-identity pair is based on the feature values associated with that account-identity pair, wherein when a prediction is over a threshold for the account-identity pair the account of the account-identify pair is associated with the identity of the account-identity pair.
16. The non-transitory computer readable medium of claim 15 , wherein the instructions are further for:
selecting a training set of account-identity pairs associated with the non-authoritative data source and the authoritative data source; and
training the machine learning (ML) model specific to the non-authoritative data source and the authoritative data source based on the training set.
17. The non-transitory computer readable medium of claim 16 , wherein the ML model is a Random Forest, an XGBoost model, a Deep Learning model, or a logistic regression model.
18. The non-transitory computer readable medium of claim 16 , wherein the training set of account-identity pairs are selected based on feature values associated with the training set of account identity pairs.
19. The non-transitory computer readable medium of claim 15 , wherein the first set of identifiers and second set of identifiers are column identifiers.
20. The non-transitory computer readable medium of claim 19 , wherein the column identifiers are associated with a schema.
21. The non-transitory computer readable medium of claim 15 , wherein the instructions are further for filtering the set of account-identity pairs to determine the one or more account-identity pairs.