Data security in enrollment management systems
There is a need for improving data security in enrollment management systems. This need can be addressed by, for example, solutions for determining an enrollment recommendation for a primary member profile based on preconfigured enrollment modeling data. In one example, a method includes retrieving enrollment modeling data for a group of member profiles, determining a plurality of related member profiles for the primary member profile from the group of member profiles, determining a cross-member enrollment prediction for the primary member profile by comparing enrollment modeling data of the primary member profile and enrollment modeling data of each related member profile, determining a member-specific enrollment recommendation by comparing enrollment modeling data of the primary member profile and enrollment coverage criteria for each enrollment plan, and determining the enrollment recommendation based on the cross-member enrollment prediction and the member-specific enrollment prediction.
1. A computer-implemented method comprising:
accessing, by one or more processors, a cross-member enrollment prediction model from a cross-member enrollment recommendation unit, wherein the cross-member enrollment prediction model is generated based at least in part on a portion of second enrollment modeling data that comprises one or more transactional records or demographic information for one or more of a plurality of secondary member profiles that are related to a primary member profile based at least in part on first enrollment modeling data for the primary member profile and the second enrollment modeling data for the plurality of secondary member profiles with respect to (i) a transactional frequency category, (ii) one or more dominant transactional patterns, and (iii) one or more demographic features;
determining, by the one or more processors and via the cross-member enrollment recommendation unit, a cross-member enrollment recommendation with respect to the primary member profile based at least in part on the cross-member enrollment prediction model and one or more cross-member predictive features of the first enrollment modeling data;
determining, by the one or more processors and via a member-specific enrollment recommendation unit, a member-specific enrollment recommendation with respect to the primary member profile based at least in part on one or more member-specific features that are determined based at least in part on the first enrollment modeling data and one or more enrollment coverage criteria for a plurality of enrollment plans;
and
providing, by the one or more processors and via a recommendation generation unit to a remote client device, the cross-member enrollment recommendation and the member-specific enrollment recommendation to an enrollment recommendation user interface at the remote client device.
2. The computer-implemented method of claim 1 , wherein the second enrollment modeling data comprises one or more transactional records for the plurality of secondary member profiles and the one or more demographic features.
3. The computer-implemented method of claim 1 further comprising:
determining, for a secondary member profile of the plurality of secondary member profiles, a transactional frequency category based at least in part on one or more transactional records for the secondary member profile;
determining, for the secondary member profile, one or more dominant transactional patterns based at least in part on the one or more transactional records for the secondary member profile; and
determining whether the secondary member profile is related to the primary member profile based at least in part on: (i) whether one or more demographic features for the primary member profile match one or more demographic features for the secondary member profile, (ii) whether a transactional frequency category for the primary member profile matches a transactional frequency category for the secondary member profile, and (iii) whether one or more dominant transactional patterns for the primary member profile match one or more dominant transactional patterns for the secondary member profile.
4. The computer-implemented method of claim 1 further comprising determining one or more training features for a secondary member profile of the plurality of secondary member profiles by determining the one or more training features based at least in part on: (i) one or more predictively critical transactional patterns for the secondary member profile that are based at least in part on one or more transactional patterns of the secondary member profile, (ii) one or more predictively frequent transactional frequency metrics for the secondary member profile that are based at least in part on the one or more transactional patterns of the secondary member profile, and (iii) one or more predictively critical demographic features for the secondary member profile that are based at least in part on the one or more demographic features of the secondary member profile.
5. The computer-implemented method of claim 1 further comprising determining one or more cross-member features based at least in part on: (i) one or more predictively critical transactional patterns for the primary member profile that are based at least in part on one or more transactional patterns of the primary member profile, (ii) one or more transactional frequency metrics for the primary member profile that are based at least in part on the one or more transactional patterns of the primary member profile, and (iii) one or more predictively critical demographic features for the primary member profile that are based at least in part on the one or more demographic features of the primary member profile.
6. The computer-implemented method of claim 1 further comprising determining one or more member-specific features for the primary member profile by:
determining one or more dominant transactional patterns for the primary member profile based at least in part on one or more transactional records for the primary member profile; and
determining a transactional count frequency for the primary member profile based at least in part on the one or more transactional records.
7. The computer-implemented method of claim 6 , wherein determining the one or more dominant transactional patterns for the primary member profile comprises:
determining one or more dominant service providers associated with the primary member profile; and
determining one or more dominant prescription drugs associated with the primary member profile.
8. The computer-implemented method of claim 6 , wherein:
determining the one or more dominant transactional patterns for the primary member profile comprises determining one or more transactional patterns based at least in part on a first temporal subset of the one or more transactional records for the primary member profile, and
determining the transactional count frequency for the primary member profile comprises determining the transactional count frequency based at least in part on a second temporal subset of the one or more transactional records for the primary member profile, the first temporal subset is associated with a first temporal range, the second temporal subset is associated with a second temporal range, and the first temporal range is more recent than the second temporal range.
9. The computer-implemented method of claim 8 , wherein the first temporal subset comprises a most recent two years.
10. The computer-implemented method of claim 8 , wherein the first temporal subset comprises a most recent five years.
11. A system comprising one or more processors and at least one memory storing processor-executable instructions that, when executed by any of the one or more processors, causes the one or more processors to perform operations comprising:
accessing a cross-member enrollment prediction model from a cross-member enrollment recommendation unit, wherein the cross-member enrollment prediction model is generated based at least in part on a portion of second enrollment modeling data that comprises one or more transactional records or demographic information for one or more of a plurality of secondary member profiles that are related to a primary member profile based at least in part on first enrollment modeling data for the primary member profile and the second enrollment modeling data for the plurality of secondary member profiles with respect to (i) a transactional frequency category, (ii) one or more dominant transactional patterns, and (iii) one or more demographic features;
determining, via the cross-member enrollment recommendation unit, a cross-member enrollment recommendation with respect to the primary member profile based at least in part on the cross-member enrollment prediction model and one or more cross-member predictive features of the first enrollment modeling data;
determining, via a member-specific enrollment recommendation unit, a member-specific enrollment recommendation with respect to the primary member profile based at least in part on one or more member-specific features that are determined based at least in part on the first enrollment modeling data and one or more enrollment coverage criteria for a plurality of enrollment plans;
and
providing, via a recommendation generation unit to a remote client device, the cross-member enrollment recommendation and the member-specific enrollment recommendation to an enrollment recommendation user interface at the remote client device.
12. The system of claim 11 , wherein the second enrollment modeling data comprises one or more transactional records for the plurality of secondary member profiles and the one or more demographic features.
13. The system of claim 11 , wherein the operations further comprise:
determining, for a secondary member profile of the plurality of secondary member profiles, a transactional frequency category based at least in part on one or more transactional records for the secondary member profile;
determining, for the secondary member profile, one or more dominant transactional patterns based at least in part on the one or more transactional records for the secondary member profile; and
determining whether the secondary member profile is related to the primary member profile based at least in part on: (i) whether one or more demographic features for the primary member profile match one or more demographic features for the secondary member profile, (ii) whether a transactional frequency category for the primary member profile matches a transactional frequency category for the secondary member profile, and (iii) whether one or more dominant transactional patterns for the primary member profile match one or more dominant transactional patterns for the secondary member profile.
14. The system of claim 11 , wherein the operations further comprise determining one or more training features based at least in part on: (i) one or more predictively critical transactional patterns for a secondary member profile that are based at least in part on one or more transactional patterns of the secondary member profile, (ii) one or more predictively frequent transactional frequency metrics for the secondary member profile that are based at least in part on the one or more transactional patterns of the secondary member profile, and (iii) one or more predictively critical demographic features for the secondary member profile that are based at least in part on the one or more demographic features of the secondary member profile.
15. The system of claim 11 , wherein the operations further comprise determining one or more cross-member features based at least in part on: (i) one or more predictively critical transactional patterns for the primary member profile that are based at least in part on one or more transactional patterns of the primary member profile, (ii) one or more transactional frequency metrics for the primary member profile that are based at least in part on the one or more transactional patterns of the primary member profile, and (iii) one or more predictively critical demographic features for the primary member profile that are based at least in part on the one or more demographic features of the primary member profile.
16. The system of claim 11 , wherein the operations further comprise:
determining one or more dominant transactional patterns for the primary member profile based at least in part on one or more transactional records for the primary member profile; and
determining a transactional count frequency for the primary member profile based at least in part on the one or more transactional records.
17. The system of claim 16 , wherein the operations further comprise:
determining one or more dominant service providers associated with the primary member profile; and
determining one or more dominant prescription drugs associated with the primary member profile.
18. One or more non-transitory computer-readable storage media including instructions that, when executed by one or more processors, cause the one or more processors to perform operations comprising:
accessing a cross-member enrollment prediction model from a cross-member enrollment recommendation unit, wherein the cross-member enrollment prediction model is generated based at least in part on a portion of second enrollment modeling data that comprises one or more transactional records or demographic information for one or more of a plurality of secondary member profiles that are related to a primary member profile based at least in part on first enrollment modeling data for the primary member profile and the second enrollment modeling data for the plurality of secondary member profiles with respect to (i) a transactional frequency category, (ii) one or more dominant transactional patterns, and (iii) one or more demographic features;
determining, via the cross-member enrollment recommendation unit, a cross-member enrollment recommendation with respect to the primary member profile based at least in part on the cross-member enrollment prediction model and one or more cross-member predictive features of the first enrollment modeling data;
determining, via a member-specific enrollment recommendation unit, a member-specific enrollment recommendation with respect to the primary member profile based at least in part on one or more member-specific features that are determined based at least in part on the first enrollment modeling data and one or more enrollment coverage criteria for a plurality of enrollment plans;
and
providing, via a recommendation generation unit to a remote client device, the cross-member enrollment recommendation and the member-specific enrollment recommendation to an enrollment recommendation user interface at the remote client device.
19. The one or more non-transitory computer-readable storage media of claim 18 , wherein the second enrollment modeling data comprises one or more transactional records for the plurality of secondary member profiles and the one or more demographic features.
20. The one or more non-transitory computer-readable storage media of claim 18 , wherein the operations further comprise:
determining, for a secondary member profile of the plurality of secondary member profiles, a transactional frequency category based at least in part on one or more transactional records for the secondary member profile;
determining, for the secondary member profile, one or more dominant transactional patterns based at least in part on the one or more transactional records for the secondary member profile; and
determining whether the secondary member profile is related to the primary member profile based at least in part on: (i) whether one or more demographic features for the primary member profile match one or more demographic features for the secondary member profile, (ii) whether a transactional frequency category for the primary member profile matches a transactional frequency category for the secondary member profile, and (iii) whether one or more dominant transactional patterns for the primary member profile match one or more dominant transactional patterns for the secondary member profile.