Machine learning, causal inference, and probabilistic combinatorial techniques for forecasting and ranking prediction-based actions
Various embodiments of the present disclosure provide computer forecasting techniques for initiating presentation of an interactive user interface. The techniques may include receiving one or more candidate prediction-based actions and generating a plurality of causal risk-based impact scores with respect to a candidate prediction-based action. The techniques include generating a plurality of causal quality-based impact scores and an action sequence for a plurality of evaluation entities and generating a causal net impact score based on (i) an aggregation of the plurality of causal risk-based impact scores and the plurality of causal quality-based impact scores and (ii) a sequence impact metric corresponding to the action sequence. The techniques include generating a sequence ranking for the action sequence and initiating a presentation of an interactive user interface reflective of the action sequence and the sequence ranking.
1 . A computer-implemented method comprising:
displaying, by one or more processors, an interactive user interface comprising: (i) an action sequence, (ii) one or more sequence attributes for the action sequence, the one or more sequence attributes including an action status, and (iii) a sequence ranking for the action sequence, wherein the action sequence and the sequence ranking are generated by:
receiving one or more candidate prediction-based actions for a plurality of evaluation entities within an entity group,
generating, for the plurality of evaluation entities:
a plurality of causal risk-based impact scores associated with a candidate prediction-based action of the one or more candidate prediction-based actions,
a plurality of causal quality-based impact scores, and
the action sequence with respect to the candidate prediction-based action,
generating, for the action sequence, a causal net impact score based on (i) aggregating the plurality of causal risk-based impact scores and the plurality of causal quality-based impact scores and (ii) a sequence impact metric corresponding to the action sequence, and
generating, the sequence ranking based on the causal net impact score and one or more causal net impact scores associated with one or more other action sequences for the plurality of evaluation entities;
responsive to detecting an update to the action status associated with the action sequence, generating, by the one or more processors, an updated interactive user interface comprising: (i) an updated action sequence, (ii) one or more updated sequence attributes for the updated action sequence, the one or more updated sequence attributes including an updated action status, (iii) an updated sequence ranking associated with the updated action sequence, and (iv) a sequence of selectable icons corresponding to the updated action sequence and the one or more other action sequences; and
providing, by the one or more processors and via the updated interactive user interface, the sequence of selectable icons to a user.
2 . The computer-implemented method of claim 1 , wherein generating the causal net impact score comprises:
identifying one or more entity-specific impact metrics for one or more of the plurality of evaluation entities based on the candidate prediction-based action and the action sequence; and
generating the sequence impact metric based on an aggregation of the one or more entity-specific impact metrics.
3 . The computer-implemented method of claim 1 , wherein the action sequence is generated using a Bayesian optimization or a stochastic gradient descent sampling technique.
4 . The computer-implemented method of claim 1 , wherein the interactive user interface indicates the causal net impact score, the one or more other action sequences, and one or more sequence rankings and the one or more causal net impact scores for each of the one or more other action sequences.
5 . The computer-implemented method of claim 1 , wherein the interactive user interface comprises a plurality of selectable evaluation entities corresponding to the plurality of evaluation entities of the entity group.
6 . The computer-implemented method of claim 5 , wherein the action sequence comprises a plurality of action statuses, each action status of the plurality of action statuses indicating whether the candidate prediction-based action will be performed for a corresponding evaluation entity of the plurality of evaluation entities.
7 . The computer-implemented method of claim 5 , wherein a selection of a selectable evaluation entity of the plurality of selectable evaluation entities causes the update to the action status for an evaluation entity corresponding to the selectable evaluation entity.
8 . The computer-implemented method of claim 7 , wherein, in response to the selection of the selectable evaluation entity, the computer-implemented method comprises:
generating the updated action sequence based on the update to the action status for the evaluation entity corresponding to the selectable evaluation entity;
generating the updated sequence ranking associated with the updated action sequence; and
displaying the updated interactive user interface.
9 . The computer-implemented method of claim 6 , wherein, in response to a selection of a selectable evaluation entity of the plurality of selectable evaluation entities, the computer-implemented method comprises:
providing an enrollment request for the candidate prediction-based action to an evaluation entity corresponding to the selectable evaluation entity.
10 . The computer-implemented method of claim 9 , wherein the action status of the plurality of action statuses and for the evaluation entity is updated in response to detecting a denial response from the evaluation entity corresponding to the selectable evaluation entity.
11 . The computer-implemented method of claim 9 , further comprising:
receiving an enrollment response from the evaluation entity in response to the enrollment request;
generating the updated action sequence based on the enrollment response;
generating the updated sequence ranking for the updated action sequence; and
displaying the updated interactive user interface.
12 . A system comprising:
one or more processors; and
one or more memories storing processor-executable instructions that, when executed, cause the one or more processors to perform operations comprising:
displaying an interactive user interface comprising: (i) an action sequence, (ii) one or more sequence attributes for the action sequence, the one or more sequence attributes including an action status, and (iii) a sequence ranking for the action sequence, wherein the action sequence and the sequence ranking are generated by:
receiving one or more candidate prediction-based actions for a plurality of evaluation entities within an entity group,
generating, for the plurality of evaluation entities:
a plurality of causal risk-based impact scores associated with a candidate prediction-based action of the one or more candidate prediction-based actions,
a plurality of causal quality-based impact scores, and
the action sequence with respect to the candidate prediction-based action,
generating, for the action sequence, a causal net impact score based on (i) aggregating the plurality of causal risk-based impact scores and the plurality of causal quality-based impact scores and (ii) a sequence impact metric corresponding to the action sequence, and
generating the sequence ranking based on the causal net impact score and one or more causal net impact scores associated with one or more other action sequences for the plurality of evaluation entities;
responsive to detecting an update to the action status associated with the action sequence, generating an updated interactive user interface comprising: (i) an updated action sequence, (ii) one or more updated sequence attributes for the updated action sequence, the one or more updated sequence attributes including an updated action status, (iii) an updated sequence ranking associated with the updated action sequence, and (iv) a sequence of selectable icons corresponding to the updated action sequence and the one or more other action sequences; and
providing, via the updated interactive user interface, the sequence of selectable icons to a user.
13 . The system of claim 12 , wherein generating the causal net impact score comprises:
identifying one or more entity-specific impact metrics for one or more of the plurality of evaluation entities based on the candidate prediction-based action and the action sequence; and
generating the sequence impact metric based on an aggregation of the one or more entity-specific impact metrics.
14 . The system of claim 12 , wherein the action sequence is generated using a Bayesian optimization or a stochastic gradient descent sampling technique.
15 . The system of claim 12 , wherein the interactive user interface indicates the causal net impact score, the one or more other action sequences, and one or more sequence rankings and the one or more causal net impact scores for each of the one or more other action sequences.
16 . The system of claim 12 , wherein the interactive user interface comprises a plurality of selectable evaluation entities corresponding to the plurality of evaluation entities of the entity group.
17 . The system of claim 16 , wherein the action sequence comprises a plurality of action statuses, each action status of the plurality of action statuses indicating whether the candidate prediction-based action will be performed for a corresponding evaluation entity of the plurality of evaluation entities.
18 . The system of claim 16 , wherein a selection of a selectable evaluation entity of the plurality of selectable evaluation entities causes the update to the action status for an evaluation entity corresponding to the selectable evaluation entity.
19 . One or more non-transitory computer-readable media storing processor-executable instructions that, when executed by one or more processors, cause the one or more processors to perform operations comprising:
displaying an interactive user interface comprising: (i) an action sequence, (ii) one or more sequence attributes for the action sequence, the one or more sequence attributes including an action status, and (iii) a sequence ranking for the action sequence, wherein the action sequence and the sequence ranking are generated by:
receiving one or more candidate prediction-based actions for a plurality of evaluation entities within an entity group,
generating, for the plurality of evaluation entities,
a plurality of causal risk-based impact scores associated with a candidate prediction-based action of the one or more candidate prediction-based actions,
a plurality of causal quality-based impact scores, and
the action sequence with respect to the candidate prediction-based action,
generating, for the action sequence, a causal net impact score based on (i) aggregating the plurality of causal risk-based impact scores and the plurality of causal quality-based impact scores and (ii) a sequence impact metric corresponding to the action sequence, and
generating the sequence ranking based on the causal net impact score and one or more causal net impact scores associated with one or more other action sequences for the plurality of evaluation entities;
responsive to detecting an update to the action status associated with the action sequence, generating an updated interactive user interface comprising: (i) an updated action sequence, (ii) one or more updated sequence attributes for the updated action sequence, the one or more updated sequence attributes including an updated action status, (iii) an updated sequence ranking associated with the updated action sequence, and (iv) a sequence of selectable icons corresponding to the updated action sequence and the one or more other action sequences; and
providing, via the updated interactive user interface, the sequence of selectable icons to a user.
20 . The one or more non-transitory computer-readable media storing the processor-excutable instructions of claim 19 , wherein the interactive user interface comprises a plurality of selectable evaluation entities corresponding the plurality of evaluation entities of the entity group.