Utilizing machine learning models to generate interactive digital text threads with personalized digital text reply options
The present disclosure relates to systems, non-transitory computer-readable media, and methods for utilizing a machine learning model to determine predicted client intent classifications and generate personalized digital text reply options within an automated interactive digital text thread. For example, disclosed systems utilize the machine learning model to generate predicted client intent classifications and corresponding intent classification probabilities. The disclosed systems utilize the predicted client disposition classifications and the disposition classification probabilities to generate personalized digital text reply options. Moreover, the disclosed systems can provide personalized digital text reply options to a client device within an automated interactive digital text thread, bypassing the inefficiency of menu options or protocols utilized to guide clients to terminal information.
1. A computer-implemented method comprising:
extracting client features corresponding to a client device participating in an interactive digital text thread;
identifying a hierarchical intent architecture comprising a plurality of intent classifications organized in a plurality of hierarchical layers;
generating, using the hierarchical intent architecture and based on the client features, a plurality of predicted client intent classifications associated with a plurality of intent classification probabilities;
applying a first weight to a first intent classification probability of the plurality of intent classification probabilities associated with a first predicted client intent classification from a first hierarchical layer of the plurality of hierarchical layers, the first weight corresponding to the first hierarchical layer;
applying a second weight to a second intent classification probability of the plurality of intent classification probabilities associated with a second predicted client intent classification from a second hierarchical layer of the plurality of hierarchical layers, the second weight corresponding to the second hierarchical layer;
selecting the first predicted client intent classification from the plurality of predicted client intent classifications based on comparing the first intent classification probability, with the first weight applied, to the plurality of intent classification probabilities; and
providing, for display via the client device, at least two personalized digital text reply options corresponding to the first predicted client intent classification and at least one additional predicted client intent classification via the interactive digital text thread.
2. The computer-implemented method of claim 1 , further comprising applying a third weight to a third intent classification probability of the plurality of intent classification probabilities associated with a third predicted client intent classification.
3. The computer-implemented method of claim 1 , wherein identifying the hierarchical intent architecture comprises:
identifying a first layer of the hierarchical intent architecture comprising a first parent intent classification and a second parent intent classification; and
identifying a second layer of the hierarchical intent architecture comprising a first child intent classification of the first parent intent classification and a second child intent classification of the second parent intent classification.
4. The computer-implemented method of claim 3 , wherein applying the first weight to the first intent classification probability of the plurality of intent classification probabilities comprises applying the first weight to the first parent intent classification.
5. The computer-implemented method of claim 4 , wherein applying the second weight to the second intent classification probability of the plurality of intent classification probabilities associated with the second predicted client intent classification comprises applying the second weight to the first child intent classification of the first parent intent classification, wherein the second weight is greater than the first weight.
6. The computer-implemented method of claim 4 , wherein applying the second weight to the second intent classification probability of the plurality of intent classification probabilities associated with the second predicted client intent classification comprises applying the second weight to the second child intent classification of the second parent intent classification, wherein the second weight is greater than the first weight.
7. The computer-implemented method of claim 1 , wherein selecting the first predicted client intent classification comprises:
generating a first selection score by applying the first weight to the first intent classification probability;
generating a second selection score by applying the second weight to the second intent classification probability; and
comparing the first selection score and the second selection score.
8. The computer-implemented method of claim 1 , wherein generating the plurality of predicted client intent classifications comprises utilizing learned parameters of a machine learning model to analyze the client features to generate the plurality of intent classification probabilities.
9. A non-transitory computer-readable medium storing instructions that, when executed by at least one processor, cause a computer system to:
extract client features corresponding to a client device participating in an interactive digital text thread;
identify a hierarchical intent architecture comprising a plurality of intent classifications organized in a plurality of hierarchical layers;
generate, using the hierarchical intent architecture and based on the client features, a plurality of predicted client intent classifications associated with a plurality of intent classification probabilities;
apply a first weight to a first intent classification probability of the plurality of intent classification probabilities associated with a first predicted client intent classification from a first hierarchical layer of the plurality of hierarchical layers, the first weight corresponding to the first hierarchical layer;
apply a second weight to a second intent classification probability of the plurality of intent classification probabilities associated with a second predicted client intent classification from a second hierarchical layer of the plurality of hierarchical layers, the second weight corresponding to the second hierarchical layer;
select the first predicted client intent classification from the plurality of predicted client intent classifications based on comparing the first intent classification probability, with the first weight applied, to the plurality of intent classification probabilities; and
provide, for display via the client device, at least two personalized digital text reply options corresponding to the first predicted client intent classification and at least one additional predicted client intent classification via the interactive digital text thread.
10. The non-transitory computer-readable medium of claim 9 , wherein the instructions, when executed by the at least one processor, further cause the computer system to apply a third weight to a third intent classification probability of the plurality of intent classification probabilities associated with a third predicted client intent classification.
11. The non-transitory computer-readable medium of claim 9 , wherein the instructions, when executed by the at least one processor, further cause the computer system to identify the hierarchical intent architecture by:
identifying a first layer of the hierarchical intent architecture comprising a first parent intent classification and a second parent intent classification; and
identifying a second layer of the hierarchical intent architecture comprising a first child intent classification of the first parent intent classification and a second child intent classification of the second parent intent classification.
12. The non-transitory computer-readable medium of claim 11 , wherein the instructions, when executed by the at least one processor, further cause the computer system to apply the first weight to the first intent classification probability of the plurality of intent classification probabilities by applying the first weight to the first parent intent classification.
13. The non-transitory computer-readable medium of claim 12 , wherein the instructions, when executed by the at least one processor, further cause the computer system to apply the second weight to the second intent classification probability of the plurality of intent classification probabilities associated with the second predicted client intent classification by applying the second weight to the first child intent classification of the first parent intent classification, wherein the second weight is greater than the first weight.
14. The non-transitory computer-readable medium of claim 9 , wherein the instructions, when executed by the at least one processor, further cause the computer system to select the first predicted client intent classification by:
generating a first selection score by applying the first weight to the first intent classification probability;
generating a second selection score by applying the second weight to the second intent classification probability; and
comparing the first selection score and the second selection score.
15. A system comprising:
at least one processor; and
at least one non-transitory computer-readable storage medium storing instructions that, when executed by the at least one processor, cause the system to:
extract client features corresponding to a client device participating in an interactive digital text thread;
identify a hierarchical intent architecture comprising a plurality of intent classifications organized in a plurality of hierarchical layers;
generate, using the hierarchical intent architecture and based on the client features, a plurality of predicted client intent classifications associated with a plurality of intent classification probabilities;
apply a first weight to a first intent classification probability of the plurality of intent classification probabilities associated with a first predicted client intent classification from a first hierarchical layer of the plurality of hierarchical layers, the first weight corresponding to the first hierarchical layer;
apply a second weight to a second intent classification probability of the plurality of intent classification probabilities associated with a second predicted client intent classification from a second hierarchical layer of the plurality of hierarchical layers, the second weight corresponding to the second hierarchical layer;
select the first predicted client intent classification from the plurality of predicted client intent classifications based on comparing the first intent classification probability, with the first weight applied, to the plurality of intent classification probabilities; and
provide, for display via the client device, at least two personalized digital text reply options corresponding to the first predicted client intent classification and at least one additional predicted client intent classification via the interactive digital text thread.
16. The system of claim 15 , further comprising instructions that, when executed by the at least one processor, cause the system to apply a third weight to a third intent classification probability of the plurality of intent classification probabilities associated with a third predicted client intent classification.
17. The system of claim 15 , further comprising instructions that, when executed by the at least one processor, cause the system to identify the hierarchical intent architecture by:
identifying a first layer of the hierarchical intent architecture comprising a first parent intent classification and a second parent intent classification; and
identifying a second layer of the hierarchical intent architecture comprising a first child intent classification of the first parent intent classification and a second child intent classification of the second parent intent classification.
18. The system of claim 17 , further comprising instructions that, when executed by the at least one processor, cause the system to apply the first weight to the first intent classification probability of the plurality of intent classification probabilities by applying the first weight to the first parent intent classification.
19. The system of claim 18 , further comprising instructions that, when executed by the at least one processor, cause the system to apply the second weight to the second intent classification probability of the plurality of intent classification probabilities associated with the second predicted client intent classification by applying the second weight to the first child intent classification of the first parent intent classification, wherein the second weight is greater than the first weight.
20. The system of claim 15 , further comprising instructions that, when executed by the at least one processor, cause the system to generate the plurality of predicted client intent classifications by utilizing learned parameters of a machine learning model to analyze the client features to generate the plurality of intent classification probabilities.