System and method for dynamic dialog control for contact center systems
A system and method for engaging in an automated dialog with a user. A processor retrieves a preset dialog flow that includes various blocks directing the dialog with the user. The processor provides a prompt to the user based on a current block of the dialog flow, receives an action from the user in response to the prompt, and retrieves a classification/decision tree corresponding to the dialog flow. The classification tree has a plurality of nodes mapped to the blocks of the dialog flow. Each of the nodes represents a user intent. The processor computes a probability for each of the nodes based on the action from the user. A particular one of the nodes is then selected based on the computed probabilities. A target block of the dialog flow is further identified based on the selected node, and a response is output in response to the identified target block.
1. A system for engaging in an automated dialog with a user, the system comprising:
a processor; and
a memory, wherein the memory stores instructions that, when executed by the processor, cause the processor to:
retrieve a preset dialog flow, the dialog flow having a plurality of blocks directing the dialog with the user, the plurality of blocks for being represented as a dialog tree;
traverse the dialog tree;
provide a prompt to the user based on a current block of the plurality of blocks of the dialog tree;
receive an action from the user in response to the prompt;
retrieve a classification tree corresponding to the dialog flow, the classification tree having a plurality of nodes mapped to the plurality of blocks, the classification tree being traversed separately from the dialog tree to compute a probability for each of the nodes based on the action from the user, each of the nodes representing a user intent;
select a particular node of the plurality of nodes based on the computed probabilities;
identify a target block of the dialog flow corresponding to the particular node; and
output a response in response to the identified target block.
2. The system of claim 1 , wherein the output response corresponds to an action identified in the target block.
3. The system of claim 1 , wherein the target block is a block hierarchically below an intermediate block on a path from the current block to the target block, wherein the intermediate block is skipped during the dialog in response to identifying the target block.
4. The system of claim 1 , wherein the particular node has a highest probability of the computed probabilities.
5. The system of claim 1 , wherein the response is a prompt for disambiguating between a plurality of candidate intents.
6. The system of claim 5 , wherein the prompt changes based on the computed probability of a target node of the plurality of nodes corresponding to the target block, relative to a threshold associated with the target node.
7. The system of claim 6 , wherein the instructions further cause the threshold to be dynamically updated based on response by the user to the prompt.
8. The system of claim 6 , wherein the instructions further cause the probabilities to be updated based on response by the user to the prompt.
9. The system of claim 1 , wherein the action from the user includes a natural language utterance.
10. The system of claim 1 , wherein the probability is computed based on a current probability value and a prior probability value.
11. A system for conducting an automated dialog with a user, the system comprising:
a processor; and
a memory, wherein the memory stores instructions that, when executed by the processor, cause the processor to:
retrieve a decision tree, the decision tree directing the dialog with the user;
traverse the decision tree;
provide a prompt to the user based on a current position of the decision tree;
receive an action from the user in response to the prompt;
invoke a classification tree for computing a probability for each intent of a plurality of intents associated with the decision tree based on the action from the user, the classification tree being traversed separately from the decision tree;
select a particular intent of the plurality of intents based on the computed probabilities;
identify a response to be output to the user based on the selected particular intent; and
output the identified response for progressing the dialog to a next position of the decision tree.
12. The system of claim 11 , wherein the selected intent has a highest probability of the computed probabilities.
13. The system of claim 11 , wherein the response is a prompt for disambiguating between the plurality of intents.
14. The system of claim 13 , wherein the prompt changes based on the computed probability of the selected intent relative to a particular threshold.
15. The system of claim 14 , wherein the instructions further cause the threshold to be dynamically updated based on response by the user to the prompt.
16. The system of claim 14 , wherein the instructions further cause the probabilities to be updated based on response by the user to the prompt.
17. The system of claim 11 , wherein the action from the user includes a natural language utterance.