Dialogue flow optimization and personalization

Patent metadata

Field	Value
Publication number	US-11025775-B2
Application number	US-201916567513-A
Country	US
Kind code	B2
Filing date	Sep 11, 2019
Priority date	Oct 21, 2015
Publication date	Jun 1, 2021
Grant date	Jun 1, 2021

Abstract

A method for generating a dialogue tree for an automated self-help system of a contact center from a plurality of recorded interactions between customers and agents of the contact center includes: computing, by a processor, a plurality of feature vectors, each feature vector corresponding to one of the recorded interactions; computing, by the processor, similarities between pairs of the feature vectors; grouping, by the processor, similar feature vectors based on the computed similarities into groups of interactions; rating, by the processor, feature vectors within each group of interactions based on one or more criteria, wherein the criteria include at least one of interaction time, success rate, and customer satisfaction; and outputting, by the processor, a dialogue tree in accordance with the rated feature vectors for configuring the automated self-help system.

First claim

The invention claimed is:

1. A method for operating an automated self-help system of a contact center, the automated self-help system comprising at least one dialogue tree, the method comprising:
establishing, by a processor, an interaction between a customer and the automated self-help system;
identifying, by the processor, a plurality of customer features of the customer, wherein the plurality of customer features are known about the customer and associated with previously recorded interactions of the customer;
creating a plurality of partitions of the previously recorded interactions based on a plurality of groups of personalization features;
mining a dialogue tree for sequences from each of the plurality of partitions to compute a at least one dialogue tree, wherein sequences are pruned to filter successful interactions and unsuccessful interactions;
outputting the at least one dialogue tree for configuring the automated self-help system; and
controlling the automated self-help system in accordance with the customer features and the at least one dialogue tree, wherein a warning is triggered upon selection of at least one dialogue tree comprising a bad sequence.

2. The method of claim 1 , wherein the at least one dialogue tree comprises a plurality of dialogue trees, each of the dialogue trees being associated with a group of personalization features, and
wherein the controlling the automated self-help system in accordance with the customer features further comprises identifying a dialogue tree of the dialogue trees associated with the group of personalization features corresponding to the customer features.

3. The method of claim 1 , wherein the at least one dialogue tree comprises a plurality of agent nodes arranged in a plurality of agent layers and a plurality of customer nodes arranged in a plurality of customer layers,
wherein at least one of the customer nodes is coupled to a plurality of successive agent nodes,
wherein each edge of the edges between the at least one of the customer nodes and the successive agent nodes includes a condition corresponding to a group of personalization features, and
wherein the controlling the automated self-help system in accordance with the customer features further comprises identifying an edge of the edges wherein the group of personalization features associated with the edge corresponds to the customer features.

4. A system comprising:
a processor; and
a memory, wherein the memory stores instructions that, when executed by the processor, cause the processor to:
establish an interaction between a customer and an automated self-help system of a contact center, the automated self-help system comprising at least one dialogue tree;
identify a plurality of customer features of the customer, wherein the plurality of customer features are known about the customer and associated with the previously recorded interactions of the customer;
create a plurality of partitions of the previously recorded interactions based on a plurality of groups of personalization features;
mine a dialogue tree for sequences from each of the plurality of partitions to compute a at least one dialogue tree, wherein sequences are pruned to filter successful interactions and unsuccessful interactions;
output the at least one dialogue tree for configuring the automated self-help system; and
control the automated self-help system in accordance with the customer features and the at least one dialogue tree, wherein a warning is triggered upon selection of at least one dialogue tree comprising a bad sequence.

5. The system of claim 4 , wherein the at least one dialogue tree comprises a plurality of dialogue trees, each of the dialogue trees being associated with a group of personalization features, and
wherein the instructions that cause the processor to control the automated self-help system in accordance with the customer features include instructions that, when executed by the processor, cause the processor to identify a dialogue tree of the dialogue trees associated with the group of personalization features corresponding to the customer features.

6. The system of claim 4 , wherein the at least one dialogue tree comprises a plurality of agent nodes arranged in a plurality of agent layers and a plurality of customer nodes arranged in a plurality of customer layers,
wherein at least one of the customer nodes is coupled to a plurality of successive agent nodes,
wherein each edge of the edges between the at least one of the customer nodes and the successive agent nodes includes a condition corresponding to a group of personalization features, and
wherein the instructions that cause the processor to control the automated self-help system in accordance with the customer features include instructions that, when executed by the processor, cause the processor to identify an edge of the edges wherein the group of personalization features associated with the edge corresponds to the customer features.

7. The method of claim 1 , wherein the successful interaction comprises one or more of: interactions did not follow with a repeat interaction after a given period of time, sentiment analysis of positive emotion, and explicit indications from the customer.

8. The method of claim 1 , wherein the unsuccessful interaction comprises one or more of: sentiment analysis of negative emotion and an escalation request.

9. The system of claim 4 , wherein the successful interaction comprises one or more of: interactions did not follow with a repeat interaction after a given period of time, sentiment analysis of positive emotion, and explicit indications from the customer.

10. The system of claim 4 , wherein the unsuccessful interaction comprises one or more of: sentiment analysis of negative emotion and an escalation request.

Assignees

• Genesys Telecommunications Laboratories Inc

Inventors

• Tapuhi Tamir
• Konig Yochai
• Lev-Tov Amir
• Faizakof Avraham
• Lev Yoni

Classifications

• H04M7/0012

  Details of application programming interfaces [API] for telephone networks; Arrangements which combine a telephonic communication equipment and a computer, i.e. computer telephony integration [CPI] arrangements · CPC title

• G10L15/22

  Procedures used during a speech recognition process, e.g. man-machine dialogue · CPC title

• H04M3/4938 Primary

  comprising a voice browser which renders and interprets, e.g. VoiceXML · CPC title

• G06F16/3329

  Natural language query formulation · CPC title

• H04M3/5166

  in combination with interactive voice response systems or voice portals, e.g. as front-ends · CPC title