US 11,455,475 B2Granted

Human-to-human conversation analysis

AssigneeVerint Americas Inc.

InventorCharles C. Wooters

Life of this patent

Estimating the term…

Loading patent…

Abstract

Customer support, and other types of activities in which there is a dialogue between two humans can generate large volumes of conversation records. Automated analysis of these records can provide information about high-level features of, for example, the workings of a customer service department. Analysis of these conversations between a customer and a customer-support agent may also allow identification of customer support activities that can be provided by virtual agents instead of actual human agents. The analysis may evaluate conversations in terms of complexity, duration, and sentiment of the participants. Additionally, the conversations may also be analyzed to identify the existence of selected concepts or keywords. Workflow characteristics, the extent to which the conversation represents a multi-step process intended to accomplish a task, may also be determined for the conversations. Characteristics of individual conversations may be combined to obtain generalized or representative features for a set of a conversation records.

Claims30

Dependent claims

A method of automatically analyzing dialogue between humans to establish a metric indicative of suitability of virtual agent to interact with a customer in place of a human agent, value of the metric based on stored human-to-human conversation records, the method comprising:

accessing a corpus of human-to-human conversations;
scoring, using one or more processing units, individual conversations in the corpus according to one or more metrics, each of the one or more metrics being indicative of at least one dialogue characteristic selected from duration, sentiment, and complexity, to produce a respective metric score for each selected dialogue characteristic for individual conversations in the corpus;
and
producing a workflow score for the individual conversations in the corpus by scoring, using one or more processors, the individual conversations in the corpus according to a likelihood that at least a portion of the one or more individual conversations includes an algorithmic series of communications between two humans to accomplish a task, wherein the workflow score is based on a number and spacing of questions between the two humans for each individual conversation;
combining the respective metric scores for each selected dialogue characteristic-for individual conversations of the corpus to generate a representative value of the one or more metrics for the corpus;
and
outputting the representative value of the one or more metrics and the workflow score, the workflow score indicative of the suitability of the virtual agent to interact with the customer in place of the human agent for the individual conversation.

4 more elements

A system for automatically analyzing dialogue between humans to establish a metric indicative of suitability of virtual agent to interact with a customer in place of a human agent, value of the metric based on stored human-to-human conversation records, the system comprising:

one or more processors;
and
memory storing computer-executable instructions that, when executed by one or more processors, cause the one or more processors to:
access a corpus of human-to-human conversations;
score, using one or more processing units, individual conversations in the corpus according to one or more metrics, each of the one or more metrics being indicative of at least one dialogue characteristic selected from duration, sentiment, and complexity, to produce a respective metric score for each selected dialogue characteristic for individual conversations in the corpus;
and
produce a workflow score for the individual conversations in the corpus by scoring, using one or more processors, the individual conversations in the corpus according to a likelihood that at least a portion of the one or more individual conversations includes an algorithmic series of communications between the two humans to accomplish a task, wherein the workflow score is based on a number and spacing of questions between the two humans for each individual conversation;
combine the respective metric scores for each selected dialogue characteristic for individual conversations of the corpus to generate a representative value of the one or more metrics for the corpus;
and
output the representative value of the one or more metrics and the workflow score the workflow score indicative of the suitability of the virtual agent to interact with the customer in place of the human agent for the individual conversation.

7 more elements

Dependencies are inferred from the claim text.

Facts

Application: 16/723,128Filed 20 Dec 2019 · continuation
Priority: 31 Aug 2012
Published: US 2020/0125805 A123 Apr 2020
Granted: 27 Sep 2022
Term ends: Not estimatedThe source records don’t support an estimate.

Classification3 codes

G06F 40/40
Handling natural language data: processing or translation of natural language
G06F 40/35
Semantic analysis: discourse or dialogue representation
G06F
Computing and data processing

Family

Application61/696,040Us provisional application · filed 31 Aug 2012
Application13/778,824Parent of this division · filed 27 Feb 2013
Application16/504,773Parent of this continuation · filed 8 Jul 2019
US 10,346,542 B2Parent of this division · filed 9 Jul 2019
US 2020/0125805 A1Pre-grant publication · published 23 Apr 2020
US 10,515,156 B2Parent of this continuation
Application16/504,773Parent of this division
US 2014/0067375 A1Published application
US 2019/0332676 A1Published application
US 11,455,475 B2This patent

From the US filing history. Foreign family members aren’t listed.