Detecting vocabulary skill level and correcting misalignment in remote interactions

Patent metadata

Field	Value
Publication number	US-11316979-B2
Application number	US-202016984724-A
Country	US
Kind code	B2
Filing date	Aug 4, 2020
Priority date	Aug 4, 2020
Publication date	Apr 26, 2022
Grant date	Apr 26, 2022

Abstract

In order to provide effective communications, the individuals engaged in the communication should have appropriately matched language proficiencies. By ensuring that a customer of a contact center is matched with an agent having, or presenting, content of a communication utilizing language proficiency appropriate for the customer, effective communications may be provided. Should an agent deviate and provide communication content having language proficiency that is misaligned with the customer, automatic corrective action may be taken to realign the language proficiency presented to the customer.

First claim

What is claimed is:

1. A system, comprising:
a network interface to a network;
a memory;
a processor coupled to the memory programmed with executable instructions; and
wherein the processor executing the executable instructions performs:
receiving a communication comprising a customer communication device utilized by a customer and an automated resource holding the communication;
accessing a customer language proficiency rating of the customer and comprising a technical sophistication rating;
accessing a pool of language proficiency ratings associated with each agent in a pool of agents;
selecting an agent from the pool of agents that has an associated language proficiency rating comprising a technical sophistication rating that best matches the customer language proficiency rating; and
connecting the communication to comprise an agent communication device associated with the selected agent.

2. The system of claim 1 , wherein the processor executing the executable instructions performs accessing the customer language proficiency rating of the customer, further comprising:
determining the customer language proficiency rating of the customer, comprising:
accessing a scoring model having scored words;
prior to the communication, accessing a prior communication having content provided by the customer;
scoring the prior communication with the scoring model to determine a proficiency score of the communication; and
making the proficiency score of the communication accessible as the customer language proficiency rating of the customer.

3. The system of claim 2 , wherein the prior communication comprises one or more of a social media posting, email, text message, voice message, voice call, transcription of a message, or transcription of a voice call.

4. The system of claim 1 , wherein the processor executing the executable instructions performs:
accessing the pool of language proficiency ratings associated with each agent in the pool of agents, comprising:
accessing a scoring model;
for each agent in the pool of agents:
prior to the communication, accessing a prior communication having content provided by ones of the agents in the pool of agents;
scoring a word within the prior communication against the scoring model;
determining an overall score; and
making the overall score accessible as an entry in the pool of language proficiency ratings for each of the agents in the pool of agents.

5. The system of claim 1 , wherein at least one of the customer language proficiency rating or ones of the pool of language proficiency ratings associated with each agent in the pool of agents, comprises one or more of a vocabulary rating, a pronunciation rating, a grammar rating, or a speed rating.

6. The system of claim 1 , wherein selecting the agent from the pool of agents that has the associated language proficiency rating that best matches the customer language proficiency rating, further comprises the processor executing the executable instruction to perform:
accessing an acceptable range from the customer language proficiency rating; and
selecting the agent from the pool of agents that has the associated language proficiency rating that is within the acceptable range.

7. The system of claim 1 , wherein the best match is determined to be present when the agent from the pool of agents has the associated language proficiency equal to the customer language proficiency rating within a previously determined range.

8. The system of claim 1 , wherein selecting the agent from the pool of agents that has the associated language proficiency rating that best matches the customer language proficiency rating, further comprises the processor executing the executable instruction to perform:
accessing a desired variation from the customer language proficiency rating; and
selecting the agent from the pool of agents that has the associated language proficiency rating having the desired variation from the customer language proficiency rating.

9. A system, comprising:
a network interface to a network;
a memory;
a processor coupled to the memory programmed with executable instructions; and
wherein the processor executing the executable instructions performs:
receiving a communication comprising customer content, from a customer utilizing a customer communication device, and agent content, from an agent utilizing an agent commination device;
accessing a customer language proficiency rating of the customer;

monitoring the agent content;
determining an agent language proficiency from the agent content;
determining whether the agent language proficiency is dissimilar from the customer language proficiency; and
upon determining that the dissimilarity between the agent language proficiency and customer language proficiency is present, causing the agent communication device to present indicia of the dissimilarity.

10. The system of claim 9 , wherein the processor executing the executable instructions performs accessing the customer language proficiency of the customer, comprising:
determining the customer language proficiency of the customer, comprising:
accessing a scoring model;
prior to the communication, accessing a prior communication having content provided by the customer; and
scoring the prior communication with the scoring model to produce a customer language proficiency score.

11. The system of claim 10 , wherein the prior communication comprises one or more of a social media posting, email, text message, voice message, voice call, transcription of a message, or transcription of a voice call.

12. The system of claim 9 , wherein the processor executing the executable instructions performs determining the agent language proficiency from the agent content, further comprising:
accessing a scoring model; and
scoring the agent content with the scoring model to produce the agent language proficiency.

13. The system of claim 9 , wherein at least one of the customer language proficiency or the agent language proficiency, comprises one or more of a vocabulary rating, a technical sophistication rating, a fluency rating, a pronunciation rating, a grammar rating, or a speed rating.

14. The system of claim 9 , wherein determining whether the agent language proficiency is dissimilar from the customer language proficiency, comprises determining whether at last one word of the words provided by the agent as a portion of the communication has an agent language proficiency rating is outside acceptable range from.

15. The system of claim 9 , wherein the determining whether the agent language proficiency is dissimilar from the customer language proficiency, comprises determining an agent language proficiency scoring of at least one word of words provided by the agent as a portion of the communication being dissimilar from a customer language proficiency scoring of at least one word from at least one prior customer communication.

16. A computer-implemented method of training a neural network for language proficiency comprising:
collecting a first set of words having a first language proficiency and a meaning;
modifying the first set of words to a set of different words having a same meaning;
creating a first training set comprising the first set of words, the different set of words, and a set of words having a different meaning;
training the neural network in a first state using the first training set;
creating a second training set for a second stage of training comprising the first training set and the different set of words incorrectly identified as being synonyms with the first set

Assignees

• Avaya Man Lp

Inventors

• Matula Valentine C
• Negi Manish
• Ray Divakar Kumar
• Chavez David

Classifications

• G06N3/09

  Supervised learning · CPC title

• G06N3/006

  based on simulated virtual individual or collective life forms, e.g. social simulations or particle swarm optimisation [PSO] · CPC title

• G06F40/30

  Semantic analysis · CPC title

• G06F40/20

  Natural language analysis (semantic analysis of natural language G06F40/30) · CPC title

• G06N3/08

  Learning methods · CPC title