System and method for determining and utilizing repeated conversations in contact center quality processes

1. A computerized-method for calculating a Repeated-Information-Request (RIR) score of an interaction in a contact center, by which a related interaction-recording is filtered for evaluation, the computerized-method comprising:

in a computerized system comprising a processor, a data storage of interaction-recordings between an agent and a customer and of metadata related to each interaction-recording, and a memory to store the data storage, said processor is configured to:

train a Natural Language Understanding (NLU) module to classify statements in the interaction-recording with a repetition intent as “request to repeat” and “not a request to repeat”,

wherein a repetition intent is a request to repeat information during the interaction-recording:

operate a Repeated-Information-Request (RIR) score calculation module, said operating of the RIR score calculation module comprising:

(i) retrieving from the data storage an interaction-recording and an interaction-recording-length thereof;

(ii) using the trained NLU module on a transcript of the retrieved interaction-recording to: a. classify statements as “request to repeat” and “not a request to repeat”; and b. count a number of statements that were classified as “request to repeat”;

(iii) calculating a RIR score according to the number of statements that were classified as “request to repeat” by the agent and the interaction-recording-length; and

(iv) storing the RIR score of the retrieved interaction-recording in the data storage, wherein the RIR score is sent to an interaction repetition filter platform by which the interaction repetition filter platform is preconfigured to distribute the interaction-recording based on the RIR score, for evaluation to one or more downstream applications, and

wherein the distributed interaction-recording for evaluation is reviewed by an evaluator for due consideration and follow-on remedial measures to improve operational efficiency of the contact center and to improve customer satisfaction,

wherein the due consideration is, selected from at least one of: (i) identifying low level of performance of agents; and (ii) inefficient processes in the contact center, and

wherein the follow-on remedial measures are selected from at least one of: (i) assigning the agent to a coaching plan based on the identified low level of performance: and (ii) optimizing processes in the contact center,

wherein the interaction repetition filter platform is distributing the interaction-recording based on the RIR score for evaluation to one or more downstream applications, by:

comparing the calculated RIR score to a precalculated Repeat Request Threshold (RRT) to determine if the calculated RIR score is below the precalculated RRT;

when it is determined that the calculated RIR score is below the precalculated RRT, distributing the interaction-recording for evaluation: and

when it is determined that the calculated RIR score is not below the precalculated RRT, the interaction-recording is not distributed for evaluation.

2. The computerized-method of claim 1 , wherein the downstream application is at least one of: (i) quality planner microservice; (ii) workforce optimization; and (iii) supervisor dashboard.

3. The computerized-method of claim 1 , wherein the RIR score is calculated by dividing the interaction-recording-length by the counted number of requests to repeat information.

4. The computerized-method of claim 1 , wherein the RRT is an average. of an aggregation of an interaction-recording-length divided by a corresponding number of repeat requests of historical interactions.

5. The computerized-method of claim 1 , wherein the operating of the RIR score calculation module is performed per media type of the interaction-recording and per agent routing skill.

6. The computerized-method of claim 5 , wherein the media type is related to a communication channel by which an interaction related to the interaction-recording has been conducted.

7. The computerized-method of claim 6 , wherein the communication channel is selected from at least one of: audio call, video call, chat and email.

8. A computerized-system for calculating a Repeated-Information-Request (RIR) score of an interaction in a contact center, by which a related interaction-recording is filtered for evaluation, the computerized-system comprising:

a processor:;

a data storage of interaction-recordings between an agent and a customer and of metadata related to each interaction-recording; and

a memory to store the data storage,

said processor is configured to:

train a Natural Language Understanding (NLU) module to classify statements in the interaction-recording with a repetition intent as “request to repeat” and “not a request to repeat”

wherein a repetition intent is a request to repeat information during the interaction-recording; and

operate, a Repeated-Information-Request (RIR) score calculation module, said RIR score calculation module is configured to:

(i) retrieve from the data storage an interaction-recording and an interaction-recording-length thereof;

(ii) use the trained NLU module on a transcript of the retrieved interaction-recording to: a. classify statements as “request to repeat” and “not a request to repeat”; and b. count a number of statements that were classified as “request to repeat”;

(iii) calculate a RIR score according to the number of statements that were classified as “request to repeat” and the interaction-recording-length; and

(iv) store the RIR score of the retrieved interaction-recording in the data. storage,

wherein the RIR score is sent to an interaction repetition filter platform by which the interaction repetition filter platform is preconfigured to distribute the interaction-recording based on the RIR score, for evaluation to one or more downstream applications, and

wherein the distributed interaction-recording for evaluation is reviewed by an evaluator for due consideration and follow-on remedial measures to improve operational efficiency of the contact center and to improve customer satisfaction,

wherein the due consideration is selected from at least one of: (i) identifying low level of performance of agents; and (ii) inefficient processes in the contact center, and

wherein the follow-on remedial measures are selected from at least one of: (i) assigning the agent to a coaching plan based on the identified low level of performance; and (ii) optimizing processes in the contact center,

wherein the interaction repetition filter platform is distributing the interaction-recording based on the RIR score for evaluation to one or more downstream applications, by:

comparing the calculated RIR score to a precalculated Repeat Request Threshold (RRT) to determine if the calculated MR score is below the RRT;

when it is determined that the calculated RIR score is below the precalculated RRT, distributing the interaction-recording for evaluation; and

when it is determined that the calculated RIR score is not below the precalculated RRT, the interaction-recording is not distributed for evaluation.

9. The computerized-method of claim 1 , wherein the RRT is calculated by using formula I:

RTT

=

(

∑

i

=

1

n

Li

Ri

)

/

n

(

I

)

whereby:

n is a total number of historical interactions,

Li is a length of an interaction-recording, and

Ri is ‘repeat request count’ of the interaction.