MAINTAINING CONTEXT IN TRANSACTION CONVERSATIONS
Disclosed are examples of systems, apparatus, methods and computer program products for maintaining context in a transaction conversation. A notification of a customer interaction from a customer is received within a retailer network. Based on at least one messaging channel and at least one interaction type in the notification, the system determines that an active conversation with the customer is open. A natural language processing (NLP) model is then selected based on one or more conversation markers in the active conversation. At least one customer intent is determined with respect to one or more entities, and a response to the customer interaction is generated based on the at least one customer intent and the one or more entities.
1 . A method for maintaining context in an individualized transaction conversation within a retail network, the method comprising:
receiving a notification of a customer interaction from a customer within the retailer network, the notification including at least one messaging channel and at least one interaction type;
determining, based on the at least one messaging channel and at least one interaction type, that an active conversation with the customer is open;
selecting, based on one or more conversation markers in the active conversation, a natural language processing (NLP) model from a list of NLP models;
determining at least one customer intent with respect to one or more entities; and
generating a response to the customer interaction based on the at least one customer intent and the one or more entities.
2 . The method of claim 1 , further comprising:
accessing a database of customer profiles; and
matching the customer with at least one customer profile from the database of customer profiles, the customer profile including conversation history for the customer.
3 . The method of claim 2 , wherein determining that an active conversation with the customer is open is based on the conversation history for the customer.
4 . The method of claim 1 , further comprising:
switching from the at least one messaging channel to one or more additional channels, wherein the switching involves maintaining context based on the at least one customer intent and the one or more entities.
5 . The method of claim 1 , further comprising:
sending one or more queries to the customer, the one or more queries relating to customer intents; and
receiving one or more responses to the queries, wherein the determining the at least one customer intent is based on the one or more responses to the queries.
6 . The method of claim 5 , further comprising:
selecting at least one additional NLP model from the list of NLP models based on the one or more responses to the queries.
7 . The method of claim 1 , wherein determining that an active conversation with the customer is open is based on a conversational threshold period, such that an active conversation is open if a previous conversation message was sent within the conversational threshold period. The method of claim 1 , further comprising:
determining, from the NLP model, required customer intent data that has not been obtained; and
sending one or more queries to the customer with respect to the required customer intent data.
9 . The method of claim 1 , wherein one or more steps of the method are performed using machine learning and/or predictive analysis techniques.
10 . A system for maintaining context in an individualized transaction conversation within a retail network, the system configurable to cause:
receiving a notification of a customer interaction from a customer within the retailer network, the notification including at least one messaging channel and at least one interaction type;
determining, based on the at least one messaging channel and at least one interaction type, that an active conversation with the customer is open;
selecting, based on one or more conversation markers in the active conversation, a natural language processing (NLP) model from a list of NLP models;
determining at least one customer intent with respect to one or more entities; and
generating a response to the customer interaction based on the at least one customer intent and the one or more entities.
11 . The system of claim 10 , further configurable to cause:
accessing a database of customer profiles; and
matching the customer with at least one customer profile from the database of customer profiles, the customer profile including conversation history for the customer.
12 . The system of claim 11 , further configurable to cause:
switching from the at least one messaging channel to one or more additional channels, wherein the switching involves maintaining context based on the at least one customer intent and the one or more entities.
13 . The system of claim 10 , further configurable to cause:
sending one or more queries to the customer, the one or more queries relating to customer intents; and
receiving one or more responses to the queries, wherein the determining the at least one customer intent is based on the one or more responses to the queries.
14 . The system of claim 10 , further configurable to cause:
determining, from the NLP model, required customer intent data that has not been obtained; and
sending one or more queries to the customer with respect to the required customer intent data.
15 . The system of claim 10 , wherein one or more steps of the method are performed using machine learning and/or predictive analysis techniques.
16 . A computer program product comprising computer-readable program code capable of being executed by one or more processors when retrieved from a non-transitory computer-readable medium, the program code comprising instructions configurable to cause:
receiving a notification of a customer interaction from a customer within a retailer network, the notification including at least one messaging channel and at least one interaction type;
determining, based on the at least one messaging channel and at least one interaction type, that an active conversation with the customer is open;
selecting, based on one or more conversation markers in the active conversation, a natural language processing (NLP) model from a list of NLP models;
determining at least one customer intent with respect to one or more entities; and
generating a response to the customer interaction based on the at least one customer intent and the one or more entities.
17 . The computer program product of claim 10 , the program code comprising instructions further configurable to cause:
switching from the at least one messaging channel to one or more additional channels, wherein the switching involves maintaining context based on the at least one customer intent and the one or more entities.
18 . The computer program product of claim 10 , the program code comprising instructions further configurable to cause:
sending one or more queries to the customer, the one or more queries relating to customer intents; and
receiving one or more responses to the queries, wherein the determining the at least one customer intent is based on the one or more responses to the queries.
19 . The computer program product of claim 10 , the program code comprising instructions further configurable to cause:
determining, from the NLP model, required customer intent data that has not been obtained; and
sending one or more queries to the customer with respect to the required customer intent data.
20 . The computer program product of claim 10 , wherein one or more steps of the method are performed using machine learning and/or predictive analysis techniques.