Automatic rule generation for next-action recommendation engine

Patent metadata

Field	Value
Publication number	US-2022198529-A1
Application number	US-202117563874-A
Country	US
Kind code	A1
Filing date	Dec 28, 2021
Priority date	Jul 24, 2019
Publication date	Jun 23, 2022
Grant date	—

Abstract

A system can recommend a next action for a user. A memory can store user data corresponding to the user and can include historic interaction points. A behavior pattern can be identified based on two or more interaction points stored in the user data. An intent of the user based on the behavior pattern can be identified. The intent can be based on a previous behavior pattern of another user. Several probabilities that the user will meet one or more objectives can be determined based on the intent. The probabilities can be scored using and used to assign a policy to the first user. A next action can be recommended based on the policy and executed with respect to the user. The outcome of the recommended next action can be stored to the user data.

First claim

1 - 20 . (canceled)

21 . A computer-implemented method using a contextual multi-armed bandit model, the method comprising:
receiving customer data corresponding to a first customer of a plurality of customer, the customer data stored in a memory storage device and including a plurality of historic customer interaction points; identifying a first customer behavior based on two or more customer interaction points of the plurality of historic customer interaction points; identifying a goal of the first customer based on the first customer behavior, the goal based on a second customer behavior of a second customer of the plurality of customer; determining, through use of a multi-armed bandit model, a plurality of customer propensities that the first customer will meet each of a plurality of objectives based on the goal; assigning a policy from a plurality of policies to the first customer based on scoring each of the plurality of objectives from the plurality of customer propensities, the policy based on a mapping between the customer data and one or more actions of a plurality of actions associated with the policy; outputting to the first customer a recommended next action from the plurality of actions associated with the assigned policy; receiving a new customer interaction point from the first customer and responsive to the recommended next action; and iterating the method using the contextual multi-armed bandit model to recommend a new recommended next action as customer data is updated.

22 . The method of claim 21 , wherein the interaction points include customer interactions and customer non-interactions of the first customer.

23 . The method of claim 21 , wherein outputting the recommended next action further comprises prompting the first customer to provide additional information.

24 . The method of claim 21 , wherein outputting the recommended next action further comprises prompting the first customer to complete a transaction.

25 . The method of claim 21 , wherein the assigned policy corresponds to a first objective of the plurality of objectives having a greater probability that the first customer will meet the first objective than a second objective of the plurality of objectives.

26 . The method of claim 21 , wherein the assigned policy is a policy of a second type and is assigned based on determining that a policy of a first type could not be identified.

27 . The method of claim 21 , wherein the outputting the recommended next action further comprises determining that the recommended next action is more suitable for the first customer than another action of the plurality of actions.

28 . The method of claim 21 , wherein a first objective of the plurality of objectives comprises two or more stages.

29 . The method of claim 28 , further comprising rewarding the first customer in response to advancing to a subsequent stage from a prior stage, the subsequent stage being progressively closer to fulfilling the objective than the prior stage.

30 . A system comprising:
a memory storage device configured to store customer data corresponding to a first customer of a plurality of customers and a plurality of historic customer interaction points; one or more processors implementing a contextual multi-armed bandit model configured to:
identify, using a first model, a first customer behavior based on two or more customer interaction points of the plurality of historic customer interaction points;
identify, using the first model, a goal of the first customer based on the first customer behavior, the goal based on a second customer behavior of a second customer of the plurality of customers;
determine, through use of a multi-armed bandit model and using the first model, a plurality of customer propensities that the first customer will meet each of a plurality of objectives based on the goal;
assign, using the first model, a policy from a plurality of policies to the first customer based on scoring each of the plurality of objectives from the plurality of customer propensities, the policy based on a mapping between the customer data and one or more actions of a plurality of actions associated with the policy;
output to the first customer, using a second model, a recommended next action from the plurality of actions associated with the assigned policy;
receive a new customer interaction point from the first customer responsive to the recommended next action; and
iterating the method using the contextual multi-armed bandit model to recommend a new recommended next action as customer data is updated.

31 . The system of claim 30 , wherein the customer interaction points include customer interactions and customer non-interactions of the first customer.

32 . The system of claim 30 , wherein the outputting of the recommended next action prompts the customer to provide additional information.

33 . The system of claim 30 , wherein the outputting of the recommended next action prompts the customer to complete a transaction.

34 . The system of claim 30 , wherein the assigned policy corresponds to a first objective of the plurality of objectives having a greater probability that the customer will meet the first objective than a second objective of the plurality of objectives.

35 . The system of claim 30 , wherein the assigned policy is a policy of a second type and is assigned based on determining that a policy of a first type could not be identified.

36 . The system of claim 30 , wherein the one or more processors are further configured to determine that the recommended next action is more suitable for the customer than another action of the plurality of the actions.

37 . The system of claim 31 , wherein a first objective of the plurality of objectives comprises two or more stages.

38 . The system of claim 37 , wherein the one or more processors are further configured to reward the customer in response to advancing to a subsequent stage from a prior stage, the subsequent stage being progressively closer to fulfilling the objective than the prior stage.

Assignees

• Salesforce Com Inc

Inventors

• Zhang Yuxi
• Xie Kexin
• Mallick Shrestha Basu
• Grissen Darrell

Classifications

• G06Q30/0281 Primary

  Customer communication at a business location, e.g. providing product or service information, consulting · CPC title

• G06Q30/0201

  Market modelling; Market analysis; Collecting market data · CPC title

• G06Q30/0271

  Personalized advertisement · CPC title