IP Library Granted Patent US 11,711,323
Granted Patent B2
US 11,711,323 · App. 16/689,514 · Granted Jul 25, 2023

Systems and methods for managing bot-generated interactions

Inventors: Mansu Kim (San Jose, CA); Benjamin Chokchai Markines (Sunnyvale, CA); Sudheer Babu Chittireddy (Cupertino, CA)
Assignee: Medallia, Inc.
H04L51/02
View Patent ↗
Loading inventors, assignments & file history…
Monitor This Case
Get email alerts when status or documents change.
Order Certified Copies
Most orders are placed with the USPTO same day — all within 24 business hours.
Order via The Patent Place →
Pre-filled with this patent's details
Quick Facts
Patent No.
US 11,711,323
App. No.
16/689,514
Granted
Jul 25, 2023
Kind
B2
Abstract

Embodiments discussed herein refer to systems and methods for chatbot interactions. When chatbot derived interactions are detected, the system can prevent those interactions from being further processed. This can be performed by an analysis system operative to engage in a dialog with customers. The system can manage a dialog with a first customer and evaluate the dialog to determine whether any interactions or responses are associated with a chatbot or a human. Interactions or responses determined be associated with a chatbot are dropped and not permitted to be further processed by the analysis system.

Claims (52)

1. A system, comprising:

an analysis system operative to engage in a dialog with a plurality of customers via a network, wherein the dialog comprises at least one question or a statement soliciting feedback and a user feedback response to each of the at least one question, and wherein the dialog is presented as a conversation thread, the analysis system comprising:

a dialog module operative for managing a dialog with each of the customers, wherein the dialog comprises interactive content; and

a bot management module operative to manage interactive content based on a determination of whether received interactive content is associated with a chatbot or a human, wherein the user feedback response is dropped and not permitted to be further processed by the analysis system if the interactive content determined be associated with a chatbot, and wherein the user feedback response is further processed by the analysis system if the interactive content determined to be associated with a human.

2. The system of claim 1 , wherein the bot management module comprises:

an address identification engine operative to assign an address to each customer that interfaces with the analysis system;

a blacklist registry comprising a plurality of addresses that are associated with a chatbot; and

a bot detection engine operative to determine whether the received interactive content is associated with a chatbot.

3. The system of claim 2 , wherein for first interactive content received from a first customer, the bot management module is operative to assign a first address to the first customer using the address identification engine, and wherein the bot management module is further operative to:

determine if the first address is in the blacklist registry; and

drop the first interactive content when the first address is determined in the blacklist registry.

4. The system of claim 3 , wherein the bot management module is further operative to:

use the bot detection engine to determine whether the first interactive content is associated with a chatbot when the first address is not in the blacklist registry;

enable further processing of the first interactive content when the bot detection engine determines that the first interactive content is not associated with a chatbot; and

when the bot detection engine determines that the first interactive content is associated with a chatbot:

place the first address in the blacklist registry; and

notify a user that the first address has been placed in the blacklist registry.

5. The system of claim 2 , wherein the bot detection engine is operative to analyze a response rate of the received interactive content to determine whether the received interactive content is associated with a chatbot.

6. The system of claim 5 , wherein the received interactive content is determined to be associated with a chatbot when the response rate is less than the response rate threshold.

7. The system of claim 2 , wherein the bot detection engine is operative to enumerate the received interactive content and compare the enumerated interactive content to existing data to determine whether the received interactive content is associated with a chatbot.

8. A computer-implemented method implemented in an analysis system operative to engage in a dialog with a plurality of customers via a network, wherein the dialog is presented as a conversation thread, the method comprising:

managing a dialog with a first customer, wherein the dialog comprises a user feedback response and a first interactive content provided by the first customer; and

evaluating the dialog to determine whether the first interactive content is associated with a chatbot or a human, wherein the user feedback response is dropped and not permitted to be further processed by the analysis system if the first interactive content is determined be associated with a chatbot, and wherein if the first interactive content is determined to be associated with a human, the user feedback response is further processed by the analysis system.

9. The computer-implemented method of claim 8 , wherein the analysis system comprises:

a bot management module comprising:

an address identification engine operative to assign an address to each customer that interfaces with the analysis system;

a blacklist registry comprising a plurality of addresses that are associated with a chatbot; and

a bot detection engine operative to determine whether the first interactive content is associated with a chatbot.

10. The computer-implemented system of claim 9 , further comprising:

assigning a first address to the first customer using the address identification engine;

determining if the first address is in the blacklist registry; and

dropping the first interactive content when the first address is determined in the blacklist registry.

11. The computer-implemented system of claim 10 , further comprising:

using the bot detection engine to determine whether the first interactive content is associated with a chatbot;

enabling further processing of the first interactive content when the bot detection engine determines that the first interactive content is not associated with a chatbot; and

when the bot detection engine determines that the first interactive content is associated with a chatbot:

placing the first address in the blacklist registry; and

notifying a user that the first address has been placed in the blacklist registry.

12. The computer-implemented method of claim 9 , further comprising:

analyzing a response rate of the first interactive content to determine whether the first interactive content is associated with a chatbot.

13. The computer-implemented method of claim 9 , further comprising:

using a chatbot to provide at least one question to the first customer as part of the dialog.

14. The computer-implemented method of claim 9 , further comprising:

detecting chatbot utterances and storing the chatbot utterances in a database; and

comparing the first interactive content to the chatbot utterances stored in the database to assess whether the first interactive content is associated with a chatbot.

15. The computer-implemented method of claim 9 , wherein the bot management module comprises at least one machine learning algorithm operative to evaluate received interactive content for evidence of a chatbot, the method further comprising:

receiving training data sets; and

training the at least one machine learning algorithm to evaluate received interactive content for evidence of a chatbot using the received training data sets.

16. The computer-implemented method of claim 15 , further comprising:

providing a prompt to a human user to enable the human user to specify whether the bot detection engine correctly determined whether the first interactive content is associated with a chatbot;

receiving an indication from the human user specifying whether the bot detection engine correctly determined whether the first interactive content is associated with a chatbot; and

incorporating the indication into the bot management module as feedback to the at least one machine learning algorithm.

Assignments (6)
RELEASE OF SECURITY INTEREST Recorded Apr 13, 2022
From: WELLS FARGO BANK NA
To: MEDALLION, INC
Reel/Frame 059581/0865 →
CORRECTIVE ASSIGNMENT TO CORRECT THE THE LIST OF PATENT PROPERTY NUMBER TO INCLUDE TWO PATENTS THAT WERE MISSING FROM THE ORIGINAL FILING PREVIOUSLY RECORDED AT REEL: 057968 FRAME: 0430. ASSIGNOR(S) HEREBY CONFIRMS THE ASSIGNMENT. Recorded Nov 1, 2021
From: WELLS FARGO BANK, NATIONAL ASSOCIATION, AS ADMINISTRATIVE AGENT
To: MEDALLIA, INC.
Reel/Frame 057982/0092 →
SECURITY INTEREST Recorded Oct 29, 2021
From: MEDALLIA, INC.
To: WILMINGTON TRUST, NATIONAL ASSOCIATION, AS COLLATERAL AGENT
Reel/Frame 057964/0016 →
RELEASE OF SECURITY INTEREST Recorded Oct 29, 2021
From: WELLS FARGO BANK, NATIONAL ASSOCIATION, AS ADMINISTRATIVE AGENT
To: MEDALLIA, INC.
Reel/Frame 057968/0430 →
SECURITY INTEREST Recorded Jul 28, 2021
From: MEDALLIA, INC.
To: WELLS FARGO BANK, NATIONAL ASSOCIATION
Reel/Frame 057011/0012 →
ASSIGNMENT OF ASSIGNOR'S INTEREST Recorded Nov 20, 2019
From: KIM, MANSU; MARKINES, BENJAMIN CHOKCHAI; CHITTIREDDY, SUDHEER BABU
To: MEDALLIA, INC.
Reel/Frame 051064/0887 →
Continuity (1)
Related Publication 20210152496A1 · May 20, 2021