# US 12,182,672 B2 · Systems and methods for generating dynamic conversational queries in user interfaces using state machine libraries > Agents: this data is available as JSON from the patentgrep API and MCP server — see https://patentgrep.com/llms.txt - Number: US 12,182,672 B2 - Document ID: US12182672B2 - Status: Granted - Kind: B2 - Application number: 17/307,590 - Filed: 2021-05-04 - Published: 2022-11-10 - Granted: 2024-12-31 - Earliest claimed: 2021-05-04 - Assignees: Capital One Services, LLC - Applicants: Capital One Services, LLC - Inventors: Minh Le; William Miller; Sara Mikulic; Rui Zhang; Erik Mueller - Web page: https://patentgrep.com/patent/US12182672B2 ## CPC - G06F 40/35: Discourse or dialogue representation - G06N 20/00: Machine learning - G06F 16/335: Filtering based on additional data, e.g. user or group profiles (filtering in web context G06F16/9535, G06F16/9536) - G06F 16/338: Presentation of query results - G06F 40/30: Semantic analysis - G06N 3/09: Supervised learning - G06F: ELECTRIC DIGITAL DATA PROCESSING (computer systems based on specific computational models G06N) - G06N: COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS ## Abstract Described are methods and systems are for generating dynamic conversational queries. For example, as opposed to being a simply reactive system, the methods and systems herein provide means for actively determining a user's intent and generating a dynamic query based on the determined user intent. Moreover, these methods and systems generate these queries in a conversational environment. ## Claims ### Claim 1 A system for generating dynamic conversational queries in user interfaces using state machine libraries, the system comprising: cloud-based storage circuitry configured to store a first model for determining, based on user data, a first prediction confidence that a user requires a first resolution of a plurality of available resolutions, wherein the user data comprises a history of electronic communications of the user; cloud-based control circuitry configured to: receive the user data in response to the user interacting with a user interface; generate a feature input based on the user data; input the feature input into the first model; receive an output from the first model, wherein the output comprises the first prediction confidence; compare the first prediction confidence to a first threshold confidence; in response to the first prediction confidence equaling or exceeding the first threshold confidence, select a first state machine from a plurality of state machines corresponding to a first resolution type of the first resolution; determine, using the first state machine, a first parameter for the first resolution, wherein the first parameter comprises a required electronic communication frequency; determine, based on the user data, whether the first parameter is met based on comparing the required electronic communication frequency to the history of electronic communications; and cloud-based input/output circuitry configured to generate for display, on a user interface, a first dynamic conversational query corresponding to the first resolution in response to the first parameter being met, wherein the first dynamic conversational query comprises dialog related to the first resolution. ### Claim 2 A method for generating dynamic conversational queries in user interfaces using state machine libraries, the method comprising: receiving user data; determining, based on the user data, a first prediction confidence that a user requires a first resolution of a plurality of available resolutions; comparing the first prediction confidence to a first threshold confidence; in response to the first prediction confidence equaling or exceeding the first threshold confidence, selecting a first state machine from a plurality of state machines corresponding to a first resolution type of the first resolution; determining, using the first state machine, a first parameter for the first resolution; determining, based on the user data, whether the first parameter is met; and in response to the first parameter being met, generating for display, on a user interface, a first dynamic conversational query corresponding to the first resolution. ### Claim 3 (depends on 2) The method of claim 2, further comprising: determining, based on the user data, a second prediction confidence that the user requires a second resolution of the plurality of available resolutions; comparing the second prediction confidence to a second threshold confidence; in response to the second prediction confidence equaling or exceeding the second threshold confidence, selecting a second state machine from the plurality of state machines corresponding to a second resolution type of the second resolution; determining, using the second state machine, a second parameter for the second resolution; determining, based on the user data, whether the second parameter is met; and in response to the second parameter not being met, generating for display, on the user interface, a second dynamic conversational query corresponding to meeting the second parameter. ### Claim 4 (depends on 2) The method of claim 2, wherein determining, based on the user data, the first prediction confidence that the user requires for the first resolution of the plurality of available resolutions further comprises: generating a feature input based on the user data; inputting the feature input into a first model; and receiving an output from the first model, wherein the output comprises the first prediction confidence. ### Claim 5 (depends on 4) The method of claim 4, further comprising: determining an accuracy metric for the first model; and determining the first threshold confidence based on the accuracy metric. ### Claim 6 (depends on 4) The method of claim 4, further comprising: determining a frequency at which the user requires the first resolution; and determining the first threshold confidence based on the frequency. ### Claim 7 (depends on 2) The method of claim 2, wherein selecting the first state machine from the plurality of state machines corresponding to the first resolution type of the first resolution comprising: determining a characteristic of the first resolution type; and comparing the characteristic with listings of respective state machines that correspond to each resolution type of the plurality of available resolutions to determine that the first state machine corresponds to the first resolution type. ### Claim 8 (depends on 2) The method of claim 2, further comprising: retrieving a plurality of parameters for the first resolution; determining a ranking for the plurality of parameters; and selecting to determine the first parameter based on the ranking. ### Claim 9 (depends on 2) The method of claim 2, wherein the first parameter comprises a required location of the user, and wherein the user data comprises a current location of the user. ### Claim 10 (depends on 2) The method of claim 2, wherein the first parameter comprises available calendar dates for the first resolution, and wherein the user data comprises a current date. ### Claim 11 (depends on 2) The method of claim 2, wherein the first parameter comprises a required electronic communication frequency, and wherein the user data comprises a history of electronic communications of the user. ### Claim 12 A non-transitory, computer readable medium for generating dynamic conversational queries in user interfaces using state machine libraries comprising instructions that when executed by one or more processors, cause operations comprising: receiving user data; determining, based on the user data, a first prediction confidence that a user requires a first resolution of a plurality of available resolutions; comparing the first prediction confidence to a first threshold confidence; in response to the first prediction confidence equaling or exceeding the first threshold confidence, selecting a first state machine from a plurality of state machines corresponding to a first resolution type of the first resolution; determining, using the first state machine, a first parameter for the first resolution; determining, based on the user data, whether the first parameter is met; and in response to the first parameter being met, generating for display, on a user interface, a first dynamic conversational query corresponding to the first resolution. ### Claim 13 (depends on 12) The non-transitory, computer readable medium of claim 12, further comprising: determining, based on the user data, a second prediction confidence that the user requires a second resolution of the plurality of available resolutions; comparing the second prediction confidence to a second threshold confidence; in response to the second prediction confidence equaling or exceeding the second threshold confidence, selecting a second state machine from the plurality of state machines corresponding to a second resolution type of the second resolution; determining, using the second state machine, a second parameter for the second resolution; determining, based on the user data, whether the second parameter is met; and in response to the second parameter not being met, generating for display, on the user interface, a second dynamic conversational query corresponding to meeting the second parameter. ### Claim 14 (depends on 12) The non-transitory, computer readable medium of claim 12, wherein determining, based on the user data, the first prediction confidence that the user requires the first resolution of the plurality of available resolutions further comprises: generating a feature input based on the user data; inputting the feature input into a first model; and receiving an output from the first model, wherein the output comprises the first prediction confidence. ### Claim 15 (depends on 14) The non-transitory, computer readable medium of claim 14, further comprising: determining an accuracy metric for the first model; and determining the first threshold confidence based on the accuracy metric. ### Claim 16 (depends on 14) The non-transitory, computer readable medium of claim 14, further comprising: determining a frequency at which the user requires the first resolution; and determining the first threshold confidence based on the frequency. ### Claim 17 (depends on 12) The non-transitory, computer readable medium of claim 12, wherein selecting the first state machine from the plurality of state machines corresponding to the first resolution type of the first resolution comprising: determining a characteristic of the first resolution type; and comparing the characteristic with listings of respective state machines that correspond to each resolution type of the plurality of available resolutions to determine that the first state machine corresponds to the first resolution type. ### Claim 18 (depends on 12) The non-transitory, computer readable medium of claim 12, further comprising: retrieving a plurality of parameters for the first resolution; determining a ranking for the plurality of parameters; and selecting to determine the first parameter based on the ranking. ### Claim 19 (depends on 12) The non-transitory, computer readable medium of claim 12, wherein the first parameter comprises a required location of the user, and wherein the user data comprises a current location of the user. ### Claim 20 (depends on 12) The non-transitory, computer readable medium of claim 12, wherein the first parameter comprises available calendar dates for the first resolution, and wherein the user data comprises a current date.