Granted Patent B2

US 12,625,916 · App. 18/830,992 · Granted May 12, 2026

Inventors: Chetan Singh Naruka (Gurugram, IN); Deepam Kumar Singh (Ghaziabad, IN); Shailendra Singh (Thane, IN)

Assignee: Bank of America Corporation

G06F16/9538G06F16/9537G06F16/954

Abstract

Claims (87)

1 . A system comprising: a database that storing a plurality of webpages, metadata associated with the webpages, and relationships between the webpages;

and a processor communicatively coupled to the database, and that:

receives a user request, from a user, to perform a data interaction, wherein the user request comprises a user query describing the data interaction;

assigns one or more interaction categories to at least a portion of the webpages based on the metadata of the respective webpages, wherein the interaction category indicates a type of operation that can be performed using a respective webpage, wherein assigning the one or more interaction categories to a particular webpage comprises:

assigning a binary value of one to each interaction category determined to be associated with the webpage;

and assigning a binary value of zero to each interaction category determined not be associated with the webpage;

determine a particular interaction category associated with the data interaction based at least in part upon the user request;

generates a search function based on the user query, wherein the search function comprises one or more secondary queries to search the database and extract from the database data related to the user query associated with the user request;

runs the search function in the database to extract the data related to the user query comprising a plurality of webpages that are assigned to the particular interaction category by identifying the webpages that are assigned a binary value of one for the particular interaction category, wherein the extracted data at least comprises a set of the webpages, metadata associated with the webpages in the set of webpages, and relationships between one or more of the webpages in the set of webpages;

runs a machine learning model to generate, based on the extracted data, a navigation map to perform the data interaction,

wherein running the machine learning model comprises:

inputting the set of webpages, the metadata associated with the webpages and the relationships between one or more of the webpages into the machine learning model;

 generating, based on the extracted data, a navigation map to perform the data interaction, wherein the navigation map comprises a sequence of webpages from the plurality of webpages that are assigned to the particular interaction category and information relating to one or more operations that are to be performed at each of the webpages as part of performing the data interaction,

obtains the navigation map comprising the selected sequence of webpages as an output of the machine learning model; and,

displays the obtained navigation map to the user.

2 . The system of claim 1 , wherein: the database further stores, for each webpage, an identification of one or more channels the webpage is compatible with, wherein the channels comprise a web application, a mobile application, a metaverse application, or a combination thereof; and the processor further:

receives, as part of the user request, an indication of a channel to be used to perform the data interaction;

and generates the search function to search the database for only those webpages that are compatible with the indicated channel to be used to perform the data interaction;

wherein the webpages in the sequence of webpages associated with the navigation map are compatible with the indicated channel.

3 . The system of claim 1 , wherein the navigation map provides a shortest path to perform the data interaction, wherein the shortest path comprises a minimum number of operations that can be performed using the sequence of webpages to perform the data interaction.

4 . The system of claim 1 , wherein: the database further comprises a historical activity log of previous data interactions performed by a plurality of users;

and the processor further:

determines, based on the historical activity log, a second sequence of webpages and corresponding operations that were performed by one or more other users to perform the same data interaction or a similar data interaction;

wherein the sequence of webpages include at least a portion of the second sequence of webpages and corresponding operations that were performed by the one or more other users to perform the same data interaction or the similar data interaction.

5 . The system of claim 1 , wherein: the navigation map comprises a stepwise flow of a plurality of steps to be followed for performing the data interaction;

each step of the plurality of steps comprises one or more of:

an identification of a particular webpage of the sequence of webpages;

one or more operations that are to be performed at the particular webpage;

instructions to perform the one or more operations; or an estimated time to complete the one or more operations.

6 . The system of claim 1 , wherein: the database further comprises a historical activity log of previous data interactions performed by a plurality of users;

and the processor is further: determines, based on the historical activity log, an average time taken to complete one or more operations at a webpage of the sequence of webpages that is part of the navigation map;

and include as part of information associated with the navigation map, the average time taken to complete the one or more operations at the webpage.

7 . The system of claim 1 , wherein: the database further comprises a historical activity log of previous data interactions performed by a plurality of users; and the processor further:

determines, based on the historical activity log, an average time taken to complete respective one or more operations at each webpage of the sequence of webpages that is part of the navigation map;

determines, based on the average time taken to complete the respective one or more operations at each webpage, an estimated total time needed to complete all operations using the sequence of webpages that are part of the navigation map;

and includes as part of information associated with the navigation map, the estimated total time needed to complete all the operations at the sequence of webpages that are part of the navigation map.

8 . The system of claim 1 , wherein the database is a graph database, wherein the graph database stores data as a network of data nodes and relationships between the data nodes.

9 . A method comprising: receiving a user request, from a user, to perform a data interaction, wherein the user request comprises a user query describing the data interaction;

assigning one or more interaction categories to at least a portion of a plurality of webpages based on metadata of the respective webpages, wherein the interaction category indicates a type of operation that can be performed using a respective webpage, wherein assigning the one or more interaction categories to a particular webpage comprises:

assigning a binary value of one to each interaction category determined to be associated with the webpage;

and assigning a binary value of zero to each interaction category determined not be associated with the webpage;

determine a particular interaction category associated with the data interaction based at least in part upon the user request;

generating a search function based on the user query, wherein the search function comprises one or more secondary queries to search a database and extract from the database data related to the user query associated with the user request, wherein the database stores the plurality of webpages, the metadata associated with the webpages, and relationships between the webpages;

running the search function in the database to extract the data related to the user query comprising a plurality of webpages that are assigned to the particular interaction category by identifying the webpages that are assigned a binary value of one for the particular interaction category, wherein the extracted data at least comprises a set of the webpages, metadata associated with the webpages in the set of webpages, and relationships between one or more of the webpages in the set of webpages;

running a machine learning model to generate, based on the extracted data, a navigation map to perform the data interaction,

wherein running the machine learning model comprises:

inputting the set of webpages, the metadata associated with the webpages and the relationships between one or more of the webpages into the machine learning model;

 generating, based on the extracted data, a navigation map to perform the data interaction, wherein the navigation map comprises a sequence of webpages from the plurality of webpages that are assigned to the particular interaction category and information relating to one or more operations that are to be performed at each of the webpages as part of performing the data interaction,

obtaining generate the navigation map comprising the selected sequence of webpages as an output of the machine learning model;

displaying the obtained navigation map to the user.

10 . The method of claim 9 , wherein: the database further stores, for each webpage, an identification of one or more channels the webpage is compatible with, wherein the channels comprise a web application, a mobile application, a metaverse application, or a combination thereof;

and further comprising: receiving, as part of the user request, an indication of a channel to be used to perform the data interaction; and generating the search function to search the database for only those webpages that are compatible with the indicated channel to be used to perform the data interaction;

wherein the webpages in the sequence of webpages associated with the navigation map are compatible with the indicated channel.

11 . The method of claim 9 , wherein the navigation map provides a shortest path to perform the data interaction, wherein the shortest path comprises a minimum number of operations that can be performed using the sequence of webpages to perform the data interaction.

12 . The method of claim 9 , wherein: the database further comprises a historical activity log of previous data interactions performed by a plurality of users;

and further comprising: determining, based on the historical activity log, a second sequence of webpages and corresponding operations that were performed by one or more other users to perform the same data interaction or a similar data interaction;

wherein the sequence of webpages include at least a portion of the second sequence of webpages and corresponding operations that were performed by the one or more other users to perform the same data interaction or the similar data interaction.

13 . The method of claim 9 , wherein: the navigation map comprises a stepwise flow of a plurality of steps to be followed for performing the data interaction; each step of the plurality of steps comprises one or more of:

an identification of a particular webpage of the sequence of webpages;

one or more operations that are to be performed at the particular webpage;

instructions to perform the one or more operations; or an estimated time to complete the one or more operations.

14 . The method of claim 9 , wherein: the database further comprises a historical activity log of previous data interactions performed by a plurality of users; and further comprising: determining, based on the historical activity log, an average time taken to complete one or more operations at a webpage of the sequence of webpages that is part of the navigation map;

and including as part of information associated with the navigation map, the average time taken to complete the one or more operations at the webpage.

15 . The method of claim 9 , wherein: the database further comprises a historical activity log of previous data interactions performed by a plurality of users;

and further comprising: determining, based on the historical activity log, an average time taken to complete respective one or more operations at each webpage of the sequence of webpages that is part of the navigation map;

determining, based on the average time taken to complete the respective one or more operations at each webpage, an estimated total time needed to complete all operations using the sequence of webpages that are part of the navigation map;

and including as part of information associated with the navigation map, the estimated total time needed to complete all the operations at the sequence of webpages that are part of the navigation map.

16 . The method of claim 9 , wherein the database is a graph database that is configured to store data as a network of data nodes and relationships between the data nodes.

17 . A non-transitory computer-readable medium storing instructions that when executed by a processor cause the processor to:

receive a user request to perform a data interaction, wherein the user request comprises a user query describing the data interaction;

assign one or more interaction categories to at least a portion of a plurality of webpages based on metadata of the respective webpages, wherein the interaction category indicates a type of operation that can be performed using a respective webpage, wherein assigning the one or more interaction categories to a particular webpage comprises:

assigning a binary value of one to each interaction category determined to be associated with the webpage;

and assigning a binary value of zero to each interaction category determined not be associated with the webpage;

determine a particular interaction category associated with the data interaction based at least in part upon the user request;

generate a search function based on the user query, wherein the search function comprises one or more secondary queries configured to search a database and extract from the database data related to the user query associated with the user request, wherein the database stores the plurality of webpages, the metadata associated with the webpages, and relationships between the webpages;

run the search function in the database to extract the data related to the user query comprising a plurality of webpages that are assigned to the particular interaction category by identifying the webpages that are assigned a binary value of one for the particular interaction category, wherein the extracted data at least comprises a set of webpages, metadata associated with the webpages in the set of the webpages, and relationships between one or more of the webpages in the set of webpages;

run a machine learning model to generate, based on the extracted data, a navigation map to perform the data interaction,

wherein running the machine learning model comprises:

inputting the set of webpages, the metadata associated with the webpages and the relationships between one or more of the webpages into the machine learning model;

 generating, based on the extracted data, a navigation map to perform the data interaction, wherein the navigation map comprises a sequence of webpages from the plurality of webpages that are assigned to the particular interaction category and information relating to one or more operations that are to be performed at each of the webpages as part of performing the data interaction,

obtain generate the navigation map comprising the selected sequence of webpages as an output of the machine learning model;

display the obtained navigation map to the user.

18 . The non-transitory computer-readable medium of claim 17 , wherein: the database further stores, for each webpage, an identification of one or more channels the webpage is compatible with, wherein the channels comprise a web application, a mobile application, a metaverse application, or a combination thereof; and the instructions further cause the processor is to: receive, as part of the user request, an indication of a channel to be used to perform the data interaction; and generate the search function to search the database for only those webpages that are compatible with the indicated channel to be used to perform the data interaction; wherein the webpages in the sequence of webpages associated with the navigation map are compatible with the indicated channel.

19 . The non-transitory computer-readable medium of claim 17 , wherein the navigation map provides a shortest path to perform the data interaction, wherein the shortest path comprises a minimum number of operations that can be performed using the sequence of webpages to perform the data interaction.

20 . The non-transitory computer-readable medium of claim 17 , wherein: the database further comprises a historical activity log of previous data interactions performed by a plurality of users;

and the instructions further cause the processor is to: determine, based on the historical activity log, a second sequence of webpages and corresponding operations that were performed by one or more other users to perform the same data interaction or a similar data interaction;

wherein the sequence of webpages include at least a portion of the second sequence of webpages and corresponding operations that were performed by the one or more other users to perform the same data interaction or the similar data interaction.

Assignments (1)

Continuity (1)

References Cited (28)