Auto identification and mapping of functional attributes from visual representation

• US 10,810,273 B2
• Filed: 06/13/2017
• Issued: 10/20/2020
• Est. Priority Date: 06/13/2017
• Status: Active Grant

• Current Assignee: Bank of America Corp.
• Original Assignee: Bank of America Corp.
• Inventors: Chakraborty, Pinak, Bansal, Gaurav, Mahajan, Nipun, Raghuvanshi, Yogesh
• Primary Examiner(s): Syed, Farhan M
• Application Number:
  US15/621,349
• Publication Number:
  US 20180357324A1
• Time in Patent Office: 1,225 Days
• CPC Class Codes: G06F 16/9535 Search customisation based on user profiles and personalisation
• CPC Class Codes: G06F 16/955 using information identifiers, e.g. uniform resource locators [URL]
• CPC Class Codes: G06F 3/04895 Guidance during keyboard input operation, e.g. prompting
• CPC Class Codes: G06F 40/143 Markup, e.g. Standard Generalized Markup Language [SGML] or Document Type Definition [DTD]
• CPC Class Codes: G06F 40/289 Phrasal analysis, e.g. finite state techniques or chunking
• CPC Class Codes: G06F 40/30 Semantic analysis

Abstract

Aspects of the disclosure relate to computing systems for receiving user requests through channels of a webpage and performing computational functions corresponding to the user requests. In one embodiment, a computing platform may detect a change corresponding to input parameters for performing a user request through a first representation of a webpage. The computing platform may update, based on the detected change corresponding to the input parameters, a natural language processing application associated with a second representation of the webpage. The computing platform may generate, based on the updated natural language processing application associated with the second representation of the webpage, a logical table including one or more questions to ask a user through the second representation of the webpage in response to one or more user inputs.

Claims

1. A computing platform, comprising:

at least one processor;

memory storing computer-readable instructions that, when executed by the at least one processor, cause the computing platform to;

detect a change corresponding to input parameters for a first type of input for performing a user request through a first representation of a webpage, detecting the change corresponding to input parameters including identifying metadata for the input parameters;

parse a page document corresponding to the first representation of the webpage to generate a document object model (DOM) of the page document including one or more tag sequences;

isolate each of the one or more tag sequences related to user input fields for performing the user request;

extract, from each of the one or more isolated tag sequences, information corresponding to the input parameters and associated attributes of each of the user input fields; and

based on the extracted information, assign each of the input parameters as either mandatory or optional;

update, based on the detected change corresponding to the input parameters and assessment of the metadata for the input parameters, a natural language processing application associated with a second representation of the webpage, updating the natural language processing application including updating input parameters associated with a second type of input different from the first type of input;

generate, based on the updated natural language processing application associated with the second representation of the webpage, a logical table including one or more questions for prompting a user to provide one or more answers for the input parameters through the second representation of the webpage;

receive, through the second representation of the webpage, a first user input corresponding to the user request;

responsive to receiving the first user input, prompt the user with the one or more questions regarding the input parameters for the user request;

responsive to prompting the user with the one or more questions, receive at least a second user input including the one or more answers for the input parameters; and

in response to receiving at least the second user input including the one or more answers, execute the user request with the one or more answers for the input parameters.

2. The computing platform of claim 1, wherein the computer-readable instructions, when executed by the at least one processor, further cause the computing platform to:

retrieve historical conversation logs from one or more databases stored on the computing platform;

isolate the historical conversation logs corresponding to the user request;

parse the isolated historical conversation logs corresponding to the user request to produce a part-of-speech (POS) tagged sentence for each of the historical conversation logs;

generate a mapping of the input parameters and associated attributes with the POS tagged sentence for each of the historical conversation logs;

build, based on the mapping, a confidence association of the input parameters and associated attributes with the POS tagged sentence for each of the historical conversation logs; and

update, based on the mapping and the confidence association, the natural language processing application associated with the second representation of the webpage.

3. The computing platform of claim 2, wherein the computer-readable instructions, when executed by the at least one processor, further cause the computing platform to:

identify, based on the mapping of the input parameters and associated attributes with the POS tagged sentence for each of the historical conversation logs, a particular question of the one or more questions for each of the input parameters corresponding to the user request;

determine a priority ranking for each of the input parameters corresponding to the user request based on an average position of the input parameters in the POS tagged sentence in each of the historical conversation logs; and

order the one or more questions based on the priority ranking for each of the input parameters.

4. The computing platform of claim 3, wherein the computer-readable instructions, when executed by the at least one processor, further cause the computing platform to:

parse the first user input to produce a POS tagged sentence for the first user input; and

identify, based on the POS tagged sentence for the first user input, input parameters for which the user has provided information.

5. The computing platform of claim 4, wherein the computer-readable instructions, when executed by the at least one processor, further cause the computing platform to:

determine the input parameters for which the user has not provided information

identify the priority ranking for each of the input parameters for which the user has not provided information;

based on the priority ranking for each of the input parameters, prompt the user to provide the second user input by presenting a question corresponding to the input parameter having a highest priority ranking; and

receive, through the second representation of the webpage, the second user input.

6. The computing platform of claim 5, wherein the computer-readable instructions, when executed by the at least one processor, further cause the computing platform to:

identify that the user has provided information for each of the input parameters corresponding to the user request; and

in response to identifying that the user has provided information for each of the input parameters corresponding to the user request, perform the user request.

7. A method, comprising:

at a computing platform comprising at least one processor, and memory;

detecting a change corresponding to input parameters for a first type of input for performing a user request through a first representation of a webpage, detecting the change corresponding to input parameters including identifying metadata for the input parameters;

parsing a page document corresponding to the first representation of the webpage to generate a document object model (DOM) of the page document including one or more tag sequences;

isolating each of the one or more tag sequences related to user input fields for performing the user request;

extracting, from each of the one or more isolated tag sequences, information corresponding to the input parameters and associated attributes of each of the user input fields; and

based on the extracted information, assigning each of the input parameters as either mandatory or optional;

updating, based on the detected change corresponding to the input parameters and assessment of the metadata for the input parameters, a natural language processing application associated with a second representation of the webpage, updating the natural language processing application including updating input parameters associated with a second type of input different from the first type of input;

generating, based on the updated natural language processing application associated with the second representation of the webpage, a logical table including one or more questions for prompting a user to provide one or more answers for the input parameters through the second representation of the webpage;

receiving, through the second representation of the webpage, a first user input corresponding to the user request;

responsive to receiving the first user input, prompting the user with the one or more questions regarding the input parameters for the user request;

responsive to prompting the user with the one or more questions, receiving at least a second user input including the one or more answers for the input parameters; and

in response to receiving at least the second user input including the one or more answers, executing the user request with the one or more answers for the input parameters.

8. The method of claim 7, further comprising:

retrieve historical conversation logs from one or more databases stored on the computing platform;

isolating the historical conversation logs corresponding to the user request;

parsing the isolated historical conversation logs corresponding to the user request to produce a part-of-speech (POS) tagged sentence for each of the historical conversation logs;

generating a mapping of the input parameters and associated attributes with the POS tagged sentence for each of the historical conversation logs;

building, based on the mapping, a confidence association of the input parameters and associated attributes with the POS tagged sentence for each of the historical conversation logs; and

updating, based on the mapping and the confidence association, the natural language processing application associated with the second representation of the webpage.

9. The method of claim 8, further comprising:

identifying, based on the mapping of the input parameters and associated attributes with the POS tagged sentence for each of the historical conversation logs, a particular question of the one or more questions for each of the input parameters corresponding to the user request;

determining a priority ranking for each of the input parameters corresponding to the user request based on an average position of the input parameters in the POS tagged sentence in each of the historical conversation logs; and

ordering the one or more questions based on the priority ranking for each of the input parameters.

10. The method of claim 9, further comprising:

parsing the first user input to produce a POS tagged sentence for the first user input; and

identifying, based on the POS tagged sentence for the first user input, input parameters for which the user has provided information.

11. The method of claim 10, further comprising:

determining the input parameters for which the user has not provided information;

identifying the priority ranking for each of the input parameters for which the user has not provided information;

based on the priority ranking for each of the input parameters, prompting the user to provide the second user input by presenting a question corresponding to the input parameter having a highest priority ranking; and

receiving, through the second representation of the webpage, the second user input.

12. The method of claim 11, further comprising:

identifying that the user has provided information for each of the input parameters corresponding to the user request; and

in response to identifying that the user has provided information for each of the input parameters corresponding to the user request, performing the user request.

13. One or more non-transitory computer-readable media storing instructions that, when executed by a computing platform comprising at least one processor, and memory, cause the computing platform to:

detect a change corresponding to input parameters for a first type of input for performing a user request through a first representation of a webpage, detecting the change corresponding to input parameters including identifying metadata for the input parameters;

parse a page document corresponding to the first representation of the webpage to generate a document object model (DOM) of the page document including one or more tag sequences;

isolate each of the one or more tag sequences related to user input fields for performing the user request;

extract, from each of the one or more isolated tag sequences, information corresponding to the input parameters and associated attributes of each of the user input fields;

based on the extracted information, assign each of the input parameters as either mandatory or optional;

update, based on the detected change corresponding to the input parameters and assessment of the metadata for the input parameters, a natural language processing application associated with a second representation of the webpage, updating the natural language processing application including updating input parameters associated with a second type of input different from the first type of input;

generate, based on the updated natural language processing application associated with the second representation of the webpage, a logical table including one or more questions for prompting a user to provide one or more answers for the input parameters through the second representation of the webpage;

receive, through the second representation of the webpage, a first user input corresponding to the user request;

responsive to receiving the first user input, prompt the user with the one or more questions regarding the input parameters for the user request;

responsive to prompting the user with the one or more questions, receive at least a second user input including the one or more answers for the input parameters; and

in response to receiving at least the second user input including the one or more answers, execute the user request with the one or more answers for the input parameters.

14. The one or more non-transitory computer-readable media of claim 13, storing additional instructions that, when executed by the computing platform, cause the computing platform to:

retrieve historical conversation logs from one or more databases stored on the computing platform;

isolate the historical conversation logs corresponding to the user request;

parse the isolated historical conversation logs corresponding to the user request to produce a part-of-speech (POS) tagged sentence for each of the historical conversation logs;

generate a mapping of the input parameters and associated attributes with the POS tagged sentence for each of the historical conversation logs;

build, based on the mapping, a confidence association of the input parameters and associated attributes with the POS tagged sentence for each of the historical conversation logs; and

update, based on the mapping and the confidence association, the natural language processing application associated with the second representation of the webpage.

15. The one or more non-transitory computer-readable media of claim 14, storing additional instructions that, when executed by the computing platform, cause the computing platform to:

identify, based on the mapping of the input parameters and associated attributes with the POS tagged sentence for each of the historical conversation logs, a particular question of the one or more questions for each of the input parameters corresponding to the user request;

determine a priority ranking for each of the input parameters corresponding to the user request based on an average position of the input parameters in the POS tagged sentence in each of the historical conversation logs; and

order the one or more questions based on the priority ranking for each of the input parameters.

16. The one or more non-transitory computer-readable media of claim 15, storing additional instructions that, when executed by the computing platform, cause the computing platform to:

parse the first user input to produce a POS tagged sentence for the first user input; and

identify, based on the POS tagged sentence for the first user input, input parameters for which the user has provided information.

17. The one or more non-transitory computer-readable media of claim 16, storing additional instructions that, when executed by the computing platform, cause the computing platform to:

determine the input parameters for which the user has not provided information;

identify the priority ranking for each of the input parameters for which the user has not provided information;

based on the priority ranking for each of the input parameters, prompt the user to provide the second user input by presenting a question corresponding to the input parameter having a highest priority ranking;

receive, through the second representation of the webpage, the second user input;

identify that the user has provided information for each of the input parameters corresponding to the user request; and

in response to identifying that the user has provided information for each of the input parameters corresponding to the user request, perform the user request.