ARTIFICIAL INTELLIGENT AGENT CROWDSOURCING
Aspects of the present disclosure relate to systems and methods for updating description of a place of interest. The systems and methods comprise receiving a proposed place edit to a description of a place as a report from a reporting agent, inserting the report with the proposed place edit in a queue of proposed place edits to the place, receiving one or more votes on accepting the proposed place edit by a voting agent, and updating the description of the place according to the accepted edit. The reporting agent performs according to a first machine learning model that automatically generates an edit to a description of a place of interest based on information obtained by crawling one or more websites that publish information about places of interest. The voting agent comprises a second machine learning model that predicts acceptance of a given edit to a description of a place.
1 . A method for editing description of a place of interest by crowdsourcing, comprising:
receiving a piece of proposal for editing description of the place of interest;
accepting, based on verification of authenticity of the piece of proposal, the piece of proposal; and
updating the description of the place of interest according to the accepted piece of proposal.
2 . The method according to claim 1 , wherein the received piece of proposal is automatically created by a reporting agent by executing a first machine learning model, and the first machine learning model retrieves information about the place of interest over a network and creates the piece of proposal.
3 . The method according to claim 1 , wherein
the accepting the piece of proposal further comprises interactively accepting the piece of proposal through an interactive operations by a superuser with an elevated privilege to vote on the piece of proposal for acceptance.
4 . The method according to claim 1 , wherein accepting the piece of proposal further comprises:
automatically accepting, by a voting agent through executing a second machine learning model, the piece of proposal, wherein the voting agent comprises a website search agent and a verification agent, wherein the website search agent automatically performs searching and identifying, based on the piece of proposal, a website to retrieve content, and wherein the verification agent performs, based on the retrieved content from the website, automatic verification of the piece of proposal as ground truth data.
5 . The method according to claim 4 , further comprising:
calibrating, based on an automatic execution of the accepting the piece of proposal by the voting agent, one or more parameters of the second machine learning model through an interactive user interface to train the voting agent, thereby enabling the voting agent to accept the piece of proposal with accuracy.
6 . The method according to claim 4 , further comprising:
automatically creating, by a reporting agent through executing the first machine learning model, the piece of proposal for editing the description of the place of interest;
updating, based on a result of validating accuracy of the automatically created piece of proposal, one or more parameters of the first machine learning model, thereby training the first machine learning model to enable the reporting agent to perform creating another piece of proposal for editing the description of the place of interest with accuracy.
7 . The method according to claim 4 , further comprising:
updating, based on a result of the automatically accepting the piece of proposal by the voting agent, one or more parameters of the second machine learning model to calibrate the voting agent, thereby causing the voting agent to perform a subsequent automatic acceptance of another proposal for editing the description of the place of interest with accuracy.
8 . The method according to claim 4 , wherein the first machine learning model automatically generates, based on received information about the place of interest as input, the piece of proposal for editing the description of the place of interest.
9 . The method according to claim 4 , wherein the second machine learning model automatically marks, based on verifying content of the website describing the place of interest, acceptance of the piece of proposal to edit the description of the place of interest.
10 . A computer-readable non-transitory recording medium storing a computer-executable program instructions that when executed by a processor cause a computer system to execute operations comprising:
receiving a piece of proposal for editing description of a place of interest;
accepting, based on verification of authenticity of the piece of proposal, the piece of proposal; and
updating the description of the place of interest according to the accepted piece of proposal.
11 . The computer-readable non-transitory recording medium according to claim 10 , wherein the received piece of proposal has been automatically created by a reporting agent by executing a first machine learning model, and the first machine learning model retrieves information about the place of interest over a network and creates the piece of proposal.
12 . The computer-readable non-transitory recording medium according to claim 10 , wherein the accepting the piece of proposal further comprises interactively accepting the piece of proposal through an interactive operations by a superuser with an elevated privilege to vote on the piece of proposal for acceptance.
13 . The computer-readable non-transitory recording medium according to claim 10 , wherein the accepting the piece of proposal further comprises automatically accepting, by a voting agent through executing a second machine learning model, the piece of proposal, the voting agent comprises a website search agent and a verification agent, the website search agent automatically performs searching and identifying, based on the piece of proposal, a website to retrieve content, and the verification agent performs, based on the retrieved content from the website, automatic verification of the piece of proposal as ground truth data.
14 . The computer-readable non-transitory recording medium according to claim 13 , the computer-executable program instructions when executed further causing the computer system to execute operations comprising:
calibrating, based on an automatic execution of the accepting the piece of proposal by the voting agent, one or more parameters of the second machine learning model through an interactive user interface to train the voting agent, thereby enabling the voting agent to accept the piece of proposal with accuracy.
15 . The computer-readable non-transitory recording medium according to claim 13 , the computer-executable program instructions when executed further causing the computer system to execute operations comprising:
automatically creating, by a reporting agent through executing the first machine learning model, the piece of proposal for editing the description of the place of interest;
updating, based on a result of validating accuracy of the automatically created piece of proposal, one or more parameters of the first machine learning model, thereby training the first machine learning model to enable the reporting agent to perform creating another piece of proposal for editing the description of the place of interest with accuracy.
16 . The computer-readable non-transitory recording medium according to claim 13 , the computer-executable program instructions when executed further causing the computer system to execute operations comprising:
updating, based on a result of the automatically accepting the piece of proposal by the voting agent, one or more parameters of the second machine learning model to calibrate the voting agent, thereby causing the voting agent to perform a subsequent automatic acceptance of another proposal for editing the description of the place of interest with accuracy.
17 . The computer-readable non-transitory recording medium according to claim 13 , wherein the first machine learning model automatically generates, based on received information about the place of interest as input, the piece of proposal for editing the description of the place of interest.
18 . T The computer-readable non-transitory recording medium according to claim 13 , wherein the second machine learning model automatically marks, based on verifying content of the website describing the place of interest, acceptance of the piece of proposal to edit the description of the place of interest.
19 . A system comprising:
at least one processor; and
memory encoding computer executable instructions that, when executed by the at least one processor, perform operations comprising:
instantiating a reporting agent to automatically generate a proposed place edit to at least a part of a description of a place of interest; and
instantiating a voting agent, wherein the voting agent automatically marks acceptance of the proposed place edit created by the instantiated reporting agent.
20 . The system of claim 10 , wherein the proposed place edit is generated using a first machine learning model and wherein the acceptance of the proposed edit is marked using a second machine learning model.