Intelligent crowdsourced resource assistant

Patent metadata

Field	Value
Publication number	US-2019122162-A1
Application number	US-201715789418-A
Country	US
Kind code	A1
Filing date	Oct 20, 2017
Priority date	Oct 20, 2017
Publication date	Apr 25, 2019
Grant date	—

Abstract

A crowdsource assistant provides guidance to crowdsource resources navigating a crowdsource platform to bid for and be accepted to accomplish tasks. The crowdsource assistant utilizes machine learning to train similarity computation models that improve the crowdsource assistant's computing capability to generate more relevant recommendations to a resource in a faster and more efficient manner.

First claim

What is claimed is:

1 . An enterprise server computer comprising:
a communication interface configured to:
communicate with a crowdsource platform and a resource; and
receive, from the resource, profile information corresponding to the resource;
assistance circuitry coupled to the communication interface, the assistance circuitry configured to:
receive, through the communication interface, the profile information;
determine a sample resource, also on the crowdsource platform, is similar to the resource based on the profile information and sample profile information corresponding to the sample resource;
implement a comparison of the profile information with the sample profile information corresponding to the sample resource;
generate a profile recommendation based on the comparison, the profile recommendation including at least part of the sample profile information; and
presentation circuitry configured to:
generate a profile recommendation window within the crowdsource platform, the profile recommendation window including the profile recommendation.

2 . The enterprise server computer of claim 1 , wherein the assistance circuitry is further configured to:
generate a profile summary based on the profile information.

3 . The enterprise server computer of claim 1 , wherein the assistance circuitry is further configured to:
parse task attribute information corresponding to tasks posted within the crowdsource platform as being available for bidding; implement a comparison of the profile information against the task attribute information; and select an available task from the tasks posted within the crowdsource platform based on the comparison; and wherein the presentation circuitry is further configured to:
generate a task recommendation window within the crowdsource platform, the task recommendation window including the available task.

4 . The enterprise server computer of claim 3 , wherein the assistance circuitry is further configured to:
detect a selection of the available task from the resource; parse the crowdsource platform for other resources that have bid on the same available task; parse profile information corresponding to the other resources; and generate a ranking of resources, including the resource and the other resources, in an order of predicted match for the available task based on the profile information corresponding to the other resources, the profile information corresponding to the resource, and identified requirements of the available task.

5 . The enterprise server computer of claim 1 , wherein the assistance circuitry is further configured to:
parse a utilization history for the resource; and generate a marketplace recommendation for adding a new marketplace subscription based on the utilization history and the profile information.

6 . The enterprise server computer of claim 1 , wherein the assistance circuitry is further configured to:
parse a utilization history for the resource; and generate a marketplace recommendation for removing an existing marketplace subscription based on the utilization history and the profile information.

7 . The enterprise server computer of claim 1 , wherein the assistance circuitry is further configured to:
extract task attribute information from an available task posted within the crowdsource platform; determine a past task preference based on a comparison of the task attribute information with a utilization history for a resource; determine a profile preference based on a comparison of the task attribute information with profile information of the resource; determine a task poster preference based on a comparison of the resource and other resources utilized by a task poster of the available task posting; and generate a past task preference vector corresponding to the resource based on one of the past task preference, the profile preference, the task poster preference, or a combination thereof.

8 . The enterprise server computer of claim 7 , wherein the assistance circuitry is further configured to:
generate a worker-worker similarity matrix based on a similarity between the resource and other resources included within the crowdsource platform; and generate a collaborative preference based on the past task preference vector and the worker-worker similarity matrix.

9 . A method for assisting a resource on a crowdsource platform, the method comprising:
receiving, from a resource, profile information associated with the resource; determining a sample resource is similar to the resource based on the profile information corresponding to the resource and sample profile information corresponding to the sample resource, wherein both the resource and the sample resource are within the crowdsource platform; implementing a comparison of the profile information with the sample profile information corresponding to the sample resource; generating a profile recommendation based on the comparison, the profile recommendation including at least part of the sample profile information; and generating a profile recommendation window within the crowdsource platform, the profile recommendation window including the profile recommendation.

10 . The method of claim 9 , further comprising:
generating a profile summary based on the profile information.

11 . The method of claim 9 , further comprising:
parsing task attribute information corresponding to tasks posted within the crowdsource platform as being available for bidding; implementing a comparison of the profile information against the task attribute information; selecting an available task from the tasks posted within the crowdsource platform based on the comparison; and generating a task recommendation window within the crowdsource platform, the task recommendation window including the available task.

12 . The method of claim 11 , further comprising:
detecting a selection of the available task from the resource; parsing the crowdsource platform for other resources that have bid on the same available task; parsing profile information corresponding to the other resources; and generating a ranking of resources, including the resource and the other resources, in an order of predicted match for the available task based on the profile information corresponding to the other resources, the profile information corresponding to the resource, and identified requirements of the available task.

13 . The method of claim 9 , further comprising:
parsing a utilization history for the resource; and generating a marketplace recommendation for adding a new marketplace subscription based on the utilization history and the profile information.

14 . The method of claim 9 , further comprising:
parsing a utilization history for the resource; and generating a marketplace recommendation for removing an existing marketplace subscription based on the utilization history and the profile information.

15 . The method of claim 9 , further comprising:
extracting task attribute information from an available task posted within the crowdsource platform; determining a past task preference based on a comparison of the task attribute information with a utilization history for a resource; determining a profile preference based on a comparison of the task attribute information with profile information of the resource; determining a task poster preference based on a comparison of the resource and other resources utilized by a task poster of the available task posting; and generating a past task preference vector corresponding to the resource based on one o

Assignees

• Accenture Global Solutions Ltd

Inventors

• Abhinav Kumar
• Jain Sakshi
• Dubey Alpana
• Bhatia Gurpriya Kaur
• Mccartin Blake

Classifications

• G06Q10/40

  Business processes related to social networking or social networking services · CPC title

• G06N5/04

  Inference or reasoning models · CPC title

• G06Q30/08

  Auctions · CPC title

• G06F16/3344

  using natural language analysis · CPC title

• G06Q10/063118 Primary

  Staff planning in a project environment · CPC title