Adaptive crowdsourcing method based on deep reinforcement learning
A reinforcement learning and self-adaptive technology, applied in the field of worker selection and task allocation, can solve problems such as tasks affecting production efficiency, achieve more flexible task allocation, improve work efficiency and effect, and ensure quality.
Active Publication Date: 2018-09-28
ZHEJIANG UNIV
View PDF6 Cites 14 Cited by
- Summary
- Abstract
- Description
- Claims
- Application Information
AI Technical Summary
Problems solved by technology
[0004] In the application of crowdsourcing technology, the specific assignment of tasks will greatly affect production efficiency
Method used
the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
View moreImage
Smart Image Click on the blue labels to locate them in the text.
Smart ImageViewing Examples
Examples
Experimental program
Comparison scheme
Effect test
Embodiment Construction
[0038] In order to make the object, technical solution and advantages of the present invention clearer, the present invention will be further described in detail below in conjunction with the accompanying drawings.
[0039] refer to figure 1 , is the implementation flow of the deep reinforcement learning-based adaptive crowdsourcing method of the present invention. The adaptive crowdsourcing method based on deep reinforcement learning includes the following steps:
[0040] S1. First sample the information of the crowdsourcing tasks and crowdsourcing workers that need to be assigned from the crowdsourcing system;
[0041] In this step, the original features of the crowdsourcing task include task classification labels, task text content, and estimated degree of difficulty; the original features of the crowdsourcing workers include sex, age, time distribution of task completion, total historical number of assigned tasks, The total number of tasks completed in history, the assig...
the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
Login to View More PUM
Login to View More Abstract
The present invention discloses an adaptive crowdsourcing method based on deep reinforcement learning. The method is specifically as follows: 1) sampling to-be-assigned tasks and candidate crowdsourcing workers from a crowdsourcing system; 2) obtaining low-dimensional feature representation of the to-be-assigned tasks and the candidate workers through the deep learning method; 3) determining a task assignment strategy through the reinforcement learning method; 4) assigning the tasks through the crowdsourcing system according to the assignment strategy, evaluating the income obtained from the assignment according to the task completion result, feeding back the income to the reinforcement learning method, and updating the reinforcement learning parameter; and 5) starting from the 1) to continue to perform the next round of task assignment. Compared with the prior art, according to the method provided by the present invention, the deep reinforcement learning method is combined, modeling is systematically performed on the task assignment problem, appropriate crowdsourcing workers are selected for the characteristics of different tasks, an adaptive intelligent crowdsourcing method is formed, and the work efficiency and effect of crowdsourcing are creatively improved.
Description
technical field [0001] The invention relates to the application of a deep reinforcement learning method in a crowdsourcing system, in particular to a technical method for worker selection and task assignment in the crowdsourcing system. Background technique [0002] With the rapid development of the Internet and the advancement of information globalization, the crowdsourcing model came into being. Crowdsourcing is a new form of production organization brought about by the Internet, which has changed the traditional solution. It is a distributed problem-solving method, that is, using the Internet to decompose and distribute related work and break it into parts. Put idle productivity to good use by rewarding users appropriately for participation. Crowdsourcing has been identified as a potential problem-solving mechanism for governments and non-profit organizations. [0003] Crowdsourcing has a wide range of applications in data annotation, book digitization, and knowledge gr...
Claims
the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
Login to View More Application Information
Patent Timeline
Login to View More IPC IPC(8): G06N3/08G06Q10/06
CPCG06N3/084G06Q10/06311
Inventor 张寅杨璞胡滨
Owner ZHEJIANG UNIV