User-task associated crowdsourcing task recommendation method based on graph neural network

Abstract

The invention discloses a user-task associated crowdsourcing task recommendation method based on a graph neural network, and the method comprises the steps: collecting the interaction information between users generated by real users, the information of crowdsourcing tasks historically undertaken by the users, and the preference label information of the users for tasks through a crowdsourcing task platform, and respectively constructing a user association graph, a task association graph and a user-task association graph; respectively training by using a graph neural network so as to obtain preliminary vector representations of the user and the product; utilizing the new graph neural network to reserve crowdsourcing task information undertaken by the user in history; and through the learned user and product representation, obtaining a task recommendation list for the user. According to the method, the task recommendation information for the user is obtained, and for the newly-added user, if the connection between the newly-added user and the historical user is obtained, a part of recommendation cold start problems can be better solved.

Description

technical field [0001] The invention relates to the field of crowdsourcing task recommendation, is mainly used in recommendation systems, and specifically relates to the design of a graph neural network and the construction of an association graph. Background technique [0002] With the vigorous development of the Internet, individuals or organizations can use a large number of Internet users to obtain needed services and ideas instead of obtaining professional knowledge through a small number of experts. "Crowdsourcing" is a portmanteau created in 2006 by mixing the meanings of crowd and outsourcing. This is a model that assigns work to many participants and then synthesizes it into the final result. Therefore, it is necessary to ensure the quality of crowdsourcing tasks and recommend suitable tasks to suitable users. How to recommend suitable tasks to different users? Their tasks are faced with great challenges. [0003] Graph neural network is a branch of graph represen...

AI Technical Summary

Problems solved by technology

[0005] Purpose of the invention: The technical problem to be solved is that in the traditional crowdsourcing task recommendation system, the measurement of task similarity is relatively rough and there are few recommendable tasks that users can access. The present invention provides a user- Task-associated crowdsourcing task recommendation method, the invention establishes a relatively unified user and task representation, so as to better measure the relevance between users and tasks, and obtain a better and wider recommendation sequence

Images