Research On The Expansion Method Of Knowledge Graph In The Field Of Textile And Clothing Based On Crowdsourcing

Along with the spread of the Internet and the flourishing of the textile and clothing industry,data in the field of textile and clothing has increased significantly,and plenty of multi-source heterogeneous,complicated and unorganized knowledge have been accumulated on the Internet.Due to the lack of hierarchy and systematization,it is hard for users to seek and maintain knowledge,and it is very costly for users to acquire high-quality knowledge.Therefore,there is an eager need to realize the efficient retrieval and co-construction and share of resources in the textile and clothing field.Knowledge graph directly reveals the knowledge structure and its relevance in the form of visual graph with semantic information expression,which has good semantic information and hierarchical structure.Because of its friendly knowledge representation method,knowledge graph provides a low cognitive cost guarantee for the public to participate in knowledge construction,which makes crowdsourcing knowledge construction possible.Crowdsourcing is a new distributed resource construction solution under the Internet,which uses public wisdom to solve problems.Crowdsourcing can take full advantage of group intelligence to expand knowledge graphs,which is more conducive to improving the quality of knowledge graph.This paper constructs a domain knowledge graph by organizing the knowledge system in the field of textile and clothing domain.In response to the existing problems of uneven knowledge coverage,homogeneous forms and untimely updates in the construction of traditional knowledge graphs,a crowdsourcing-based knowledge graph expansion approach is proposed.By combining with the field of textile and clothing,the knowledge graph development system of textile and clothing supporting crowdsourcing mode is developed.The major work is classified into three main areas as follows.Firstly,a textile and clothing knowledge graph is constructed.By summarizing the knowledge system of textile and clothing,referring to professional books,domain theme websites and encyclopedia knowledge,the concepts,relationships and attributes and other characteristics of the domain ontology,construct the ontology concept system,and expand the graph by the method of entity extraction,which provides data base for the application service of knowledge graph of textile and clothing.Secondly,this paper studies the method of textile and clothing knowledge graph expansion based on crowdsourcing.Crowdsourcing methods are designed to guarantee the smooth execution of crowdsourcing tasks.Consider the bias of crowdsourcing participants’ understanding of the task brought by the difference of professional backgrounds,and design a quality control mechanism to ensure the accuracy and authority of the graph construction.Motivate the crowdsourcing participants by adding incentives to motivate users to complete the crowdsourcing tasks in time.Finally,a knowledge graph platform of textile and clothing based on crowdsourcing is designed and developed.The platform supports users to consult and retrieve knowledge in the form of graph structure,and supports users to cooperate and improve knowledge content through crowdsourcing tasks,so as to better connect knowledge resources and users,realize knowledge interconnection.