The Construction Of Online Review Knowledge Map Based On Multiple Data Sources

With the advent of Web 2.0,the era of the open Internet where Internet users actively generating large amounts of information and sharing it with others has begun.With the popularity of online shopping,product and user review data on various e-commerce sites has grown exponentially.Users can realize their own information and sharing needs by publishing online reviews,meanwhile express their own shopping experience and product feelings through reviews,and provide support for other users to make purchasing decisions.Enterprises can also mine user needs from the information published by these users to improve products and services.However,a large amount of data caused the problem of information overload.The quality of a large number of online reviews on the Internet was uneven,which greatly affected the reference basis for user purchase decisions and the needs of enterprises to identify users.Therefore,how to effectively organize a large number of unstructured products and reviews Information,and the in-depth mining and management of information,has become a research focus in recent years.Online reviews are presented in the form of semi-structured or unstructured data,such as text,numbers,images,etc.This information is spread across multiple data sources on the Internet,covering all aspects of products and services.It is difficult for users,merchants,or manufacturers to efficiently find the information they need in multiple data sources and conduct in-depth analysis of the information.Therefore,based on the fusion of heterogeneous review data from multiple sources,a large amount of unstructured product and review information is effectively organized,and the information is deeply mined and managed to provide users with higher-quality information services and become an urgent problem.The Knowledge Graph proposed by Google in 2012 made intelligent search possible.Knowledge Graph is a very effective way to organize information.Knowledge graphs can express various entities,relationships between entities,and attributes of entities.Specifically,knowledge graphs have two basic organizational units: <entity,relationship,entity>,<entity,attribute,entity>.The knowledge graph can fuse structured product information and unstructured review information to achieve intelligent and automated applications.At the same time,the knowledge graph's pattern layer can provide an efficient data management model.The processed data is no longer fragmentation and redundancy,perfectly solve the problem of information overload.Therefore,this paper uses knowledge map construction technology to propose a knowledge map framework that integrates multiple heterogeneous review information.It organizes and mines multi-source heterogeneous online reviews to enrich and support users' purchase decisions and mining user demand data sources.Users provide higher quality information services to solve the problem of information overload.First,according to the current research status of online reviews,analyze user information needs,including ordinary customers,e-commerce platforms,and producers,analyze different user information needs,and lay the foundation for subsequent model layer construction;Then,based on the results of the requirements analysis and referring to the existing knowledge graph construction technology,an online review knowledge graph construction scheme is designed and implemented,including the detailed process of the pattern layer,data layer,and storage.Finally,based on the above research results,this paper designs and implements the construction of a mobile review review knowledge map,and displays the review knowledge map,semantic retrieval,and data mining on the basis of this.