Research On Early Warning Method Of Construction Safety Based On Text Mining

Building construction is one of the fields with high incidence of safety prod-uction accidents,so it is an important task to continuously improve the safety management level of building construction.In addition,the informatization of the building construction industry has made the industry accumulate a large number of data resources.How to mine the hidden knowledge of these data resources and organize this knowledge to serve safety management has become urgent.The refore,this study applies the ontology theory to the safety management of building construction,and realizes the mining and structured organization of hidden kn-owledge in domain text data resources.And try to realize the early warning of building construction safety accidents based on the ontology knowledge base,so as to improve the decision-making level of building construction safety managem-ent to a certain extent.This research first collected text data of accident cases,related laws and reg-ulations through web crawlers,used text word segmentation and hybrid methods to extract domain concepts,and applied the concept lattice method to mining dec-ision rules.Then based on the cause of the accident,the architecture was defined.Construction safety accident ontology model,through the process of constructing the ontology hierarchy,defining objects and data attributes,constructing ontology instances,etc.,using Protégé5.0 software to complete the construction of construc-tion safety accident ontology,and using the SWRLTab plug-in to build a rule base,and then formed Accident ontology knowledge base;Finally,choose the ap-propriate accident case,normalize and formalize the accident information in it into Jena's reasoning machine-readable rules,and write an inference program to load the rules and the accident ontology knowledge base.The feasibility and effe-ctiveness of the early warning method is verified by comparing with the accident report records.Through the research of this paper,the in-depth use and value mining of text data in the field of construction safety are realized.The acquired domain knowledge can provide knowledge services for the field of construction safety through the organization and storage of the ontology knowledge base,and also provide early warning of construction safety accidents And intelligent decisionmaking for emergency response.These studies have important practical significance for the development of building construction safety management in the era of data intelligence.