Research On Knowledge Map Construction Method For Bad Network Information

With the development of Internet,the network to provide information services to everyone at the same time also brought some security risks.And now the network information detection and filtering is a single form,it can not meet the needs of the development of information industry.In recent years,the rapid development of knowledge map in information retrieval has brought new opportunities for the retrieval and filtering of bad information.Therefore,this paper presents a research on the construction of knowledge map for network bad information.Based on the existing technology and algorithms,this paper studies the construction of knowledge map for network bad information.The overall construction process of this paper contains three parts: data collection,knowledge extraction and map construction.The main contents of the study in the process of knowledge map construction are as follows:(1)Studying the method of constructing knowledge map for network bad information.Based on the present situation of the development of the knowledge map at this stage and the data source of the network bad information,this paper summarizes the knowledge map construction method and steps for the bad information of the network.In the construction of the knowledge map using top-down and bottom-up methods.Using the top-down approach to build the schema of the knowledge map,and use the bottom-up approach to construct the data layer of the knowledge map.(2)Research on knowledge extraction algorithm based on deep confidence network.In the stage of knowledge extraction of unstructured text,this paper uses an algorithm model based on deep confidence network.The model needs to input the parameters vector and the neuron node as parameters.Therefore,we need to use the language cloud platform of Harbin Institute of Technology to divide the corpus and part of speech,and then use the word2 vec tool of Google to extract the lexical feature vector.(3)Achieving the construction and storage of knowledge map.This paper constructs the knowledge map for the bad information of the network.In this paper,through the acquisition of Chinese Wikipedia,Sogou encyclopedia of entity classification combined with the corresponding knowledge reasoning to build a knowledge map of the model layer,and the data layer of the construction and data collection of the entity classification information inseparable.Finally,the integration of good data into the map database Neo4 j.