【摘要】：提出一種基于LDA模型(latent Dirichlet allocation model)與主題知識(shí)庫相結(jié)合的網(wǎng)絡(luò)日志內(nèi)容屬性標(biāo)注方法。IP知識(shí)庫的建立首先需要對采集的網(wǎng)絡(luò)日志進(jìn)行數(shù)據(jù)預(yù)處理;然后基于統(tǒng)計(jì)學(xué)原理標(biāo)注網(wǎng)絡(luò)日志的時(shí)間類屬性,利用IP地址庫映射方法提取網(wǎng)絡(luò)日志的地域類屬性;最后采用一種基于LDA模型與主題知識(shí)庫相結(jié)合的標(biāo)注方法對網(wǎng)絡(luò)日志的內(nèi)容類屬性進(jìn)行挖掘。結(jié)果表明,該方法原理正確,對網(wǎng)絡(luò)日志屬性的挖掘具有較好的效果。
[Abstract]:A network log content attribute annotation method based on LDA model and subject knowledge base is presented. The establishment of IP knowledge base needs to preprocess the collected network log first;Then, based on the statistical principle, the time type attribute of the network log is marked, and the regional attribute of the network log is extracted by the IP address library mapping method;Finally錛，

本文編號(hào)：2509262