CAJ | 학술논문

为了进一步提高网络蜘蛛在互联网、物联网和实时工业控制网络中信息采集的效率，分析了导致网络蜘蛛陷入局部最优解的原因，将遗传算法引入到网络蜘蛛的应用当中求解全局最优解，针对传统遗传算法中存在早熟和收敛慢的问题对选择、交叉、变异这三种核心算子进行了改进。经实验对比表明，该算法和网络蜘蛛相结合克服了以上问题，具有较高的搜索查全率和搜索准确率。
위료진일보제고망락지주재호련망、물련망화실시공업공제망락중신식채집적효솔，분석료도치망락지주함입국부최우해적원인，장유전산법인입도망락지주적응용당중구해전국최우해，침대전통유전산법중존재조숙화수렴만적문제대선택、교차、변이저삼충핵심산자진행료개진。경실험대비표명，해산법화망락지주상결합극복료이상문제，구유교고적수색사전솔화수색준학솔。
In order to further raise the information acquisition efficiency of Web spider in the Internet,IoT and industrial real-time control networks,an analysis of the causes that lead Web spider into the local optimum was presented.Mean-while,the genetic algorithm (GA)for finding the global optimum was introduced in the application of Web spider.To avoid the slow convergence rate and premature convergence of the pure GA,an improved algorithm was proposed, which refined selection,crossover,and mutation of these three basic operators of GA.The experiment results show that this algorithm overcomes the above problems in combination with Web spider.Meanwhile,the recall ratio and retrieval precision are both increased.