基于CS-SVM的网络热点话题变化趋势预测  

Network hot topic change trend prediction based on cuckoo search algorithm and support vector machine

作　　者：邱仕坦[1,2] 

机构地区：[1]福州大学信息化建设办公室,福建福州350116 [2]福建省超级计算中心,福建福州350116

出　　处：《福州大学学报（自然科学版）》2014年第3期381-386,共6页Journal of Fuzhou University(Natural Science Edition)

基　　金：福建省自然科学基金资助项目(2010J05133)

摘　　要：为了提高网络热点话题变化趋势的预测精度,提出一种布谷鸟搜索(CS)算法优化支持向量机(SVM)参数的热点话题变化趋势预测模型(CS-SVM).首先获取热点话题的时间序列,然后将SVM参数作为一个鸟巢位置,通过CS算法模拟布谷鸟种群寄生繁衍机制找到最优参数,最后根据最优SVM参数建立热点话题变化趋势预测模型,并采用仿真实验对模型性能进行测试.结果表明,相对于对比预测模型,CS-SVM提高了热点话题变化趋势预测精度,可以准确刻画热点话题变化趋势,是一种理想的复杂、多变热点话题变化趋势预测工具.Network hotspot topic prediction is a complex prediction problem because it has small sam-ple, uncertainty, in order to improve the prediction accuracy of network hotspot topic change trend,this paper proposed a network hotspot topic prediction model based on cuckoo search （CS） algorithmoptimized support vector machine （SVM）. Firstly, time series of network hotspot topic prediction areobtained, and then the parameters of support vector machine are coded as a bird＇ s nest location, andthe optimal parameters of CS algorithm is obtained by simulating the cuckoo species parasitic reproduc-tion mechanism, finally, network hotspot topic prediction model is established according to the optimalparameters, and the simulation experiments are carried out to test the performance of the proposedmodel. The results show that the proposed model can improve the prediction precision of network hots-pot topic change trend compared with other prediction models, and can accurately describe the hot top-ic change trend, so it is a good prediction tool for complex network hotspot topic.

关 键 词：热点话题 布谷鸟搜索算法 支持向量机 参数优化 

分 类 号：TP393[自动化与计算机技术—计算机应用技术]