KNN和SVM并行结合的算法

KNN and SVM Parallel algorithm

根据话题跟踪的定义和特点,分析了K最近邻(KNN)算法和支持向量机(SVM)算法的优缺点,发现它们的优缺点具有互补的可能性,提出了KNN和SVM并行结合的算法作为话题跟踪算法,设计了话题跟踪实验,实验结果证明了新算法作为话题跟踪算法,考虑了话题跟踪的特点,利用了KNN算法和SVM算法的理论优势而避免了理论的缺陷,处理话题跟踪问题时具有很好的话题跟踪效果.

According to the definition and characteristics of topic tracking,the advantages and disadvantages of the K-nearest neighbor(KNN)algorithm and support vector machines(SVM)algorithm were analyzed.It is found that their advantages and disadvantages have the possibility of complementary,proposes parallel with KNN and SVM algorithm as topic tracking algorithm,designs topic tracking experiments,and experimental results prove that the new algorithm as the topic tracking algorithm considers the characteristics of the topic tracking,takes advantage of theoretical advantages of the KNN algorithm and SVM algorithm while avoiding their theoretical defect and has a good topic tracking effect in dealing with the problems of topic Tracking.