%0 Journal Article
%T Learning algorithm model of constructive PONN based on knowledge rules<br>基于知识规则的构造性优先排序神经网络算法
%A GU Yang-bo
%A WU Yan
%A ZHU Shi-jiao
%A WANG Shou-jue
%A <br>谷秧波
%A 武妍
%A 朱世交
%A 王守觉
%J 计算机应用
%D 2008
%I 
%X From the perspective of covering points in high-dimensional space, we discussed the theory of constrictive PONN based on knowledge rules. PONN's general constructive algorithm and its two special methods based on random sampling and barycenter rules respectively were proposed in this paper. Simulations on spirals recognition and voice language identification were conducted using the special methods. Experimental results illustrate the good performance of PONN instances. Besides, voice language identification results show that PONN method based on barycenter rule outperforms SVM under certain circumstances.
%K priority ordered neural network
%K knowledge rule
%K coverage<br>优先度排序神经网络
%K 知识规则
%K 覆盖
%U http://www.alljournals.cn/get_abstract_url.aspx?pcid=5B3AB970F71A803DEACDC0559115BFCF0A068CD97DD29835&cid=8240383F08CE46C8B05036380D75B607&jid=831E194C147C78FAAFCC50BC7ADD1732&aid=3DB48778D189810C3842547D148E798D&yid=67289AFF6305E306&vid=D3E34374A0D77D7F&iid=DF92D298D3FF1E6E&sid=E521CC33DC22B84E&eid=F8C186D6055F60DE&journal_id=1001-9081&journal_name=计算机应用&referenced_num=0&reference_num=11