【摘要】 本文提出了一种新的汉字识别方法—“包含配选法”。文中介绍了此方法的原理,并报告了为验证此种方法的有效性所做的实验。实验中对常用的6000个汉字中的5659个按偏旁、部首分为198类,若把机分结果稳定地进入一类的文字视为正确分类,则正确分类率为97.29％;若把机分结果稳定地进入两类中某一类的121个汉字也算入正确分类,则正确分类率为99.43％。更多还原

【Abstract】 This paper presents a new method of Chinese character recognition-"Inclusive Matching Method". The paper first introduces the principle of the method and then the experiments undertaken to prove the effectiveness of the method. In the experiments, 5659 out of 6000 commonly used characters are classified into 198 groups in accordance with character components and radicals. If only characters classified into one group are considered to be in correct classification, the correct classifying rate is 97.29%, but if the 121 characters which enter either of the two groups are also counted in the correct classification, the correct classifying rate becomes 99.43%.更多还原