本文提出了按照语料库统计的理论对《元朝秘史》进行计算机自动校勘的方法,并列举了相比人工校勘的优越性。通过计算机自动校勘,可以发现大量人工校勘很难发现的错误,通过设置阀值和置信区间,可以控制让计算机自动更正错误或者做标记,计算机不能判断之处标记出来留待人工审核。使用该方法可以节省大量的人工校验时间,是一种处理复杂文本的较好方法。 In this paper, we propose a computer-based automatic collation of the Secret History of Yuan Dynasty according to the theory of corpus statistics, and give an example of the superiority compared with manual collation. By computer automatically collation, you can find a lot of manual correction of hard to find errors, by setting the threshold and confidence interval, you can control the computer automatically correct the error or mark, the computer can not judge the place marked for manual review. Using this method can save a lot of time manually checking, is a better way to deal with complex texts.