arXivDaily arXiv每日学术速递 周一至周五更新
arXiv cs/9907010cs.CL
收藏

具有置信限的语言识别

Language Identification With Confidence Limits

David Elworthy

摘要:

描述了一种统计分类算法及其在噪声输入中的语言识别中的应用。主要的创新是计算分类的置信限,这样当有足够的证据做出明确的决定时,算法就终止了,从而避免了具有相似特征的类别的问题。第二个应用程序,体裁识别,简要审查。结果表明,其他语言识别技术的一些问题可以避免,并说明了一个更重要的一点:统计语言过程可以用来提供有关其自身成功率的反馈。

英文摘要:

A statistical classification algorithm and its application to language identification from noisy input are described. The main innovation is to compute confidence limits on the classification, so that the algorithm terminates when enough evidence to make a clear decision has been made, and so avoiding problems with categories that have similar characteristics. A second application, to genre identification, is briefly examined. The results show that some of the problems of other language identification techniques can be avoided, and illustrate a more important point: that a statistical language process can be used to provide feedback about its own success rate.

机构、总结、中文摘要均由AI分析而来,可能存在错误,请以论文正文为准。

↑