摘要
研究了潜在语义索引在中文自动问答系统FAQ库构建中的应用,并着重阐述了句子相似度的计算方法以及使用LSI对FAQ库去重的实验选取方法,结果显示LSI方法在一定程度上优于TF×IDF方法。
We studied the application of the LSI （latent semantic index） for FAQ construction in Chinese automatic question and answer system, and emphatically introduces computing technology in sentence similarity and experiment methods of using latent semantic index for FAQ repetition. The result showed the LSI was better than the TF×IDF.
出处
《石河子大学学报（自然科学版）》
CAS
2005年第6期778-781,共4页
Journal of Shihezi University(Natural Science)
关键词
自动问答
FAQ
潜在语义索引
句子相似度
automatic question and answer
frequently-asked question
latent semantic index
sentence similarity
