本文首先在欧共体项目（LC-STAR）的资助下，开展了大规模汉语平衡语料的收集与分析工作，其主要目的是建立一个反映现代汉语语言特点的、适用于汉语语言分析、语音识别和语音合成的汉语标注语料库，并以此为基础建立相应的信息词典。-Our work supported by European Union’s project of LC-STAR, which includes collection and analysis of a large-scale balance-corpus, aims to

拼写检查，建立词库，输入单词，检查正误，如果错误就改正为词库中最接近的单词-Spell checking, thesaurus established, enter the word, check the right and wrong, if the mistake is corrected the closest word lexicon

In this paper, we systematically explore feature definition and selection strategies for sentiment polarity classification. We begin by exploring basic questions, such as whether to use stemming, term frequency versus binary weighting, negation-enr