摘要: 针对维吾尔语情感词汇获取难度大、人工扩充情感词汇工作量大且效率低的缺陷,结合维吾尔语主观文本语料的具体特点,分析维吾尔语情感词汇在情感语料中表现的特征,建立维吾尔语情感词汇的特征模板,利用条件随机场模型实现维吾尔语情感词汇的自动识别方法。实验结果验证该方法能有效自动识别情感词汇,降低人工识别情感词汇的工作量。
关键词:
维吾尔语,
情感词汇,
特征识别,
条件随机场
Abstract: For expanding uyghur emotional vocabulary and identificating emotional words automatically, efficiency of artificial works is improved. Combined with the specific characteristics of the uyghur subjective text corpus, this paper proposes the full analysis of characteristics of uyghur emotional words in the text, establishes uyghur emotional words feature template, using Conditional Random Fields(CRFs) model to achieve the uyghur emotional words automatic identification method. Experimental results show that the method can effectively recognize the emotional words, and reduce the workload of manual identification of emotional words.
Key words:
uyghur,
emotional words,
feature recognition,
Conditional Random Fields(CRFs)
中图分类号:
禹龙, 田生伟, 冯冠军. 维吾尔语情感词汇自动识别[J]. 计算机工程, 2011, 37(7): 213-215.
YU Long, TIAN Sheng-Wei, FENG Guan-Jun. Automatic Recognition of Uyghur Emotional Words[J]. Computer Engineering, 2011, 37(7): 213-215.