欢迎您！东篱公司
退出
-
申报数据库
-
立项数据库
-
成果数据库
-
项目获奖数据库
中文摘要：
本文构建了一个中文情感常识知识库,包括规则库和实例库两部分。目前的文本情感分析,主要以句子为计算单位,仅仅得出一个句子的情感,忽视了对于情感的持有者的分析,而且认为上下文句子的情感存在接续关系。本文尝试以情感的持有者作为文本情感计算的单位,认为一个情感持有者的情感存在接续关系,这一部分在规则库里进行描述。在情感计算的对象方面,当前主要以＂人＂为计算对象,而本文构建的实例库还包括一些非情感词汇,动物和特殊身份人物的情感常识知识,从而扩大了情感计算对象的范围。
英文摘要：
An Affective Commonsense Knowledgebase is constructed which consists of a rule knowledgebase and an instance knowledgebase.Now,people analyze a sentence＇s emotion without holder and assume that emotion in context influence each other.But,we think emotion must be connected with holder and this can be meaningful.We also think only the one＇s emotion has some rules in flow.All of these attempts are collected in the rule knowledgebase.On the other hand,because of the shortage of the affective computing object-only human beings,we collected some Affective Commonsense Knowledge in the instance knowledgebase,which relates to some animals,special identities and so on.We are sure that the instance knowledgebase can extend the affective computing object.
同期刊论文项目
同项目期刊论文
Web-oriented OOV translation extracting approach through cross-language query expansion for English-
BioPPISVMExtractor:A Protein-Protein Interaction Extractor for Biomedical Literature Using SVM and R
Uncertainty sampling-based active learning for protein-protein interaction extraction from biomedica
期刊信息
位置：