殷希红,乔晓东,张运良.利用多策略模糊综合评判的术语关系识别方法研究[J].情报工程,2015,1(6):074-082 |
利用多策略模糊综合评判的术语关系识别方法研究 |
Research on the Method odTerm Relationship Recognition Using Multi-Strategy Fuzzy Comprehensive Evaluation |
|
DOI:10.3772/j.issn.2095-915X.2015.06.011 |
中文关键词: 模糊综合评判,多策略,关系识别,相似度 |
英文关键词: Fuzzy comprehensive evaluation, multi-strategy, term, relation recognition, similarity |
基金项目:国家自然科学基金项目“面向特定情报分析应用的知识组织系统快速构建关键问题研究”(71203208);国家科技支撑计划项目“面向科技情报分析的信息服务资源开发与支撑技术研究 ” (2015BAH25F01) |
作者 | 单位 | 殷希红 | 中国科学技术信息研究所 | 乔晓东 | 中国科学技术信息研究所 | 张运良 | 中国科学技术信息研究所 |
|
摘要点击次数: 2618 |
全文下载次数: 1814 |
中文摘要: |
利用多策略模糊综合评判的方法进行术语关系识别,首先采用多种相似度计算方法计算术语的相似度,然后利用连续属性离散化方法确定关系类别及阈值区间的划分,利用样本分布概率确定区间对类别的隶属度,利用粒子群算法和交叉验证法确定因素权重,最后利用模糊综合评判方法将所有相似度计算方法的计算结果进行融合处理,实现术语关系的识别。本研究将以中国科学技术信息研究所已有的新能源汽车领域汉语科技词系统中的术语作为测试集,用准确率、召回率和 F 值对关系识别的结果进行评价,论证该方法的有效性。 |
英文摘要: |
The multi-strategy fuzzy comprehensive evaluation is used to recognize the term relations. First, a variety of similarity calculation methods are used to calculate the terms similarity. Then, relations and threshold interval are identified by discretization algorithm of continuous attributes, the sample distribution probability is used to determine the membership degree of the interval to the relations, and the weight of elements are determined by particle swarm algorithm and cross validation method. All the calculation results are composited by fuzzy comprehensive evaluation method to recognize the term relations. At last, the precision, recall and F value are used to evaluate the effect of the results. This experiment regards the Chinese scientific and technical vocabulary system (new energy vehicles) as the test data. The result shows that the method can recognize the term relations effectively |
查看全文
查看/发表评论 下载PDF阅读器 |
关闭 |