基于网络外包的专业技能关联知识库构建
发布时间：2018-12-08 18:32
【摘要】：使用中文文本挖掘方法来分析中国高校网页中各专业培养方案和培养目标的非结构化数据集。以K-means文本聚类算法和聚类结果归纳的各专业类别的技能关键词为基础,在集成了所有专业领域的专有特征和专家审核并结合了频率计算方法后,定义了技能指标与相应各个专业的重要性程度。最后,建立了专业和技能之间的关联知识库,为构建网络化创新外包人才技能模型建立了基础。通过实验评估发现,与基于基本中文语料库的分词方法相比较,在中文分词过程中引入专业专有特征的方法能够提供更加精确和合理的聚类结果。因此,本文提出的方法能够高效地构建专业技能关联知识库。
[Abstract]:The Chinese text mining method is used to analyze the unstructured data sets of the training programs and objectives of each major in the web pages of Chinese colleges and universities. Based on the K-means text clustering algorithm and the skill keywords of each specialty category summarized by the clustering results, after integrating the specific features of all specialized fields and expert audit and combining the frequency calculation method, The skill index and the importance of each major are defined. Finally, the related knowledge base between specialty and skill is established, which is the foundation of constructing network innovation outsourcing talent skill model. The experimental results show that compared with the word segmentation method based on the basic Chinese corpus, the method of introducing specialized and exclusive features in the process of Chinese word segmentation can provide more accurate and reasonable clustering results. Therefore, the method proposed in this paper can efficiently construct the knowledge base of professional skills association.
【作者单位】： 上海交通大学安泰经济与管理学院;上海大学管理学院;
【基金】：国家自然科学基金青年项目(71301102);国家自然科学基金资助项目(71171131) 国家自然科学基金委创新研究群体资助项目(71421002) 长江学者和创新团队发展计划资助项目(IRT13030)
【分类号】：G250.74;TP391.1
，
本文编号：2368808
[Abstract]:The Chinese text mining method is used to analyze the unstructured data sets of the training programs and objectives of each major in the web pages of Chinese colleges and universities. Based on the K-means text clustering algorithm and the skill keywords of each specialty category summarized by the clustering results, after integrating the specific features of all specialized fields and expert audit and combining the frequency calculation method, The skill index and the importance of each major are defined. Finally, the related knowledge base between specialty and skill is established, which is the foundation of constructing network innovation outsourcing talent skill model. The experimental results show that compared with the word segmentation method based on the basic Chinese corpus, the method of introducing specialized and exclusive features in the process of Chinese word segmentation can provide more accurate and reasonable clustering results. Therefore, the method proposed in this paper can efficiently construct the knowledge base of professional skills association.
【作者单位】： 上海交通大学安泰经济与管理学院;上海大学管理学院;
【基金】：国家自然科学基金青年项目(71301102);国家自然科学基金资助项目(71171131) 国家自然科学基金委创新研究群体资助项目(71421002) 长江学者和创新团队发展计划资助项目(IRT13030)
【分类号】：G250.74;TP391.1
，
本文编号：2368808
本文链接：https://www.wllwen.com/kejilunwen/ruanjiangongchenglunwen/2368808.html
