摘要
知识图谱是把复杂的领域知识通过数据挖掘、信息处理、知识计量和图形绘制而显示出来,解释知识领域的动态发展规律。知识图谱把所有不同种类的信息(heterogeneous information)连接在一起得到一个关系网络并从"关系"的角度去分析问题。知识图谱目前被广泛应用于智能搜索、智能问答等领域。提出了一种基于知识图谱的智能决策支持框架,用于解决传统决策支持系统存在的问题。通过大数据、知识图谱等海量知识分析和模型构建技术,结合决策支持系统,增强对问题的分解与处理、形成具有关系型网络的知识系统。最后结合电信领域中的经典决策案例,搭建基于知识图谱的欺诈电话智能决策支撑平台。和传统的决策支持系统比较,该研究方法的优点在于结合大数据处理方法提升了知识建模的算力和决策支持的效率,使实时处理大规模信息数据成为现实;基于知识图谱的关系型网络,提升了决策模型的准确性和关联相关性。
Knowledge graph is the display of complex domain knowledge through data mining,information processing,knowledge measurement and graphical rendering to explain the dynamic development law of knowledge field.Knowledge graph links heterogeneous Information together to form a relationship network and analyze the problem from the perspective of"relationship".Knowledge graph is widely used in intelligent search,intelligent question answering and other fields.We present an intelligent decision support framework based on knowledge graph to solve the problems existing in traditional decision support systems.Through massive knowledge analysis and model building technologies such as big data and knowledge graph,and combined with decision support system,the knowledge system with relational network can be formed by enhancing the decomposition and processing of problems.At the end,a fraud telephone intelligent decision support platform based on knowledge graph is built by combining the classical decision cases in the telecommunication field.Compared with the traditional decision support system,the advantages of the research method proposed are that it combines the large data processing method to improve the computational power of knowledge modeling and the efficiency of decision support,and makes real-time processing of large-scale information data become a reality.The relational network based on knowledge graph improves the accuracy and relevance of decision model.
作者
魏瑾
李伟华
潘炜
WEI Jin;LI Wei-hua;PAN Wei(Northwestern Polytechnical University,Xi'an 710129,China)
出处
《计算机技术与发展》
2020年第1期1-6,共6页
Computer Technology and Development
基金
国家国防部“十三五”预研项目(31511090403)
