Dynamic Knowledge Graph Algorithm for Electricity Knowledge
Jingyi Su, Nan Zhang, Zhen Sun, Wenqing Yang · 2023
When employees of electricity companies want to know relevant information in the electricity field in real time, such as the latest policies, news, standards, cases, etc., traditional information retrieval alone cannot meet people's needs for fast and accurate knowledge acquisition. This paper introduces electricity knowledge graph as auxiliary information to help knowledge recommendation. To maintain the freshness and completeness of knowledge recommendation, it is necessary to improve the freshness of knowledge graph, thus we study the dynamic updating method of knowledge graph before knowledge recommendation in electricity field. First, we study the dynamic updating method of the knowledge graph in the electricity field, and then continues to study the knowledge recommendation method based on the knowledge graph that can be dynamically updated. In the automatic updating part of the knowledge graph, the missing information is supplemented according to the existing information in the knowledge graph to achieve internal updating. When the known knowledge is limited and it is difficult to further supplement by inferring the missing knowledge, the hot new knowledge is discovered according to the network new word discovery, and the domain entity is identified, the dynamic multivariate attribute labeling algorithm (DWAL) is added to the knowledge graph to update the knowledge. Considering the dynamic of knowledge recommendation, this paper introduces a time correction factor in the process of user interest modeling to dynamically reflect the changes of users' interests, which successfully reflects the dynamic changes of users' interests and ultimately improves the quality of recommendation.