| In the logistics industry,transportation as the core link of the industry work,its operation process is indispensable and the quality of service to grasp the upstream and downstream work circulation lifeline.Transportation supervision is an important prerequisite and basis for the development of social economy and maintenance of urban order.With the continuous improvement of people’s living standards in China,technological innovation and information intelligence are widely concerned,and artificial intelligence applications are becoming more and more popular.Among them,intelligent transportation is the specific embodiment and carrier of the various scientific and technological innovations in the field of transportation,strengthening transportation supervision services,improving the governance capacity and service level of intelligent transportation,intelligent technology integration not only to reduce human labor,but also to enhance work productivity.About how to make good use of artificial intelligence technology in transportation governance issues,creating a good solution for users without increasing labor costs and reducing efficiency,has become a key issue of common concern in the political and academic circles.The question and answer system of transportation supervision service can solve for users,real-time online consultation of problems encountered in transportation in logistics operations,where this paper studies the technology and application design of knowledge graph can solve the problems of non-response,non-answer,only support single-round question and answer,and reply effect for slightly complex problems in use.Based on this,this study mainly answers three questions: How should the transportation field in logistics operation better discriminate information on user consultation topics? How to improve the accuracy of information retrieval through key algorithm design? How to make the information extraction technique in knowledge graph better applied in question and answer system and grasp the information for answer feedback?The accuracy of the entity information of natural language consultation question will affect the overall effect of the information extraction technology in the Q&A knowledge graph required to build the intelligent Q&A system,which not only directly affects the extraction of event topics,elements and entity relationships,but also indirectly affects the accuracy of answer search,decision event classification and knowledge logic management in the Q&A knowledge graph.In order to further improve the accuracy of the entity information extraction of customer voice consultation questions and enhance the overall effect of the information extraction technology in the knowledge graph of the intelligent question answering system,the authors optimized semantic annotation and applied the Bi LSTM-CRF(Bidirectional Long Short-Term Memory Conditional Random Filed)with BERT(Bidirectional Encoder Representation from Transformers)model to learn entity extraction of sentences.Entity extraction and semantic annotation were validated in specific experiments with the voice text of customer voice consultation question events as the data source.The results show that,based on semantic annotation optimization,the BERT model works better than the single Bi LSTM-CRF method for semantic annotation extraction.The proposed model harmonic mean reaches 91.53%which is of practical significance for enhancing the event entity information extraction. |
PDF Full Text Request