Research On Motivation Recognition Method Based On Event Logic Graph ——Taking The Field Of Outbound Tourism As An Instance

Tourism motivation is the starting point for tourists to formulate various travel activities,and affects all aspects of tourists' travel behavior.How to identify tourists' travel motivations from their behaviors and choices,and make rational allocation of tourism resources based on tourists' travel motivations,as well as reasonable planning of tourism policies,are urgent issues for the recovery and development of tourism in the post-epidemic era.question.As an important part of the tourism industry and the branch of tourism that is most affected by the epidemic,outbound tourism needs to explore and pay attention to the motivation of tourists for outbound tourism and make new adjustments and development based on the needs of tourists.Traditional tourism motivation identification is based on questionnaire survey and theoretical analysis methods to identify and analyze tourists' outbound travel motivation.With the vigorous development of various travel online websites,tourists are accustomed to sharing their travel experiences and experiences on travel websites,generating a large amount of motivation-related information with great analytical value.Various travel notes and tourist attraction reviews contain a large amount of information related to the motivation of tourists.How to organize and analyze these unstructured data,so as to excavate the intrinsic motivational factors of tourists' travel behavior,has gradually attracted the attention and exploration of scholars.This paper constructs an automated data collection framework to collect travel data of outbound travel from online travel websites,extracts and analyzes the event information in the travel notes through natural language processing technology,and builds a map of outbound travel events.The rational allocation of resources and the rational planning of tourism policies provide data support for the identification of tourism motives.Specifically,in the process of constructing the outbound tourism graph,this paper firstly performs word segmentation and semantic role labeling on the collected data,and then uses pattern matching and dependency parsing analysis to extract the succession relationship and event combination respectively.Through the elimination of meaningless events and event fusion,the outbound tourism Event logic graph is generated and stored in the graph database.Finally,based on the event logic graph of outbound travel affairs,a graph convolutional network model was constructed to identify the motivation of the event chain corresponding to the travel data,and the experimental effect of motivation identification exceeded the five baseline models selected in this paper.