Mining And Application Of Tourism Information Based On Multi-topic Emotional Dictionary

With the widespread application of the mobile Internet in people's daily lives,users using social media store travel information on online platforms.These travel information mostly contain rich evaluation information on different topics in the scenic area.Processing and mining this information can explore the underlying rules,this can quickly provide tourists with personal travel information to meet their needs for personalized travel.At the same time,it can help scenic area managers to monitor the entire tourism management object and provide data support for corresponding decision analysis.Social media data is difficult to use directly because of their unstructured characteristics.Rich topical evaluation information is hidden in unstructured and huge text corpora.In current research,mining travel information usually use a general dictionary for sentiment calculations,but the general dictionary has a single mining dimension,which cannot meet the mining criteria for multi-topic tourism evaluation information.This paper combines natural language processing and machine learning and other algorithms to build a set of automated mining processes for multi-topic evaluation of emotional information in tourism.The main work and achievements include the following aspects:(1)This paper builds a system for automatic collection and processing of social media data.In the data collection process,collection procedures are developed for different tourism websites.During the data processing process,the current data processing method and accuracy is improved based on the particularities of data in the tourism field.(2)The paper based on the characteristics of the tourism field and the national A-level scenic spot evaluation criteria formulates the theme classification of the scenic area.This paper based on the current general sentiment dictionary and integrated multiple machine learning models designed a set of multi-topic sentiment dictionary based on the online review in the tourism field.(3)According to the multi-topic sentiment dictionary constructed in this paper,the themes and corresponding emotional attitudes of tourist reviews in the corpus are identified.The dictionary can recognize the positive and negative comments of tourists on different topics of the scenic spot.This paper integrated the algorithm flow to build a tourism information mining and analysis system for the tourism field.The system implements functions such as data acquisition,data management,sentiment analysis,and scenic spot comparison analysis,and it has a good application analysis effect.(4)This paper first summarizes the spatiotemporal pattern of tourists' attention from a macro perspective,and then explores the evolution pattern of the spatiotemporal pattern of attention in the scenic area based on the inter-annual changes in the amount of tourist review data.Five representative attractions are selected to explore which topics are concentrated in high attention.Secondly,from a micro perspective,through in-depth analysis of changes in the fine-grained emotions of various topics,and providing targeted comments for tourists and scenic area managers,it can save a lot of manpower costs and time for tourists and scenic area managers.