Research On Tourism Recommendation Based On Association Rules Algorithm

With the rapidly development and widely application of the mobile network, the traditional services of providing tourism information could not satisfy people’s needs. Tourism industry is an important part of the national Five-year Plan and the concept of "Smart Tourism" is coming into being. It is often reflected in the wisdom of tourism services, management and marketing. The wisdom of providing tourism information services is the most important part of the three, How to provide wisdom tourism service to people is the challenge coming to us. Nowadays, big data is also a hot issue, and tourism data is also fits for the big data’s feature. Traditional methods could not solved the problems, and new technology tables and methods must be applied. Cloud Computing and Internet of Things technology are mostly widely used to deal with problems about tourism services these days, including the tourism service issues, data processing issues and resource issues to provide guidance to travel managers and provide a new travel patterns and methods for the public service.As the best platform for big data processing and storage in the era of big data, Hadoop distributed processing platform offers the possibility to make use of its unique programming model and data storage. Traditional tourism information services can be migrated to a cloud computing environment and carried out. More useful information can be got from a large number of travel data by writing the appropriate data mining algorithms to provide more satisfactory and efficient service for the user.Attribute data in this paper are got from well-known tourism websites Baidu and the hornet’s nest sites. With improved data mining algorithms, data is analyzed through the mining process in order to provide travel information services for specific visitors. This paper mainly includes the following specific research work:(1) Study the Hadloop’s working principle, study MapReduce programming model and runtime processes and HDFS works and storage principle in depth and write MapReduce JOB corresponding task program to cope with the problems to lay the foundation for the subsequent improved algorithm.(2) Study traditional association rules algorithm in depth and analyze pros and cons of algorithm. Based on existing algorithms, the author put forward Term consolidation pruning-based association rules algorithm and Adjacent group-based association algorithm. The former solves the mining process repeated mining problem and the latter solves the problem of reducing the generation of redundant rules. The final results show that the improved algorithm can guarantee excavation accuracy while improving operational efficiency of the algorithm and mining.(3) Get data from well-known tourist travel and tourism websites Baidu and hornet’s nest sites, study the methods of data acquisition and processing, do the work of data cleaning and preprocessing on specific issues, transform data into transaction data to prepare the ground for subsequent experimental data;(4) Finally, the author uses visitors tourism data getting from the third process and analyzes the data with improved arithmetic, then the author brings out a feasible method for recommending tourist attractions and meals, lodging according to tourists’ specific conditions.