【摘要】：隨著移動(dòng)互聯(lián)網(wǎng)的快速發(fā)展和普及應(yīng)用,傳統(tǒng)的旅游信息服務(wù)已經(jīng)不能滿足大眾的需求,旅游產(chǎn)業(yè)作為國家十二五規(guī)劃的重點(diǎn)產(chǎn)業(yè)之一,“智慧旅游”的概念應(yīng)運(yùn)而生,它往往體現(xiàn)在旅游服務(wù)、管理和營銷的智慧。然而這三方面最被關(guān)心的就是旅游信息服務(wù)的智慧,如何做到給用戶智慧的旅游服務(wù),這是我們將要面臨的挑戰(zhàn)。如今,大數(shù)據(jù)也是熱點(diǎn)問題,旅游數(shù)據(jù)符合大數(shù)據(jù)的特點(diǎn),大量的旅游數(shù)據(jù)雜亂無章、數(shù)據(jù)巨大,如何處理和利用好這些數(shù)據(jù),傳統(tǒng)的方法已經(jīng)不能夠解決問題,必須采用新的技術(shù)平臺(tái)和方法。目前通常采用的方法是利用新的云計(jì)算和物聯(lián)網(wǎng)技術(shù),來處理旅游業(yè)中的旅游服務(wù)問題,將旅游產(chǎn)業(yè)中的旅游服務(wù)問題、數(shù)據(jù)處理問題和資源整合問題在新的大數(shù)據(jù)環(huán)境下利用數(shù)據(jù)挖掘的技術(shù)手段來為旅游管理者提供指導(dǎo),并為大眾服務(wù)提供一種全新的旅游形態(tài)和方法。在大數(shù)據(jù)時(shí)代,Hadoop平臺(tái)作為優(yōu)秀的分布式處理平臺(tái),給大數(shù)據(jù)的處理和存儲(chǔ)提供了可能性,利用它特有的編程模型和數(shù)據(jù)存儲(chǔ)方式,可以將傳統(tǒng)的旅游信息服務(wù)移植到云計(jì)算環(huán)境下進(jìn)行,通過編寫相應(yīng)的數(shù)據(jù)挖掘算法,可以從大量的旅游數(shù)據(jù)中獲取有效的信息,針對用戶提供更加滿意、高效的服務(wù)。本文從知名旅游網(wǎng)站百度旅游和馬蜂窩網(wǎng)站中獲取游客的屬性數(shù)據(jù),利用改進(jìn)的數(shù)據(jù)挖掘算法,在對數(shù)據(jù)分析挖掘處理之后,針對具體的游客提供旅游信息服務(wù)。本文具體的研究工作如下：(1)研究Hadoop的工作原理,深入學(xué)習(xí)MapReduce的編程模型和運(yùn)行流程和HDFS工作原理和存儲(chǔ)原理,針對問題,編寫對應(yīng)的MapReduce JOB任務(wù)程序,為后續(xù)的算法改進(jìn)打下了基礎(chǔ)。(2)深入學(xué)習(xí)傳統(tǒng)的關(guān)聯(lián)規(guī)則算法,并且分析了算法的優(yōu)缺點(diǎn),在現(xiàn)有算法的基礎(chǔ)上,提出了基于項(xiàng)合并剪枝的關(guān)聯(lián)規(guī)則算法和基于頻繁閉項(xiàng)集鄰接圖的關(guān)聯(lián)規(guī)則算法,前者解決了在挖掘過程中的重復(fù)挖掘問題,后者減少冗余規(guī)則的生成問題,最終的實(shí)驗(yàn)結(jié)果表明,改進(jìn)后的算法可以保證挖掘精準(zhǔn)度的同時(shí),提高算法的運(yùn)行和挖掘效率。(3)從知名旅游網(wǎng)站百度旅游和馬蜂窩網(wǎng)站獲取游客的旅游數(shù)據(jù),研究了數(shù)據(jù)的獲取和處理方法,針對具體問題對數(shù)據(jù)進(jìn)行了清洗和預(yù)處理工作,轉(zhuǎn)化為可以處理的事務(wù)數(shù)據(jù),為后續(xù)的實(shí)驗(yàn)做好基礎(chǔ)。(4)最后,我們利用(3)中獲取到的游客的旅游數(shù)據(jù),應(yīng)用改進(jìn)后的算法進(jìn)行挖掘分析,在游客的景點(diǎn)推薦和吃住行混合推薦方面給出了一種可行的方法,符合具體的實(shí)際情況。
[Abstract]:With the rapid development and popularization of the mobile Internet, the traditional tourism information service has not been able to meet the needs of the public. As one of the key industries in the 12th Five-Year Plan, the concept of "intelligent tourism" has emerged as the times require. It is often reflected in travel services, management and marketing wisdom. However, these three aspects are most concerned about the wisdom of tourism information services, how to provide intelligent travel services to users, which is the challenge we will face. Nowadays, big data is also a hot issue. Tourism data conform to the characteristics of big data. A large amount of tourism data is messy and huge. How to deal with and make good use of these data, the traditional methods can no longer solve the problem. New technology platforms and methods must be adopted. The current approach is to use new cloud computing and Internet of things technologies to deal with tourism services in the tourism industry, and tourism services in the tourism industry. Data processing problem and Resource Integration problem in the new environment of big data, the technology of data mining is used to provide guidance for tourism managers, and to provide a new form and method of tourism for mass service. In big data's time, Hadoop platform, as an excellent distributed processing platform, provided the possibility for the processing and storage of big data, and made use of its unique programming model and data storage method. Traditional tourism information services can be transplanted to cloud computing environment. By writing the corresponding data mining algorithm, we can obtain effective information from a large number of travel data, and provide more satisfied and efficient services for users. In this paper, the attribute data of tourists are obtained from well-known tourist websites Baidu Travel and Horse Honeycomb, and the improved data mining algorithm is used to provide tourist information services for specific tourists after processing the data analysis and mining. The specific research work of this paper is as follows: (1) study the working principle of Hadoop, deeply study the programming model and running flow of MapReduce, the working principle and storage principle of HDFS, and write the corresponding MapReduce JOB task program for the problem. It lays a foundation for further improvement of the algorithm. (2) deeply study the traditional association rules algorithm, and analyze the advantages and disadvantages of the algorithm, on the basis of the existing algorithms, An association rule algorithm based on item merging and pruning is proposed, and an association rule algorithm based on frequent closed itemsets adjacent graph is proposed. The former solves the problem of repeated mining in mining process, and the latter reduces the generation of redundant rules. The final experimental results show that the improved algorithm can ensure the accuracy of mining and improve the efficiency and efficiency of the algorithm. (3) the travel data of tourists are obtained from well-known tourism websites Baidu Travel and Horse Honeycomb. The methods of data acquisition and processing are studied. The data is cleaned and preprocessed for specific problems, which can be converted into transactional data. (4) finally, Using the tourist data obtained in (3), we apply the improved algorithm to mining and analysis, and give a feasible method in the tourist attraction recommendation and the mixed recommendation of eating, lodging and traveling, which is in line with the actual situation.
【學(xué)位授予單位】：陜西師范大學(xué)
【學(xué)位級(jí)別】：碩士
【學(xué)位授予年份】：2015
【分類號(hào)】：TP311.13;TP391.3

【參考文獻(xiàn)】

相關(guān)期刊論文 前3條

1 王端祥;;貴州旅游電子商務(wù)現(xiàn)狀及發(fā)展對策研究[J];貴州師范學(xué)院學(xué)報(bào);2010年05期

2 石培華;馮凌;;新時(shí)期中國旅游研究十大創(chuàng)新方向展望[J];北京第二外國語學(xué)院學(xué)報(bào);2010年05期

3 何華春,石軍南;基于GIS系統(tǒng)的旅游區(qū)研究——以武功山風(fēng)景旅游區(qū)為例[J];經(jīng)濟(jì)地理;2003年06期

，

本文編號(hào)：2230292