發(fā)布時間：2018-06-18 08:18

  本文選題：cn域名 + 日志數(shù)據(jù)　； 參考：《南京師范大學》2013年碩士論文

【摘要】：互聯(lián)網(wǎng)作為全球性的信息網(wǎng)絡對人們的生產(chǎn)和生活產(chǎn)生著深刻影響,在用戶從互聯(lián)網(wǎng)上獲取信息的同時,服務器會對用戶的訪問行為進行相應的記錄,生成互聯(lián)網(wǎng)日志數(shù)據(jù)�；ヂ�(lián)網(wǎng)日志數(shù)據(jù)蘊含了大量的空間信息。目前,對于互聯(lián)網(wǎng)日志數(shù)據(jù)的研究主要是利用數(shù)據(jù)挖掘、機器學習等技術對用戶訪問行為和系統(tǒng)安全性等進行分析和監(jiān)測,而在空間層面上的研究還相對較少。對互聯(lián)網(wǎng)日志數(shù)據(jù)的研究與地理空間相結合,利用地理信息系統(tǒng)中的技術手段和方法對互聯(lián)網(wǎng)日志數(shù)據(jù)進行空間上的挖掘,可以有效的獲取其隱含的空間規(guī)律,在為互聯(lián)網(wǎng)日志數(shù)據(jù)的分析提供了更寬的視野和角度的同時,也可促進以信息流為研究對象的信息地理學研究,具有重要的理論意義和實踐價值。 本文以從中國互聯(lián)網(wǎng)絡信息中心獲取的連續(xù)24小時的cn域名服務器日志數(shù)據(jù)為研究基礎數(shù)據(jù),針對cn域名服務器日志數(shù)據(jù)的特點,實現(xiàn)對cn域名服務器日志數(shù)據(jù)的海量數(shù)據(jù)處理、地理編碼和空間化,并從全球、地區(qū)和局域三個尺度對其進行空間分析,旨在理論上為信息地理學拓展研究領域,在技術上構建網(wǎng)絡日志數(shù)據(jù)處理和空間化表達的技術體系,在實踐上為網(wǎng)絡基礎設施建設提供參考。主要研究內(nèi)容及結論如下： (1)對信息地理學、互聯(lián)網(wǎng)DNS服務以及屬性數(shù)據(jù)空間化的相關理論體系進行梳理�？偨Y了信息地理學的內(nèi)涵以及研究方向。歸納了互聯(lián)網(wǎng)DNS服務的概念、體系結構以及工作原理,指出域名服務器日志數(shù)據(jù)具有結構化和海量性的特點。概括了屬性數(shù)據(jù)空間化的涵義和基本方法。在理論梳理的基礎之上,構建頂級域名網(wǎng)絡訪問空間特征分析的理論體系。 (2)針對cn域名服務器日志數(shù)據(jù)結構化、海量性的特點以及傳統(tǒng)地理信息系統(tǒng)在海量數(shù)據(jù)處理方面的不足,構建了海量數(shù)據(jù)處理框架,實現(xiàn)對日志數(shù)據(jù)的高效處理。經(jīng)過對日志數(shù)據(jù)中地理信息的提取和對訪問量空間分布的影響因素的分析,設計了cn域名服務器日志數(shù)據(jù)的空間化方法并對其進行實現(xiàn)。 (3)從全球、地區(qū)以及局域三個尺度對cn域名服務器日志數(shù)據(jù)進行了空間分析,揭示了cn域名下網(wǎng)絡服務的訪問量在空間上的分布格局。探索了這種分布格局形成的原因以及反映出的問題。這些分析結果將為未來中文網(wǎng)絡在中國乃至全球的發(fā)展提供參考和指導。
[Abstract]:As a global information network, the Internet has a profound impact on people's production and life. When users obtain information from the Internet, the server records the user's access behavior and generates the Internet log data. Internet log data contains a lot of spatial information. At present, the research on Internet log data mainly uses data mining, machine learning and other technologies to analyze and monitor user access behavior and system security. Combining the research of Internet log data with geographical space, and using the technical means and methods of GIS to mine the Internet log data in space, we can effectively obtain its implicit spatial rule. It not only provides a wider perspective and angle for the analysis of Internet log data, but also promotes the study of information geography with information flow as its research object, which has important theoretical significance and practical value. In this paper, the basic data of CN domain name server log data obtained from China Internet Network Information Center for 24 hours are studied, aiming at the characteristics of CN domain name server log data. To realize the massive data processing, geo-coding and spatialization of CN domain name server log data, and to carry out spatial analysis from three scales of global, regional and local, in order to expand the research field of information geography theoretically. The technical system of web log data processing and spatial expression is constructed in technology, and the reference is provided for the construction of network infrastructure in practice. The main research contents and conclusions are as follows: 1) combing the related theories of information geography, Internet DNS service and attribute data spatialization. The connotation and research direction of information geography are summarized. This paper summarizes the concept, architecture and working principle of Internet DNS service, and points out that the domain name server log data has the characteristics of structure and magnanimity. The meaning and basic method of attribute data spatialization are summarized. On the basis of theoretical combing, this paper constructs a theoretical system for analyzing the spatial characteristics of top-level domain name (TLDN) access space. (2) aiming at the structure of CN domain name server log data, Based on the characteristics of magnanimity and the shortcomings of traditional GIS in mass data processing, a massive data processing framework is constructed to achieve efficient processing of log data. After extracting the geographical information from the log data and analyzing the factors affecting the spatial distribution of the access amount, the spatial method of CN domain name server log data is designed and realized. The spatial analysis of CN domain name server log data is carried out at the three scales of region and local area, and the spatial distribution pattern of the traffic volume of network service under CN domain name is revealed. The reasons for the formation of this distribution pattern and the problems reflected are explored. These results will provide reference and guidance for the future development of Chinese language network in China and even the world.
【學位授予單位】：南京師范大學
【學位級別】：碩士
【學位授予年份】：2013
【分類號】：TP311.13;P208

【參考文獻】

相關期刊論文 前5條

1 黃瑩;包安明;陳曦;劉海隆;楊光華;;基于綠洲土地利用的區(qū)域GDP公里格網(wǎng)化研究[J];冰川凍土;2009年01期

2 甄峰;信息時代新空間形態(tài)研究[J];地理科學進展;2004年03期

3 汪明峰,寧越敏;互聯(lián)網(wǎng)與中國信息網(wǎng)絡城市的崛起[J];地理學報;2004年03期

4 季成;李曉東;袁堅;尉遲學彪;山秀明;;基于k-means算法的DNS查詢模式分析[J];清華大學學報(自然科學版);2010年04期

5 蔡俊;宋順林;;基于Web日志的頻繁偏愛路徑挖掘算法[J];計算機工程與設計;2009年24期

，

本文編號：2034808