發(fā)布時(shí)間：2018-01-29 21:31

  本文關(guān)鍵詞： SaaS 數(shù)據(jù)挖掘 weka 分類預(yù)測(cè) 聚類 關(guān)聯(lián)規(guī)則　出處：《西安電子科技大學(xué)》2014年碩士論文　論文類型：學(xué)位論文

【摘要】：SaaS模式作為一種完全創(chuàng)新的軟件應(yīng)用模式，已經(jīng)被認(rèn)為是IT行業(yè)發(fā)展的無可爭(zhēng)議的方向。與此同時(shí)，SaaS平臺(tái)上的數(shù)據(jù)在急劇的增長，海量數(shù)據(jù)處理和海量數(shù)據(jù)應(yīng)用成為了SaaS平臺(tái)一個(gè)重要的問題。SaaS平臺(tái)上的數(shù)據(jù)具有復(fù)雜、海量、異構(gòu)、多源等特點(diǎn)，這樣對(duì)SaaS平臺(tái)數(shù)據(jù)分析面臨著巨大的挑戰(zhàn)，所以需要借助數(shù)據(jù)挖掘軟件來完成SaaS平臺(tái)數(shù)據(jù)分析。由于SaaS發(fā)展起來的時(shí)間并不長，針對(duì)SaaS平臺(tái)的功能正在不斷的擴(kuò)展完善，本文根據(jù)用戶及SaaS廠商的需要提出SaaS平臺(tái)數(shù)據(jù)挖掘的解決方案。 本文首先分析了SaaS平臺(tái)的大數(shù)據(jù)，并對(duì)SaaS平臺(tái)數(shù)據(jù)存儲(chǔ)方式及數(shù)據(jù)特點(diǎn)進(jìn)行了分析，根據(jù)上述分析提出了SaaS平臺(tái)數(shù)據(jù)挖掘功能需求及系統(tǒng)需求。 其次，本文對(duì)數(shù)據(jù)挖掘技術(shù)進(jìn)行了簡(jiǎn)單描述，簡(jiǎn)要介紹了數(shù)據(jù)挖掘的流程及經(jīng)典數(shù)據(jù)挖掘算法。針對(duì)SaaS平臺(tái)，在比較當(dāng)前流行的數(shù)據(jù)挖掘軟件后本系統(tǒng)選擇了公開的數(shù)據(jù)挖掘軟件Weka，并針對(duì)SaaS平臺(tái)注冊(cè)公司信息數(shù)據(jù)、公司使用用戶數(shù)據(jù)、SaaS模塊租用信息數(shù)據(jù)分別提出了相對(duì)應(yīng)的數(shù)據(jù)挖掘算法。 然后，本文提出SaaS平臺(tái)數(shù)據(jù)挖掘系統(tǒng)設(shè)計(jì)方案，采用分模塊的設(shè)計(jì)思想，SaaS平臺(tái)數(shù)據(jù)挖掘系統(tǒng)分為存儲(chǔ)各種數(shù)據(jù)的數(shù)據(jù)模塊、進(jìn)行數(shù)據(jù)轉(zhuǎn)換及預(yù)處理的數(shù)據(jù)服務(wù)模塊、進(jìn)行數(shù)據(jù)挖掘算法實(shí)現(xiàn)及應(yīng)用程序?qū)崿F(xiàn)模塊。 最后，本文實(shí)現(xiàn)了SaaS平臺(tái)數(shù)據(jù)挖掘過程，封裝了SaaS平臺(tái)數(shù)據(jù)挖掘算法，并通過實(shí)驗(yàn)驗(yàn)證了本系統(tǒng)的有效性及實(shí)用性。實(shí)驗(yàn)結(jié)果表明：租用者在選擇SaaS模塊時(shí)，SaaS提供商可推薦相應(yīng)關(guān)聯(lián)性大的模塊；公司根據(jù)員工信息對(duì)員工進(jìn)行分類，公司可以更方便的對(duì)員工進(jìn)行分層次管理；通過對(duì)注冊(cè)公司性質(zhì)的預(yù)測(cè)，，SaaS提供商可更好的做好SaaS商品營銷策略。
[Abstract]:As a completely innovative software application model, SaaS model has been regarded as the indisputable direction of IT industry development. At the same time, the data on SaaS platform is growing rapidly. Mass data processing and data application has become an important problem in SaaS platform. The data on SaaS platform is complex, massive, heterogeneous, multi-source and so on. In this way, the data analysis of SaaS platform is facing a huge challenge, so we need to use data mining software to complete the data analysis of SaaS platform. Because SaaS is not developed for a long time. In view of the continuous expansion and perfection of the function of SaaS platform, this paper puts forward the solution of SaaS platform data mining according to the needs of users and SaaS manufacturers. This paper first analyzes the big data of the SaaS platform, and analyzes the data storage mode and data characteristics of the SaaS platform. According to the above analysis, the functional requirements and system requirements of SaaS platform data mining are put forward. Secondly, this paper gives a brief description of data mining technology, introduces the flow of data mining and classic data mining algorithms, aiming at the SaaS platform. After comparing the current popular data mining software, the system chooses the open data mining software Weka. and for the SaaS platform to register company information data, the company uses user data. The corresponding data mining algorithms are proposed for SaaS module rental information data. Then, this paper puts forward the design scheme of SaaS platform data mining system, which is divided into data modules to store all kinds of data. Data service module for data conversion and preprocessing, data mining algorithm implementation and application program implementation module. Finally, this paper implements the SaaS platform data mining process, encapsulates the SaaS platform data mining algorithm. The validity and practicability of the system are verified by experiments. The experimental results show that the renters can recommend the relevant modules with great relevance when choosing the SaaS module. The company classifies the employee according to the employee information, the company can carry on the stratification management to the staff more conveniently; By predicting the nature of registered companies, SaaS providers can do a better job of marketing strategy.
【學(xué)位授予單位】：西安電子科技大學(xué)
【學(xué)位級(jí)別】：碩士
【學(xué)位授予年份】：2014
【分類號(hào)】：TP311.13;TP393.09

【參考文獻(xiàn)】

相關(guān)期刊論文 前10條

1 翁曉奇;李妙旎;;基于SaaS應(yīng)用模式的企業(yè)信息化前景分析[J];硅谷;2009年19期

2 鞏軍全;;數(shù)據(jù)挖掘及其軟件選擇應(yīng)用[J];消費(fèi)導(dǎo)刊;2009年11期

3 高陽;廖家平;吳偉;;基于決策樹的ID3算法與C4.5算法[J];湖北工業(yè)大學(xué)學(xué)報(bào);2011年02期

4 趙進(jìn);;SaaS成熟度模型淺析[J];程序員;2008年08期

5 陳慧萍;林莉莉;王建東;苗新蕊;;WEKA數(shù)據(jù)挖掘平臺(tái)及其二次開發(fā)[J];計(jì)算機(jī)工程與應(yīng)用;2008年19期

6 顧天竺;沈潔;陳曉紅;李慧;張舒;吳顏;;基于XML的異構(gòu)數(shù)據(jù)集成模式的研究[J];計(jì)算機(jī)應(yīng)用研究;2007年04期

7 陳鵬;薛恒新;;面向中小企業(yè)信息化的SaaS應(yīng)用研究[J];中國制造業(yè)信息化;2008年01期

8 孟小峰;慈祥;;大數(shù)據(jù)管理:概念、技術(shù)與挑戰(zhàn)[J];計(jì)算機(jī)研究與發(fā)展;2013年01期

9 莫展宏;;國內(nèi)外SaaS模式的發(fā)展現(xiàn)狀分析[J];商場(chǎng)現(xiàn)代化;2012年07期

10 趙法信,王國業(yè);數(shù)據(jù)挖掘中聚類分析算法研究[J];通化師范學(xué)院學(xué)報(bào);2005年02期

相關(guān)博士學(xué)位論文 前1條

1 何月順;關(guān)聯(lián)規(guī)則挖掘技術(shù)的研究及應(yīng)用[D];南京航空航天大學(xué);2010年

本文編號(hào)：1474405