本文選題：用戶行為分析 + 交互過(guò)程　； 參考：《北京交通大學(xué)》2015年碩士論文

【摘要】：隨著網(wǎng)站用戶數(shù)量的增多,管理者與運(yùn)營(yíng)人員等急需了解站點(diǎn)被訪問(wèn)的情況、用戶的行為表現(xiàn)等,以輔助站點(diǎn)管理和提供決策支持。在網(wǎng)站各職能崗位對(duì)用戶行為分析的需求中,識(shí)別用戶與網(wǎng)頁(yè)交互的“規(guī)律”則是最亟待實(shí)現(xiàn)的。用戶在網(wǎng)站的每一次操作都可以通過(guò)頁(yè)面或事件追蹤進(jìn)行記錄,一次完整會(huì)話就可以統(tǒng)計(jì)成為一條事件軌跡,從海量的事件軌跡中挖掘用戶與網(wǎng)頁(yè)的實(shí)際交互過(guò)程是解決該問(wèn)題的關(guān)鍵所在。挖掘結(jié)果將會(huì)反映出大部分用戶對(duì)網(wǎng)站操作的適應(yīng)與否,還會(huì)間接反映用戶的操作習(xí)慣以及對(duì)網(wǎng)站的流程理解。 目前在網(wǎng)頁(yè)用戶行為分析領(lǐng)域?qū)换ミ^(guò)程的研究還只停留在了熱力圖、訪問(wèn)頁(yè)面序列圖、點(diǎn)擊流模型等以頁(yè)面為基礎(chǔ)的分析方案。除了不能細(xì)化至頁(yè)面事件外,沒(méi)有模型化的分析結(jié)果、缺少與預(yù)期流程的差異比較、無(wú)法完成性能分析也成為了這些方案的主要缺點(diǎn)。 本文以挖掘用戶交互過(guò)程為核心需求點(diǎn),首先提出了對(duì)流程圖圖元的擴(kuò)展,用以更好地表達(dá)預(yù)期流程；然后從網(wǎng)站訪問(wèn)日志中獲得會(huì)話的事件軌跡后通過(guò)a算法挖掘出實(shí)際交互過(guò)程模型；繼而與預(yù)期流程圖進(jìn)行比較,分析差異、計(jì)算相似度；最后通過(guò)事件時(shí)間戳定量統(tǒng)計(jì)出實(shí)際交互過(guò)程的性能。除了交互過(guò)程分析外,系統(tǒng)還將實(shí)現(xiàn)流量分析、內(nèi)容效率分析、用戶分析三類網(wǎng)站基礎(chǔ)數(shù)據(jù)統(tǒng)計(jì)業(yè)務(wù),以滿足更多需求。 論文選題源自本人在實(shí)習(xí)工作中實(shí)際參與的項(xiàng)目,本人在項(xiàng)目中負(fù)責(zé)所有的功能設(shè)計(jì)和本文所提到模塊的實(shí)現(xiàn)開(kāi)發(fā)。本文在分析了系統(tǒng)需求后,詳細(xì)介紹了交互過(guò)程分析的算法設(shè)計(jì),最后以系統(tǒng)設(shè)計(jì)與實(shí)現(xiàn)部分本人開(kāi)發(fā)的模塊為核心進(jìn)行了技術(shù)架構(gòu)及功能實(shí)現(xiàn)的詳細(xì)介紹。 本文最后通過(guò)一個(gè)真實(shí)的在線購(gòu)票網(wǎng)站作為實(shí)驗(yàn)對(duì)象來(lái)演示α算法挖掘交互過(guò)程的可行性,證明了工作具有一定的研究和工程應(yīng)用價(jià)值。
[Abstract]:With the increase of the number of users in the website , the manager and the operator need to know the situation of the site being accessed , the behavior of the user and so on , so as to assist the site management and provide the decision support .

At present , the research on the interaction process in the field of webpage user behavior analysis is still only on the page - based analysis scheme such as the thermal diagram , the access page sequence diagram , the click flow model and the like . In addition to being unable to drill down to the page event , there is no modeling analysis result , and the lack of the difference comparison with the expected process , and the failure to complete the performance analysis is also the main disadvantage of these schemes .

In this paper , based on the mining user interaction process as the core requirement point , the extension of the flowchart primitive is first proposed to better express the expected flow ;
then obtaining the event track of the session from the website access log , and digging out the actual interaction process model through the a algorithm ;
then comparing with the expected flow chart , analyzing the difference and calculating the similarity ;
In addition to the interactive process analysis , the system will realize the flow analysis , the content efficiency analysis , the user analysis three kinds of website basic data statistical business to meet more needs .

This paper analyzes the system demand , introduces the algorithm design of interactive process analysis , and finally introduces the technical architecture and function realization in the system design and the module developed by myself .

Finally , a real online ticket booking website is used as the experimental object to demonstrate the feasibility of the alpha algorithm mining interaction process , which proves that the work has certain research and engineering application value .

【學(xué)位授予單位】：北京交通大學(xué)
【學(xué)位級(jí)別】：碩士
【學(xué)位授予年份】：2015
【分類號(hào)】：TP393.092

【參考文獻(xiàn)】

相關(guān)期刊論文 前2條

1 孫玲芳;夏聰;;Web使用挖掘在用戶行為分析中的應(yīng)用[J];江蘇科技大學(xué)學(xué)報(bào)(自然科學(xué)版);2011年03期

2 朱鵬;黃必清;薛霄;吳澄;吳蕓;;基于過(guò)程挖掘的醫(yī)療服務(wù)過(guò)程建模[J];計(jì)算機(jī)集成制造系統(tǒng);2010年12期

，

本文編號(hào)：1781847