發(fā)布時(shí)間：2018-03-09 09:10

  本文選題：增量學(xué)習(xí)　切入點(diǎn)：精準(zhǔn)廣告　出處：《山西財(cái)經(jīng)大學(xué)》2010年碩士論文　論文類型：學(xué)位論文

【摘要】： 網(wǎng)絡(luò)技術(shù)的飛速發(fā)展,廣告成為網(wǎng)絡(luò)盈利的一個(gè)主要手段。網(wǎng)絡(luò)廣告為越來越多的企業(yè)和機(jī)構(gòu)所了解,并且大部分企業(yè)和機(jī)構(gòu)都進(jìn)行了網(wǎng)絡(luò)廣告的投放。但是,網(wǎng)絡(luò)廣告形式多樣,具有動(dòng)態(tài)性。網(wǎng)絡(luò)廣告投放的隨意性和泛濫性,使網(wǎng)絡(luò)用戶應(yīng)接不暇,產(chǎn)生了厭煩的情緒,使網(wǎng)絡(luò)廣告的投放達(dá)不到預(yù)期的效果。針對(duì)這種情況,出現(xiàn)了精準(zhǔn)廣告投放這一概念。精準(zhǔn)廣告投放即向網(wǎng)絡(luò)用戶投放其感興趣的,并真正能夠及時(shí)供用戶所需的廣告信息。而通過對(duì)用戶的行為分析,對(duì)用戶進(jìn)行智能地分類,成為實(shí)現(xiàn)精準(zhǔn)廣告投放的有效方法,那么對(duì)網(wǎng)絡(luò)上的海量用戶進(jìn)行智能分類就成為進(jìn)行精準(zhǔn)廣告投放的重要研究?jī)?nèi)容。 由于海量的網(wǎng)絡(luò)用戶信息,采用數(shù)據(jù)挖掘方法,可以有效地對(duì)用戶進(jìn)行智能分類。然而由于所獲得信息的實(shí)時(shí)性與在線性,即在互聯(lián)網(wǎng)下的用戶的行為信息是不斷變化的,致使這些海量的信息并不能一次性全部獲得。而對(duì)于分批獲得數(shù)據(jù),一般的分類算法,需要不斷地重新的更新分類模型,而耗費(fèi)了大量的時(shí)間。增量學(xué)習(xí)則是對(duì)訓(xùn)練數(shù)據(jù)集的學(xué)習(xí)過程逐步展開,后續(xù)的學(xué)習(xí)結(jié)果是建立在先前學(xué)習(xí)結(jié)果的基礎(chǔ)上的。同時(shí),鑒于貝葉斯分類方法能夠充分利用先驗(yàn)知識(shí)學(xué)習(xí)這一特點(diǎn),在一定程度上解決了先驗(yàn)信息傳遞的問題。 因此,本文提出一種基于貝葉斯的增量學(xué)習(xí)算法用于對(duì)精準(zhǔn)廣告投放系統(tǒng)中用戶的分類,它是一個(gè)利用樣本知識(shí)來修正當(dāng)前知識(shí)的連續(xù)的、動(dòng)態(tài)的過程。通過對(duì)在線用戶的行為特征不斷進(jìn)行分析,根據(jù)獲得的數(shù)據(jù)信息利用貝葉斯增量學(xué)習(xí)對(duì)用戶進(jìn)行分類,不斷地更新分類模型,以達(dá)到更好的分類效果,從而更有效地實(shí)現(xiàn)精準(zhǔn)廣告的投放。本文選擇了Book-Crossing(BX)數(shù)據(jù)集作為實(shí)驗(yàn)研究對(duì)象,它包括278858個(gè)用戶(匿名但有人口統(tǒng)計(jì)信息),提供了對(duì)于271379本圖書的1149780評(píng)分信息。通過利用SQL Server 2005處理成實(shí)驗(yàn)所需的數(shù)據(jù)格式。研究結(jié)果表明,增量學(xué)習(xí)對(duì)于在線的實(shí)時(shí)學(xué)習(xí)能夠解決其他分類器所帶來的時(shí)間與精力的耗費(fèi),并能獲得較好的分類效果,從而精確地、及時(shí)地對(duì)網(wǎng)絡(luò)用戶的分類,達(dá)到精準(zhǔn)廣告投放的目的。本文對(duì)精準(zhǔn)廣告投放系統(tǒng)進(jìn)行了設(shè)計(jì),從系統(tǒng)分析到系統(tǒng)功能的實(shí)現(xiàn)都進(jìn)行了論述,并提出主要是將用戶推薦模塊作為一個(gè)通用接口,不僅在本系統(tǒng)中可以應(yīng)用,在其他的網(wǎng)站中,只要將代碼嵌入并做相應(yīng)的調(diào)整,就同樣可以實(shí)現(xiàn)對(duì)當(dāng)下網(wǎng)站的用戶的分類。本文結(jié)論部分對(duì)整個(gè)文章進(jìn)行了總結(jié),并提出下一步工作。
[Abstract]:With the rapid development of network technology, advertising has become one of the main means of making profits on the Internet. More and more enterprises and institutions have learned about online advertising, and most of them have carried out online advertising. However, Network advertisement has various forms and is dynamic. The randomness and flood of network advertisement put in, make the network user be overwhelmed, produce weariness mood, make the network advertisement put in can't reach the expected effect. In view of this kind of situation, The concept of precision advertising has emerged. Precision advertising, that is, delivering information that is of interest to users on the Internet and can really provide users with the advertising information they need in a timely manner, can be intelligently classified by analyzing the behavior of users. It has become an effective method to achieve precision advertising, so the intelligent classification of mass users on the network becomes an important research content of precision advertising. Because of the huge amount of network user information, using data mining method, users can be classified intelligently. However, because of the real-time and linearity of the information obtained, that is, the behavior information of users under the Internet is constantly changing. As a result, these huge amounts of information can not be obtained all at once. However, for batch data, general classification algorithms need to constantly update the classification model. Incremental learning is the gradual expansion of the learning process of the training data set, and the subsequent learning results are based on the previous learning results. At the same time, In view of the fact that Bayesian classification method can make full use of the characteristics of prior knowledge learning, the problem of prior information transmission is solved to a certain extent. Therefore, an incremental learning algorithm based on Bayesian is proposed to classify the users in the precision advertisement delivery system, which uses sample knowledge to modify the continuity of current knowledge. Dynamic process. Through the continuous analysis of the behavior characteristics of online users, according to the obtained data information, Bayesian incremental learning to classify users, constantly update the classification model, in order to achieve a better classification effect, In this paper, the Book-Crossing BX data set is selected as the experimental research object. It includes 278,858 users (anonymous but demographically available, provides 1149780 rating information for 271379 books). It is processed into experimental data formats using SQL Server 2005. Incremental learning for online real-time learning can solve the cost of time and energy brought by other classifiers, and can obtain better classification effect, thus accurately and timely classification of network users. In this paper, the precision advertising delivery system is designed, from the system analysis to the realization of the system functions are discussed, and it is proposed that the user recommendation module as a general interface, Not only can it be applied in this system, but also in other websites, as long as the code is embedded and adjusted accordingly, it can also realize the classification of the users of the current website. And put forward the next work.
【學(xué)位授予單位】：山西財(cái)經(jīng)大學(xué)
【學(xué)位級(jí)別】：碩士
【學(xué)位授予年份】：2010
【分類號(hào)】：TP311.13

【參考文獻(xiàn)】

相關(guān)期刊論文 前10條

1 鮑玉斌,王大玲,于戈;關(guān)聯(lián)規(guī)則和聚類分析在個(gè)性化推薦中的應(yīng)用[J];東北大學(xué)學(xué)報(bào);2003年12期

2 李娜;李愛軍;;基于用戶特征分類的精準(zhǔn)廣告投放研究[J];電腦知識(shí)與技術(shù);2010年01期

3 夏勝平;劉建軍;袁振濤;虞華;張樂鋒;郁文賢;;基于集群的增量分布式RSOM聚類方法[J];電子學(xué)報(bào);2007年03期

4 吳飛,莊越挺,潘云鶴;基于增量學(xué)習(xí)支持向量機(jī)的音頻例子識(shí)別與檢索[J];計(jì)算機(jī)研究與發(fā)展;2003年07期

5 王飛,劉大有,王淞昕;基于遺傳算法的Bayesian網(wǎng)結(jié)構(gòu)增量學(xué)習(xí)的研究[J];計(jì)算機(jī)研究與發(fā)展;2005年09期

6 姜卯生,王浩,姚宏亮;樸素貝葉斯分類器增量學(xué)習(xí)序列算法研究[J];計(jì)算機(jī)工程與應(yīng)用;2004年14期

7 冀俊忠,沙志強(qiáng),劉椿年;貝葉斯網(wǎng)模型在推薦系統(tǒng)中的應(yīng)用研究[J];計(jì)算機(jī)工程;2005年13期

8 劉波;潘久輝;;基于Ensemble的增量分類方法[J];計(jì)算機(jī)工程;2008年19期

9 丁振國(guó),陳靜;基于關(guān)聯(lián)規(guī)則的個(gè)性化推薦系統(tǒng)[J];計(jì)算機(jī)集成制造系統(tǒng)-CIMS;2003年10期

10 宮秀軍,劉少輝,史忠植;一種增量貝葉斯分類模型[J];計(jì)算機(jī)學(xué)報(bào);2002年06期

相關(guān)博士學(xué)位論文 前2條

1 沈斌;關(guān)聯(lián)規(guī)則相關(guān)技術(shù)研究[D];浙江大學(xué);2007年

2 史e，

本文編號(hào)：1587890