基于深度學習的CTR預測研究
發(fā)布時間：2018-09-04 06:27
【摘要】：伴隨著互聯(lián)網(wǎng)、云計算、物聯(lián)網(wǎng)等技術的迅猛發(fā)展,網(wǎng)絡的數(shù)據(jù)規(guī)模也在急劇增長,信息社會已經(jīng)慢慢步入“大數(shù)據(jù)”時代。網(wǎng)絡廣告投放系統(tǒng)架構于大數(shù)據(jù)的基礎上,系統(tǒng)利用機器學習對海量用戶行為進行分析挖掘并向用戶實時地推送合適的廣告。點擊率(Click Through Rate,CTR)預測是網(wǎng)絡廣告投放系統(tǒng)的核心技術,對于提升系統(tǒng)的運作效率意義重大。CTR的精準預測是制定科學的電子商務市場營銷決策的關鍵,直接影響用戶的網(wǎng)絡體驗,直接關系到互聯(lián)網(wǎng)公司的運營成本。因此,CTR的預測具有很高的商業(yè)價值和研究價值。面對網(wǎng)絡廣告投放系統(tǒng)的高精準度和高時效的要求,本文從淺層學習和深度學習兩個角度開展特征選擇、特征學習、分類預測和應用技術研究。以網(wǎng)絡廣告真實的數(shù)據(jù)集為實驗對象,分別構建淺層學習模型和深度學習模型。為了全面驗證深度學習模型,本次研究通過多視角的綜合對比實驗來證實深度學習的巨大潛力。綜合考慮,具體的研究工作主要包括以下五個方面:(1)開展數(shù)據(jù)處理和特征工程技術研究。從真實數(shù)據(jù)集出發(fā)探索研究類別不平衡性對預測模型的影響機理,不平衡數(shù)據(jù)的重采樣技術。(2)針對數(shù)據(jù)特征的高度非線性特點,開展淺層學習和深度學習理論及應用技術對比研究。為了克服淺層模型對復雜問題學習能力受限問題,構建深度學習模型,實驗通過算法實現(xiàn)證實了相對比淺層學習,深度學習的預測效果提升了約21%,具有很強的優(yōu)勢。(3)為消除類別不平衡對預測模型的影響,提出了一種深度神經(jīng)網(wǎng)絡(Deep Neural Network,DNN)的改進模型——SDNN(Deep Neural Network based on Sampling,SDNN)�；贕PU的并行計算,通過構建模型和實現(xiàn)算法驗證了在不影響預測效果的同時,SDNN預測模型訓練時間縮短了約73.28%,大幅度的提升了DNN的運算效率。針對系統(tǒng)的精準度和時效性的高要求,SDNN被證實是一種面向大數(shù)據(jù)更為高效的預測方法。(4)研究Sigmoid激活函數(shù)和Relu激活函數(shù)對DNN預測模型的影響機理。通過分別構建DNN和SDNN模型和算法的實現(xiàn),證實了相對比Sigmoid激活函數(shù),Relu激活函數(shù)更適合于層次較深的網(wǎng)絡模型,基于Relu激活函數(shù)的DNN和SDNN更適合復雜問題的建模。(5)為了避免單一SDNN訓練的局限性提升模型的泛化能力,開展關鍵參數(shù)dropout敏感性分析研究。
[Abstract]:With the rapid development of Internet, cloud computing, Internet of things and other technologies, the data scale of the network is also growing rapidly, the information society has entered the "big data" era. Based on big data, the system uses machine learning to analyze and mine massive user behavior and push appropriate advertisements to users in real time. The prediction of click rate (Click Through Rate,CTR) is the core technology of the network advertisement delivery system. It is of great significance to improve the operational efficiency of the system. The accurate prediction is the key to making scientific electronic commerce marketing decisions, which directly affects the network experience of users. Directly related to the operating costs of Internet companies. Therefore, the prediction of CTR has high commercial value and research value. In the face of the requirement of high precision and high efficiency in the network advertising system, this paper carries out the research of feature selection, feature learning, classification prediction and application technology from the two angles of shallow learning and deep learning. Taking the real data set of network advertisement as the experimental object, the shallow learning model and the depth learning model are constructed respectively. In order to fully verify the depth learning model, this study verifies the great potential of depth learning through comprehensive comparative experiments from multiple perspectives. Considering synthetically, the concrete research work mainly includes the following five aspects: (1) carry out the research of data processing and feature engineering technology. Based on the real data set, this paper explores the influence mechanism of class imbalance on prediction model, and the resampling technique of unbalanced data. (2) aiming at the highly nonlinear characteristics of data features, To carry out a comparative study of shallow and deep learning theories and applied techniques. In order to overcome the problem of limited learning ability of shallow model for complex problems and to construct a deep learning model, the experiment proves that the learning ability of shallow model is compared with that of shallow learning. The prediction effect of depth learning is improved by about 21%, which has a strong advantage. (3) in order to eliminate the influence of class imbalance on prediction model, an improved model of depth neural network (Deep Neural Network,DNN) is proposed. Based on the parallel computation of GPU, it is verified that the training time of prediction model is reduced by 73.28%, and the efficiency of DNN is greatly improved by constructing model and implementing algorithm. It has been proved that SDNN is a more efficient prediction method for big data in view of the high requirement of accuracy and timeliness of the system. (4) the influence mechanism of Sigmoid activation function and Relu activation function on DNN prediction model is studied. By constructing DNN and SDNN models and algorithms, it is proved that compared with the Sigmoid activation function, Relu activation function is more suitable for the deeper network model. DNN and SDNN based on Relu activation function are more suitable for modeling complex problems. (5) in order to avoid the limitation of single SDNN training to improve the generalization ability of the model, the key parameter dropout sensitivity analysis is carried out.
【學位授予單位】：重慶工商大學
【學位級別】：碩士
【學位授予年份】：2017
【分類號】：TP181
本文編號：2221286
[Abstract]:With the rapid development of Internet, cloud computing, Internet of things and other technologies, the data scale of the network is also growing rapidly, the information society has entered the "big data" era. Based on big data, the system uses machine learning to analyze and mine massive user behavior and push appropriate advertisements to users in real time. The prediction of click rate (Click Through Rate,CTR) is the core technology of the network advertisement delivery system. It is of great significance to improve the operational efficiency of the system. The accurate prediction is the key to making scientific electronic commerce marketing decisions, which directly affects the network experience of users. Directly related to the operating costs of Internet companies. Therefore, the prediction of CTR has high commercial value and research value. In the face of the requirement of high precision and high efficiency in the network advertising system, this paper carries out the research of feature selection, feature learning, classification prediction and application technology from the two angles of shallow learning and deep learning. Taking the real data set of network advertisement as the experimental object, the shallow learning model and the depth learning model are constructed respectively. In order to fully verify the depth learning model, this study verifies the great potential of depth learning through comprehensive comparative experiments from multiple perspectives. Considering synthetically, the concrete research work mainly includes the following five aspects: (1) carry out the research of data processing and feature engineering technology. Based on the real data set, this paper explores the influence mechanism of class imbalance on prediction model, and the resampling technique of unbalanced data. (2) aiming at the highly nonlinear characteristics of data features, To carry out a comparative study of shallow and deep learning theories and applied techniques. In order to overcome the problem of limited learning ability of shallow model for complex problems and to construct a deep learning model, the experiment proves that the learning ability of shallow model is compared with that of shallow learning. The prediction effect of depth learning is improved by about 21%, which has a strong advantage. (3) in order to eliminate the influence of class imbalance on prediction model, an improved model of depth neural network (Deep Neural Network,DNN) is proposed. Based on the parallel computation of GPU, it is verified that the training time of prediction model is reduced by 73.28%, and the efficiency of DNN is greatly improved by constructing model and implementing algorithm. It has been proved that SDNN is a more efficient prediction method for big data in view of the high requirement of accuracy and timeliness of the system. (4) the influence mechanism of Sigmoid activation function and Relu activation function on DNN prediction model is studied. By constructing DNN and SDNN models and algorithms, it is proved that compared with the Sigmoid activation function, Relu activation function is more suitable for the deeper network model. DNN and SDNN based on Relu activation function are more suitable for modeling complex problems. (5) in order to avoid the limitation of single SDNN training to improve the generalization ability of the model, the key parameter dropout sensitivity analysis is carried out.
【學位授予單位】：重慶工商大學
【學位級別】：碩士
【學位授予年份】：2017
【分類號】：TP181
【參考文獻】
相關期刊論文 前10條
1 奚雪峰;周國棟;;面向自然語言處理的深度學習研究[J];自動化學報;2016年10期
2 劉萬軍;梁雪劍;曲海成;;不同池化模型的卷積神經(jīng)網(wǎng)絡學習性能研究[J];中國圖象圖形學報;2016年09期
3 張蕾;章毅;;大數(shù)據(jù)分析的無限深度神經(jīng)網(wǎng)絡方法[J];計算機研究與發(fā)展;2016年01期
4 陳巧紅;余仕敏;賈宇波;;廣告點擊率預估技術綜述[J];浙江理工大學學報;2015年11期
5 朱志北;李斌;劉學軍;胡平;;基于LDA的互聯(lián)網(wǎng)廣告點擊率預測研究[J];計算機應用研究;2016年04期
6 王山海;景新幸;楊海燕;;基于深度學習神經(jīng)網(wǎng)絡的孤立詞語音識別的研究[J];計算機應用研究;2015年08期
7 張鵬;黃毅;阮雅端;陳啟美;;基于稀疏特征的交通流視頻檢測算法[J];南京大學學報(自然科學);2015年02期
8 徐培;蔡小路;何文偉;謝易道;;基于深度自編碼網(wǎng)絡的運動目標檢測[J];計算機應用;2014年10期
9 劉建偉;劉媛;羅雄麟;;深度學習研究進展[J];計算機應用研究;2014年07期
10 余凱;賈磊;陳雨強;徐偉;;深度學習的昨天、今天和明天[J];計算機研究與發(fā)展;2013年09期
相關碩士學位論文 前1條
1 霍艷;網(wǎng)絡廣告投放算法的研究[D];東北大學;2013年
，本文編號：2221286
本文鏈接：http://sikaile.net/kejilunwen/zidonghuakongzhilunwen/2221286.html
最近更新
教材專著
