本文選題：展示廣告 + 實(shí)時(shí)競(jìng)價(jià)��； 參考：《華東師范大學(xué)》2017年碩士論文

【摘要】：近年來,網(wǎng)絡(luò)流量急劇上升,將網(wǎng)頁上的部分資源當(dāng)作廣告位出售已經(jīng)成為越來越多的媒體進(jìn)行流量變現(xiàn)的一個(gè)重要手段。計(jì)算廣告學(xué)研究的焦點(diǎn)問題是為一組用戶與網(wǎng)頁上下文環(huán)境的組合,找到與之最匹配的廣告。精準(zhǔn)的廣告投放對(duì)用戶、媒體和廣告主均有利,作為一種新型的展示廣告投放模式,實(shí)時(shí)競(jìng)價(jià)的出現(xiàn)推動(dòng)了廣告位由線下定價(jià)到線上售賣模式的轉(zhuǎn)變,改變了廣告市場(chǎng)的格局,也極大地拓展了計(jì)算廣告學(xué)的研究領(lǐng)域,實(shí)時(shí)競(jìng)價(jià)算法研究也因此受到了學(xué)術(shù)界和工業(yè)界的廣泛關(guān)注。本文將實(shí)時(shí)競(jìng)價(jià)算法研究劃分為兩大關(guān)鍵問題:點(diǎn)擊率預(yù)測(cè)和競(jìng)價(jià)策略的設(shè)計(jì)。一方面,點(diǎn)擊率預(yù)估關(guān)系到媒體、廣告主和用戶三方的利益;另一方面廣告主需要參考點(diǎn)擊率來制定合理的競(jìng)價(jià)策略。然而,廣告歷史日志本身存在嚴(yán)重的數(shù)據(jù)稀疏性,傳統(tǒng)機(jī)器學(xué)習(xí)方法構(gòu)建的預(yù)測(cè)模型難以達(dá)到較高的準(zhǔn)確率。本文抓住廣告投放是面向用戶的商業(yè)活動(dòng)這一重要特征,提出了一種基于用戶相似度和特征分化的點(diǎn)擊率預(yù)估組合模型。該模型首先分析了用戶歷史行為特征的相似性并據(jù)此將其劃分為不同子集,接著訓(xùn)練各子集對(duì)應(yīng)的分類子模型,對(duì)于所需預(yù)測(cè)用戶、廣告、媒體的組合,首先模型需要評(píng)估用戶與各用戶子集的相似度并將其作為子分類器權(quán)重,然后統(tǒng)計(jì)在各子分類器下的點(diǎn)擊概率,最后通過對(duì)權(quán)重和各子概率的加權(quán)組合確定用戶的點(diǎn)擊率。根據(jù)實(shí)時(shí)競(jìng)價(jià)的運(yùn)行模式,廣告主通過拍賣的方式獲得廣告曝光的機(jī)會(huì)。受廣告主預(yù)算的限制,合理的競(jìng)價(jià)策略直接影響到廣告主的投資回報(bào)。出價(jià)偏高會(huì)導(dǎo)致廣告主預(yù)算消耗過快,出價(jià)太低將無法獲得廣告曝光機(jī)會(huì)。當(dāng)前主流策略研究主要集中在靜態(tài)或持續(xù)反饋模型上,考慮到互聯(lián)網(wǎng)環(huán)境的復(fù)雜性,本文在點(diǎn)擊率預(yù)估模型的基礎(chǔ)上,提出了一種基于概率反饋的動(dòng)態(tài)競(jìng)價(jià)策略。該策略引入偏離率評(píng)估當(dāng)前算法的有效性,此外,針對(duì)需要修正的狀態(tài),我們結(jié)合拍賣反饋信息給出了修正函數(shù)對(duì)其進(jìn)行調(diào)整。最后,本文在真實(shí)數(shù)據(jù)集上對(duì)提出的模型分別進(jìn)行實(shí)驗(yàn),并與目前主流方法進(jìn)行了詳細(xì)的對(duì)比分析。實(shí)驗(yàn)結(jié)果表明,本文提出的點(diǎn)擊率預(yù)估模型在Logloss、PR曲線均有突出的性能表現(xiàn)并且相對(duì)AUC值最優(yōu)情況下提升了約5%。此外,通過分析各子集的特征權(quán)重可以證明該模型能夠挖掘特征對(duì)不同群體的差異性影響。在競(jìng)價(jià)實(shí)驗(yàn)中,對(duì)比各模型下廣告主的KPI和消耗可得,本文提出的競(jìng)價(jià)策略在廣告主預(yù)算受限的情況下可以提高廣告主的投資回報(bào)率,且平均提升在三倍左右;從預(yù)算消耗趨勢(shì)上看,該策略與市場(chǎng)真實(shí)消耗情況誤差最小,并與其保持相同的消耗趨勢(shì)。
[Abstract]:In recent years, network traffic has risen sharply, and it has become an important means for more and more media to sell some resources on web pages as advertising space. The focus of computational advertising is to find the most appropriate advertisement for a group of users and web context. Accurate advertising delivery is beneficial to users, media and advertisers. As a new display advertising mode, the emergence of real-time bidding has promoted the transformation of advertising positions from offline pricing to online selling, and has changed the pattern of the advertising market. The research field of computational advertising has also been greatly expanded, and the research of real-time bidding algorithm has been widely concerned by academia and industry. In this paper, the real-time bidding algorithm is divided into two key issues: click rate prediction and bidding strategy design. On the one hand, the prediction of click rate is related to the interests of the media, advertisers and users; on the other hand, advertisers need to refer to the click rate to formulate a reasonable bidding strategy. However, there is serious data sparsity in advertising history log itself, and the prediction model constructed by traditional machine learning method is difficult to achieve high accuracy. This paper takes advantage of the important feature that advertising is a user-oriented business activity, and proposes a combination model of click rate prediction based on user similarity and feature differentiation. The model firstly analyzes the similarity of users' historical behavior characteristics and divides them into different subsets, and then trains the corresponding submodels of each subset to predict the combination of users, advertisements and media. First of all, the model needs to evaluate the similarity between users and subsets of users and take them as sub-classifier weights, then count the click probability under each sub-classifier, and finally determine the click rate of users by the weighted combination of the weights and the sub-probability. According to the real-time bidding mode, advertisers get exposure by auction. Due to the limitation of advertisers' budget, reasonable bidding strategy directly affects the advertisers' return on investment. High bids lead advertisers to spend too fast on their budgets and too low bids to get exposure. The current mainstream strategy research is mainly focused on the static or continuous feedback model. Considering the complexity of the Internet environment, this paper proposes a dynamic bidding strategy based on probability feedback based on the click rate prediction model. This strategy introduces the deviation rate to evaluate the effectiveness of the current algorithm. In addition, for the states that need to be modified, we propose a modified function to adjust the proposed algorithm in combination with the auction feedback information. Finally, the proposed model is experimented on the real data set and compared with the current mainstream methods in detail. The experimental results show that the model presented in this paper has outstanding performance in the Loglosser PR curve and is improved by about 5% relative to the optimal AUC value. In addition, by analyzing the feature weights of each subset, it can be proved that the model can mine the influence of feature on the difference of different groups. In the bidding experiment, comparing the KPI and consumption of the advertisers under each model, the bidding strategy proposed in this paper can improve the advertisers' return on investment under the condition of limited budget, and the average increase is about three times; From the point of view of the trend of budget consumption, the error between the strategy and the real consumption of the market is minimum, and the same consumption trend is maintained.
【學(xué)位授予單位】：華東師范大學(xué)
【學(xué)位級(jí)別】：碩士
【學(xué)位授予年份】：2017
【分類號(hào)】：F713.8

【參考文獻(xiàn)】

相關(guān)期刊論文 前1條

1 紀(jì)文迪;王曉玲;周傲英;;廣告點(diǎn)擊率估算技術(shù)綜述[J];華東師范大學(xué)學(xué)報(bào)(自然科學(xué)版);2013年03期

相關(guān)碩士學(xué)位論文 前1條

1 王孝舒;廣告點(diǎn)擊率預(yù)估的深層神經(jīng)網(wǎng)絡(luò)模型研究[D];北京郵電大學(xué);2015年

，

本文編號(hào)：2057327