基于排序?qū)W習和卷積神經(jīng)網(wǎng)絡(luò)的推薦算法研究
發(fā)布時間:2018-06-03 14:42
本文選題:推薦系統(tǒng) + 社交網(wǎng)絡(luò) ; 參考:《大連理工大學》2016年碩士論文
【摘要】:隨著互聯(lián)網(wǎng)技術(shù)特別是以淘寶和亞馬遜等為代表的電子商務(wù)的飛速發(fā)展,互聯(lián)網(wǎng)中的數(shù)據(jù)呈現(xiàn)爆炸性增長,信息過載問題顯得越來越嚴重。幫助我們從海量數(shù)據(jù)中篩選出有意義數(shù)據(jù)的信息過濾技術(shù)顯得越來越重要。在此背景下,推薦系統(tǒng)誕生了,并且迅速發(fā)展成為當前互聯(lián)網(wǎng)應用中的重要組成部分。推薦系統(tǒng)根據(jù)用戶行為記錄從大規(guī)模數(shù)據(jù)中找到用戶感興趣商品,它對于提高用戶的滿意度和零售商的銷售額具有重要的意義。用戶在互聯(lián)網(wǎng)中的行為主要分為兩類,分別是隱性反饋行為和顯性反饋行為。其中在隱性反饋行為中用戶沒有顯式地表達對特定商品的偏好,主要包括用戶的點擊、瀏覽、收藏等行為;而在顯性反饋行為中用戶則顯式地表達了對特定商品的偏好信息,這些行為中較為常見的主要有評分行為。針對不同類型的用戶反饋行為數(shù)據(jù)有不同的推薦方法,本文對兩種不同的用戶反饋行為進行了細致地分析和挖掘,并且分別有針對性地提出了兩種方法以提高推薦系統(tǒng)的性能。針對顯性反饋行為的評分行為,本文選取Top-K推薦作為研究目標。引入信息檢索領(lǐng)域排序?qū)W習的方法并且融合用戶的社交信息和商品標簽信息,本文擴展了一種基于列表排序?qū)W習的矩陣分解方法,一方面充分考慮用戶之間關(guān)注關(guān)系。首先通過用戶之間的關(guān)注關(guān)系計算用戶之間的信任度,接著通過用戶之間的信任度在原始模型的損失函數(shù)中添加用戶社交約束項,使相互信任的用戶偏好向量盡可能接近。另一方面,計算商品所擁有標簽的權(quán)重并以此計算商品之間的標簽相似度,再將商品的標簽約束項添加至損失函數(shù)中。在真實Epinions和百度電影數(shù)據(jù)集中的實驗結(jié)果表明,我們提出的方法的NDCG值和原始模型相比具有一定的提高,有效地提高了推薦準確率。針對隱性反饋行為,本文選取電子商務(wù)領(lǐng)域的下一個購物籃推薦作為研究目標。本文首先將用戶行為按照一定的時間窗口進行劃分,對于每個窗口從多個不同的維度抽取用戶對商品的時序偏好特征;接著運用深度學習領(lǐng)域的卷積神經(jīng)網(wǎng)絡(luò)模型,模型中的卷積層組合不同長度的特征圖來訓練分類器。在阿里巴巴移動推薦算法競賽公布的真實數(shù)據(jù)集中的實驗結(jié)果表明,和傳統(tǒng)的線性模型和樹模型等分類器相比,我們提出的卷積神經(jīng)網(wǎng)絡(luò)框架具有較強的特征萃取能力和泛化能力,提高了推薦系統(tǒng)的用戶滿意度。
[Abstract]:With the rapid development of Internet technology, especially the electronic commerce, such as Taobao and Amazon, the data in the Internet is growing explosive. The problem of information overload is becoming more and more serious. Information filtering techniques that help us filter meaningful data from mass data are becoming more and more important. In this context, the recommendation system is in the background. The system is born, and has rapidly developed into an important part of the current Internet applications. The recommended system is based on user behavior records to find users interested in goods from large-scale data. It is important to improve the satisfaction of users and the sales of retailers. The behavior of users in the Internet is divided into two categories. There is no implicit feedback behavior and explicit feedback behavior. In the implicit feedback behavior, users do not explicitly express preference for specific goods, including users' click, browse, collection and other behaviors, while in explicit feedback behavior, users express preference information about specific products, which are more common in these behaviors. There are different methods of recommendation for different types of user feedback behavior data. In this paper, two different user feedback behaviors are carefully analyzed and excavated, and two methods are proposed to improve the performance of the recommended system respectively. In this paper, the paper selects Top for the behavior of dominant feedback behavior. -K recommends as a research goal. Introducing the method of sorting learning in the field of information retrieval and integrating the user's social and commodity label information, this paper extends a matrix decomposition method based on list sorting learning. On the one hand, it takes full consideration of the concerns between users. And then the user's social constraints are added to the loss function of the original model through the trust degree between the users, so that the mutual trust user preference vector is as close as possible. On the other hand, the weight of the label is calculated and the label similarity between the goods is calculated, and the label constraint item of the commodity is added to the loss. In the loss function, the experimental results in the real Epinions and Baidu movie datasets show that the NDCG value of the proposed method is improved to a certain extent compared with the original model, which effectively improves the accuracy of the recommendation. In this paper, the next shopping basket in the field of electronic commerce is selected as the research goal. First, the user behavior is divided according to a certain time window, and each window is extracted from a number of different dimensions of the user's timing preference. Then, the convolution neural network model in the depth learning field is used to train the classifier with different length of feature graph to train the classifier. In the Alibaba movement, the classifier is moved and pushed. The experimental results of the true data set published in the recommendation algorithm contest show that, compared with the traditional linear and tree model classes, the convolution neural network framework proposed by us has strong feature extraction ability and generalization ability, and improves the user satisfaction of the recommendation system.
【學位授予單位】:大連理工大學
【學位級別】:碩士
【學位授予年份】:2016
【分類號】:TP391.3;TP183
【參考文獻】
相關(guān)期刊論文 前5條
1 李瑞敏;林鴻飛;閆俊;;基于用戶-標簽-項目語義挖掘的個性化音樂推薦[J];計算機研究與發(fā)展;2014年10期
2 閆俊;劉文飛;林鴻飛;;基于標簽混合語義空間的音樂推薦方法研究[J];中文信息學報;2014年04期
3 張子柯;周濤;張翼成;;Tag-Aware Recommender Systems:A State-of-the-Art Survey[J];Journal of Computer Science & Technology;2011年05期
4 印鑒,陳憶群,張鋼;搜索引擎技術(shù)研究與發(fā)展[J];計算機工程;2005年14期
5 姜靈敏;中國電子商務(wù)發(fā)展現(xiàn)狀與對策研究[J];商業(yè)研究;2003年01期
相關(guān)博士學位論文 前1條
1 鄧愛林;電子商務(wù)推薦系統(tǒng)關(guān)鍵技術(shù)研究[D];復旦大學;2003年
,本文編號:1973155
本文鏈接:http://www.wukwdryxk.cn/kejilunwen/zidonghuakongzhilunwen/1973155.html
最近更新
教材專著