發(fā)布時(shí)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(xué)》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(zhǎng),信息過(guò)載問(wèn)題顯得越來(lái)越嚴(yán)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(guò)濾技術(shù)顯得越來(lái)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(dāng)前互聯(lián)網(wǎng)應(yīng)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(duì)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(méi)有顯式地表達(dá)對(duì)特定商品的偏好,主要包括用戶的點(diǎn)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(dá)了對(duì)特定商品的偏好信息,這些行為中較為常見(jiàn)的主要有評(píng)分行為。針對(duì)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(duì)兩種不同的用戶反饋行為進(jìn)行了細(xì)致地分析和挖掘,并且分別有針對(duì)性地提出了兩種方法以提高推薦系統(tǒng)的性能。針對(duì)顯性反饋行為的評(píng)分行為,本文選取Top-K推薦作為研究目標(biāo)。引入信息檢索領(lǐng)域排序?qū)W習(xí)的方法并且融合用戶的社交信息和商品標(biāo)簽信息,本文擴(kuò)展了一種基于列表排序?qū)W習(xí)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(guò)用戶之間的關(guān)注關(guān)系計(jì)算用戶之間的信任度,接著通過(guò)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(xiàng),使相互信任的用戶偏好向量盡可能接近。另一方面,計(jì)算商品所擁有標(biāo)簽的權(quán)重并以此計(jì)算商品之間的標(biāo)簽相似度,再將商品的標(biāo)簽約束項(xiàng)添加至損失函數(shù)中。在真實(shí)Epinions和百度電影數(shù)據(jù)集中的實(shí)驗(yàn)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(zhǔn)確率。針對(duì)隱性反饋行為,本文選取電子商務(wù)領(lǐng)域的下一個(gè)購(gòu)物籃推薦作為研究目標(biāo)。本文首先將用戶行為按照一定的時(shí)間窗口進(jìn)行劃分,對(duì)于每個(gè)窗口從多個(gè)不同的維度抽取用戶對(duì)商品的時(shí)序偏好特征；接著運(yùn)用深度學(xué)習(xí)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(zhǎng)度的特征圖來(lái)訓(xùn)練分類器。在阿里巴巴移動(dòng)推薦算法競(jìng)賽公布的真實(shí)數(shù)據(jù)集中的實(shí)驗(yàn)結(jié)果表明,和傳統(tǒng)的線性模型和樹(shù)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(qiáng)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(xué)位授予單位】：大連理工大學(xué)
【學(xué)位級(jí)別】：碩士
【學(xué)位授予年份】：2016
【分類號(hào)】：TP391.3;TP183

【參考文獻(xiàn)】

相關(guān)期刊論文 前5條

1 李瑞敏;林鴻飛;閆俊;;基于用戶-標(biāo)簽-項(xiàng)目語(yǔ)義挖掘的個(gè)性化音樂(lè)推薦[J];計(jì)算機(jī)研究與發(fā)展;2014年10期

2 閆俊;劉文飛;林鴻飛;;基于標(biāo)簽混合語(yǔ)義空間的音樂(lè)推薦方法研究[J];中文信息學(xué)報(bào);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(jì)算機(jī)工程;2005年14期

5 姜靈敏;中國(guó)電子商務(wù)發(fā)展現(xiàn)狀與對(duì)策研究[J];商業(yè)研究;2003年01期

相關(guān)博士學(xué)位論文 前1條

1 鄧愛(ài)林;電子商務(wù)推薦系統(tǒng)關(guān)鍵技術(shù)研究[D];復(fù)旦大學(xué);2003年

，

本文編號(hào)：1973155