本文關(guān)鍵詞： 推薦系統(tǒng) 協(xié)同過(guò)濾 矩陣分解 層次分類 內(nèi)容關(guān)聯(lián) 遷移學(xué)習(xí)　出處：《北京交通大學(xué)》2016年博士論文　論文類型：學(xué)位論文

【摘要】：隨著計(jì)算機(jī)的普及和網(wǎng)絡(luò)技術(shù)的發(fā)展,互聯(lián)網(wǎng)信息服務(wù)已經(jīng)逐漸滲透到人們生活的方方面面,正在從根本上改變?nèi)藗儌鹘y(tǒng)的生活方式。特別是近年來(lái),智能手機(jī)、平板電腦等移動(dòng)設(shè)備的廣泛使用以及微信、微博等移動(dòng)應(yīng)用的興起,突破了傳統(tǒng)PC端互聯(lián)網(wǎng)訪問(wèn)的時(shí)間、空間等限制,使得人們現(xiàn)在可以更加方便、自由、快捷地通過(guò)互聯(lián)網(wǎng)獲取和分享信息。然而,伴隨著互聯(lián)網(wǎng)信息服務(wù)的蓬勃發(fā)展,其信息資源規(guī)模也發(fā)生了爆發(fā)式增長(zhǎng)。此時(shí),人們從互聯(lián)網(wǎng)中找到自己想要的信息變的愈發(fā)困難,引起了所謂的信息過(guò)載問(wèn)題。在此背景下,推薦系統(tǒng)被提出并且成為解決該問(wèn)題最有效的技術(shù)之一。目前,協(xié)同過(guò)濾是推薦系統(tǒng)中應(yīng)用最廣泛、最成功的技術(shù)。它僅需少量“用戶-物品”之間的歷史評(píng)分?jǐn)?shù)據(jù)就可以快速構(gòu)建一個(gè)可用的系統(tǒng)來(lái)預(yù)測(cè)用戶的潛在信息需求,具有簡(jiǎn)單、易用、精度高等優(yōu)點(diǎn)。然而,隨著數(shù)據(jù)規(guī)模越來(lái)越龐大、數(shù)據(jù)類型越來(lái)越豐富、應(yīng)用環(huán)境越來(lái)越復(fù)雜,傳統(tǒng)協(xié)同過(guò)濾算法正面臨更加嚴(yán)峻的數(shù)據(jù)稀疏性、冷啟動(dòng)、可擴(kuò)展性、可解釋性等問(wèn)題。最近,一些研究工作嘗試把上下文信息融合到協(xié)同過(guò)濾算法,取得了一定的性能提升。從這些初步嘗試可以看出,上下文信息與用戶興趣有緊密聯(lián)系,它們的引入有助于提高預(yù)測(cè)精度和用戶滿意度,因此融合上下文信息對(duì)于改進(jìn)協(xié)同過(guò)濾算法具有重要意義。鑒于此,本文對(duì)協(xié)同過(guò)濾算法進(jìn)行了系統(tǒng)分析,對(duì)上下文信息進(jìn)行了更加深入的探討,進(jìn)而針對(duì)不同上下文的歷史評(píng)分?jǐn)?shù)據(jù),設(shè)計(jì)了多種混合協(xié)同過(guò)濾算法能夠更高效地利用上下文信息解決當(dāng)前推薦系統(tǒng)面臨的問(wèn)題。本文主要工作和創(chuàng)新如下：1.融合物品分類結(jié)構(gòu)和內(nèi)容信息的協(xié)同過(guò)濾算法研究。目前,大部分關(guān)于可擴(kuò)展性和冷啟動(dòng)問(wèn)題的研究主要針對(duì)用戶進(jìn)行展開,而很少關(guān)注系統(tǒng)中動(dòng)態(tài)更新的物品,尤其對(duì)大規(guī)模物品缺乏可擴(kuò)展性,對(duì)新物品也不能取得令人滿意的推薦結(jié)果。本研究發(fā)現(xiàn),在有明確物品分類的前提下,同種物品之間一定會(huì)存在一些相同的內(nèi)容屬性或者其他一些潛在特征,因此用戶對(duì)同種物品應(yīng)該具有相似興趣。基于此發(fā)現(xiàn),本研究從物品關(guān)系以及物品特征入手,利用物品分類信息、物品內(nèi)容信息(關(guān)鍵字)等上下文提出一種逐步優(yōu)化用戶興趣的分層協(xié)同過(guò)濾算法。分析顯示該算法對(duì)大規(guī)模物品有可擴(kuò)展性,還能解決新物品的冷啟動(dòng)問(wèn)題,并且真實(shí)數(shù)據(jù)集上的實(shí)驗(yàn)結(jié)果表明該算法在不同比例稀疏數(shù)據(jù)情況下可以取得較高的預(yù)測(cè)精度,而且針對(duì)新物品具有較好的冷啟動(dòng)預(yù)測(cè)能力。2.融合用戶-物品內(nèi)容上下文關(guān)聯(lián)信息的協(xié)同過(guò)濾算法研究。在之前的算法中,雖然物品分類信息有助于利用物品相似性優(yōu)化用戶興趣,但是分類需要事先構(gòu)建,這種較高的數(shù)據(jù)要求限制了該算法的適用范圍,另外該算法不能對(duì)用戶進(jìn)行擴(kuò)展,也不能解決新用戶的冷啟動(dòng)問(wèn)題。為了設(shè)計(jì)更通用可擴(kuò)展的算法,本研究轉(zhuǎn)而關(guān)注內(nèi)容上下文,也就是用戶內(nèi)容信息(標(biāo)簽)和物品內(nèi)容信息(關(guān)鍵字)。用戶-物品之間的歷史評(píng)分?jǐn)?shù)據(jù)為它們的內(nèi)容上下文建立了關(guān)聯(lián)關(guān)系。基于此發(fā)現(xiàn),本研究從內(nèi)容上下文入手,將協(xié)同過(guò)濾與基于內(nèi)容的推薦算法相結(jié)合,提出一種根據(jù)內(nèi)容相似性產(chǎn)生預(yù)測(cè)結(jié)果的間接協(xié)同過(guò)濾算法。分析顯示該算法具有較強(qiáng)的可解釋性和可擴(kuò)展性,并且真實(shí)數(shù)據(jù)集上的實(shí)驗(yàn)結(jié)果表明該算法在不同比例稀疏數(shù)據(jù)情況下可以取得較高的預(yù)測(cè)精度,而且針對(duì)新用戶和新物品都具有較好的冷啟動(dòng)預(yù)測(cè)能力。3.融合子群組間潛在共享信息的協(xié)同過(guò)濾算法研究。除了直接將上下文信息與推薦算法進(jìn)行耦合外,最近出現(xiàn)了一類基于子群組的改進(jìn)算法,主要思想是根據(jù)上下文信息,將整個(gè)數(shù)據(jù)集劃分到不同子群組,然后在這些子群組上分別運(yùn)行協(xié)同過(guò)濾算法產(chǎn)生各自的預(yù)測(cè)結(jié)果。但是不均衡稀疏數(shù)據(jù)會(huì)造成子群組上協(xié)同過(guò)濾結(jié)果不穩(wěn)定的問(wèn)題。對(duì)這些子群組分析后,可以發(fā)現(xiàn)它們所包含的用戶和物品之間存在隱含聯(lián)系。基于此發(fā)現(xiàn),本研究從子群組間潛在共享信息入手,提出一種基于知識(shí)遷移的跨群組協(xié)同過(guò)濾算法,它利用少數(shù)性能較好子群組上的協(xié)同過(guò)濾結(jié)果構(gòu)建評(píng)分矩陣的多個(gè)近似,然后加權(quán)聚合這些近似產(chǎn)生預(yù)測(cè)結(jié)果。分析顯示該算法減少了一些性能較差子群組上的不必要計(jì)算,而且真實(shí)數(shù)據(jù)集上的實(shí)驗(yàn)結(jié)果表明該算法提高了預(yù)測(cè)精度,尤其是在非常稀疏數(shù)據(jù)上其性能提升尤為明顯,說(shuō)明該算法緩解了數(shù)據(jù)稀疏性問(wèn)題。
[Abstract]:With the development of computer and network technology, the Internet information service has gradually penetrated into all aspects of people's lives, is fundamentally changing people's traditional way of life. Especially in recent years, intelligent mobile phone, tablet computer and other mobile devices are widely used as well as WeChat, micro-blog and other emerging mobile applications, through access the traditional PC Internet time, space constraints, so that people can now be more convenient, free, fast access and share information through the Internet. However, with the rapid development of Internet information service, the information resource scale has undergone explosive growth. At this time, people from the Internet to find the information they want to change more difficult, cause the information overload problem. In this context, recommendation systems have been proposed to solve the problem of technology and become the most effective operation at present, Collaborative filtering is the most widely used recommendation system, the most successful technology. It only need a small amount of "user item" between the historical rating data can quickly build an available system to predict the user's potential information demand, has the advantages of simple, easy to use, high precision. However, with the increasing scale of data big data types are more and more abundant, the application environment is more and more complex, the traditional collaborative filtering algorithm is facing more severe data sparsity, cold start, scalability and interpretability. Recently, some studies try to put context information into collaborative filtering algorithm, has made a big performance improvement from these. A preliminary attempt can be seen, is closely related to context and user interest, which is helpful to improve the prediction accuracy and user satisfaction, so the fusion of context information to improve collaboration The filter algorithm has important significance. In view of this, this paper makes a systematic analysis on the collaborative filtering algorithm, the context information is further discussed, according to the historical data of different context score, design a variety of hybrid collaborative filtering algorithm can be more efficient to solve the current location recommendation system problems with context information. The main work of this paper and the innovation is as follows: collaborative filtering algorithm of 1. fusion category structure and content information. At present, most research on scalability and cold start problem is mainly for users to start, and pay little attention to goods dynamic updating of the system, especially the lack of scalability for large items of new items can not get the recommended results satisfactory. This study found that, in the premise of a clear classification of goods, the same goods between certain there will be some of the same. The attributes of the content or some other potential features, so that users of the same items should have the same interest. Based on these findings, this study from the relationship between items and items of the goods classification information, goods information (key) context we propose a hierarchical user interest gradually optimize the collaborative filtering algorithm. The analysis shows that the algorithm can extended to large items, but also resolves the problem of the cold start of new items, and the experimental results on real datasets show that the algorithm can achieve higher prediction accuracy in different proportion of sparse data, and according to the new cold start items with better prediction ability of collaborative filtering algorithm.2. fusion user context information items. Before the algorithm, although the classification of items of information to help with the similarity optimization of user interest, but is classified Prior to construction, the high data requirements limit the application range of the algorithm, the algorithm cannot the user expansion, can not solve the problem of the cold start of new users. In order to design a more general scalable algorithm, this study focus on the context, which is the user information and content information items (Tags) (key words). The history of user item rating data establish the relationship between their context. Based on these findings, this study from the context of collaborative filtering and recommendation algorithm based on the content of the combination is proposed based on content similarity prediction results. Analysis showed that the indirect collaborative filtering algorithm the algorithm has strong interpretability and scalability, and the experimental results on real datasets show that the algorithm can be taken in different proportion under the condition of sparse data Have higher prediction accuracy, but also for new users and new items have better prediction ability of cold start.3. fusion collaborative filtering algorithm of information sharing between potential sub group. In addition to direct context information and recommendation algorithm coupling, the recent emergence of a kind of improved algorithm based on sub group, the main idea is based on the the context information of the entire data set is divided into different sub groups, then the collaborative filtering algorithm to produce predictive results of their operation in these sub groups. But the imbalance of sparse data will cause the sub group on collaborative filtering unstable results. Analysis of these sub groups, there can be hidden connection between users and items found they contain. Based on these findings, this study from the group of potential information sharing, proposes a cross group collaborative filtering is based on knowledge transfer Method, it uses collaborative filtering results to construct multiple approximate score matrix a better performance on the sub group, and then weighted aggregation these approximations yield prediction results. The analysis shows that the algorithm reduces the number of sub groups of the poor performance of unnecessary calculation, and experimental results on real datasets show that the algorithm improves the prediction accuracy. Especially in very sparse data on its performance is particularly evident, indicating that the algorithm alleviates the problem of data sparsity.

【學(xué)位授予單位】：北京交通大學(xué)
【學(xué)位級(jí)別】：博士
【學(xué)位授予年份】：2016
【分類號(hào)】：TP391.3

【相似文獻(xiàn)】

相關(guān)期刊論文 前10條

1 楊風(fēng)召;;一種基于特征表的協(xié)同過(guò)濾算法[J];計(jì)算機(jī)工程與應(yīng)用;2007年06期

2 王嵐;翟正軍;;基于時(shí)間加權(quán)的協(xié)同過(guò)濾算法[J];計(jì)算機(jī)應(yīng)用;2007年09期

3 曾子明;張李義;;基于多屬性決策和協(xié)同過(guò)濾的智能導(dǎo)購(gòu)系統(tǒng)[J];武漢大學(xué)學(xué)報(bào)(工學(xué)版);2008年02期

4 張富國(guó);;用戶多興趣下基于信任的協(xié)同過(guò)濾算法研究[J];小型微型計(jì)算機(jī)系統(tǒng);2008年08期

5 侯翠琴;焦李成;張文革;;一種壓縮稀疏用戶評(píng)分矩陣的協(xié)同過(guò)濾算法[J];西安電子科技大學(xué)學(xué)報(bào);2009年04期

6 廖新考;;基于用戶特征和項(xiàng)目屬性的混合協(xié)同過(guò)濾推薦[J];福建電腦;2010年07期

7 沈磊;周一民;李舟軍;;基于心理學(xué)模型的協(xié)同過(guò)濾推薦方法[J];計(jì)算機(jī)工程;2010年20期

8 徐紅;彭黎;郭艾寅;徐云劍;;基于用戶多興趣的協(xié)同過(guò)濾策略改進(jìn)研究[J];計(jì)算機(jī)技術(shù)與發(fā)展;2011年04期

9 焦晨斌;王世卿;;基于模型填充的混合協(xié)同過(guò)濾算法[J];微計(jì)算機(jī)信息;2011年11期

10 鄭婕;鮑海琴;;基于協(xié)同過(guò)濾推薦技術(shù)的個(gè)性化網(wǎng)絡(luò)教學(xué)平臺(tái)研究[J];科技風(fēng);2012年06期

相關(guān)會(huì)議論文 前10條

1 沈杰峰;杜亞軍;唐俊;;一種基于項(xiàng)目分類的協(xié)同過(guò)濾算法[A];第二十二屆中國(guó)數(shù)據(jù)庫(kù)學(xué)術(shù)會(huì)議論文集（技術(shù)報(bào)告篇）[C];2005年

2 周軍鋒;湯顯;郭景峰;;一種優(yōu)化的協(xié)同過(guò)濾推薦算法[A];第二十一屆中國(guó)數(shù)據(jù)庫(kù)學(xué)術(shù)會(huì)議論文集（研究報(bào)告篇）[C];2004年

3 董全德;;基于雙信息源的協(xié)同過(guò)濾算法研究[A];全國(guó)第20屆計(jì)算機(jī)技術(shù)與應(yīng)用學(xué)術(shù)會(huì)議（CACIS·2009）暨全國(guó)第1屆安全關(guān)鍵技術(shù)與應(yīng)用學(xué)術(shù)會(huì)議論文集（上冊(cè)）[C];2009年

4 張光衛(wèi);康建初;李鶴松;劉常昱;李德毅;;面向場(chǎng)景的協(xié)同過(guò)濾推薦算法[A];中國(guó)系統(tǒng)仿真學(xué)會(huì)第五次全國(guó)會(huì)員代表大會(huì)暨2006年全國(guó)學(xué)術(shù)年會(huì)論文集[C];2006年

5 李建國(guó);姚良超;湯庸;郭歡;;基于認(rèn)知度的協(xié)同過(guò)濾推薦算法[A];第26屆中國(guó)數(shù)據(jù)庫(kù)學(xué)術(shù)會(huì)議論文集（B輯）[C];2009年

6 王明文;陶紅亮;熊小勇;;雙向聚類迭代的協(xié)同過(guò)濾推薦算法[A];第三屆全國(guó)信息檢索與內(nèi)容安全學(xué)術(shù)會(huì)議論文集[C];2007年

7 胡必云;李舟軍;王君;;基于心理測(cè)量學(xué)的協(xié)同過(guò)濾相似度方法(英文)[A];NDBC2010第27屆中國(guó)數(shù)據(jù)庫(kù)學(xué)術(shù)會(huì)議論文集(B輯)[C];2010年

8 林麗冰;師瑞峰;周一民;李月雷;;基于雙聚類的協(xié)同過(guò)濾推薦算法[A];2008'中國(guó)信息技術(shù)與應(yīng)用學(xué)術(shù)論壇論文集（一）[C];2008年

9 羅喜軍;王韜丞;杜小勇;劉紅巖;何軍;;基于類別的推薦——一種解決協(xié)同推薦中冷啟動(dòng)問(wèn)題的方法[A];第二十四屆中國(guó)數(shù)據(jù)庫(kù)學(xué)術(shù)會(huì)議論文集（研究報(bào)告篇）[C];2007年

10 黃創(chuàng)光;印鑒;汪靜;劉玉葆;王甲海;;不確定近鄰的協(xié)同過(guò)濾推薦算法[A];NDBC2010第27屆中國(guó)數(shù)據(jù)庫(kù)學(xué)術(shù)會(huì)議論文集A輯一[C];2010年

相關(guān)博士學(xué)位論文 前10條

1 紀(jì)科;融合上下文信息的混合協(xié)同過(guò)濾推薦算法研究[D];北京交通大學(xué);2016年

2 李聰;電子商務(wù)推薦系統(tǒng)中協(xié)同過(guò)濾瓶頸問(wèn)題研究[D];合肥工業(yè)大學(xué);2009年

3 郭艷紅;推薦系統(tǒng)的協(xié)同過(guò)濾算法與應(yīng)用研究[D];大連理工大學(xué);2008年

4 羅恒;基于協(xié)同過(guò)濾視角的受限玻爾茲曼機(jī)研究[D];上海交通大學(xué);2011年

5 薛福亮;電子商務(wù)協(xié)同過(guò)濾推薦質(zhì)量影響因素及其改進(jìn)機(jī)制研究[D];天津大學(xué);2012年

6 高e，

本文編號(hào)：1442363