本文關(guān)鍵詞： 奇異值分解 支持向量機(jī) K近鄰 協(xié)同過(guò)濾 推薦系統(tǒng)　出處：《太原理工大學(xué)》2017年碩士論文　論文類型：學(xué)位論文

【摘要】：隨著互聯(lián)網(wǎng)2.0時(shí)代的到來(lái),用戶的各類網(wǎng)絡(luò)信息數(shù)據(jù)與日俱增,信息過(guò)載的問(wèn)題日益嚴(yán)重。對(duì)于單個(gè)用戶而言,從紛繁復(fù)雜的網(wǎng)絡(luò)世界中快速捕捉到自己需要的信息越來(lái)越難;對(duì)于產(chǎn)品提供方而言,如何集成所有用戶的信息并迅速地挖掘到用戶的個(gè)人潛在需求,把用戶可能感興趣的產(chǎn)品及時(shí)推送給用戶成為大數(shù)據(jù)時(shí)代下精準(zhǔn)營(yíng)銷的一大技術(shù)難題。個(gè)性化推薦技術(shù)作為解決信息過(guò)載的有效手段和重要工具應(yīng)運(yùn)而生,在電子商務(wù)領(lǐng)域及各類社交媒體平臺(tái)展現(xiàn)出了良好的應(yīng)用前景。其中,協(xié)同過(guò)濾推薦技術(shù)作為應(yīng)用最早也最廣泛的個(gè)性化推薦技術(shù)之一,在實(shí)際應(yīng)用中取得了巨大的成功,但仍然面臨著數(shù)據(jù)稀疏與冷啟動(dòng),可擴(kuò)展性差等制約推薦精度的嚴(yán)峻的問(wèn)題。個(gè)性化推薦發(fā)展到現(xiàn)在,已經(jīng)有大量?jī)?yōu)秀的專家學(xué)者提出了很多不同的算法模型來(lái)解決傳統(tǒng)協(xié)同過(guò)濾的這些缺陷,其中混合推薦算法因其能夠有效緩解傳統(tǒng)協(xié)同過(guò)濾推薦手段單一,推薦效率不高等缺陷而成為推薦算法研究領(lǐng)域的熱門方向,受到了越來(lái)越多的關(guān)注。本文提出的基于奇異值分解與支持向量機(jī)的混合推薦算法對(duì)傳統(tǒng)協(xié)同過(guò)濾算法進(jìn)行了一些相應(yīng)的改進(jìn),主要工作如下:1.針對(duì)推薦系統(tǒng)中用戶-項(xiàng)目評(píng)分?jǐn)?shù)據(jù)的稀疏性問(wèn)題,提出采用矩陣分解技術(shù)降維來(lái)最大化提取有效信息,分解得到三個(gè)稠密的包含用戶對(duì)項(xiàng)目偏好信息的奇異矩陣,有效地緩解了原始評(píng)分矩陣的極端稀疏情況;2.針對(duì)推薦系統(tǒng)中用戶及項(xiàng)目數(shù)量急劇增長(zhǎng)引發(fā)的可擴(kuò)展性差的問(wèn)題,利用奇異值分解技術(shù)抽取用戶-項(xiàng)目數(shù)據(jù)的關(guān)鍵特征,降低用戶或項(xiàng)目的奇異向量維數(shù),相比傳統(tǒng)協(xié)同過(guò)濾一定程度上降低了相似度矩陣的計(jì)算復(fù)雜度,較好地解決了可擴(kuò)展性差的問(wèn)題;3.為了避免推薦系統(tǒng)用戶及項(xiàng)目數(shù)量龐大導(dǎo)致的內(nèi)存損耗問(wèn)題,提出基于SVD及SVM的混合推薦算法,只需存儲(chǔ)奇異值分解后的用戶或者項(xiàng)目的奇異矩陣,用戶或項(xiàng)目的特征向量維數(shù)大大降低,保證了推薦精確度的同時(shí),節(jié)省了更多存儲(chǔ)空間,這對(duì)于擁有浩如煙海數(shù)據(jù)的推薦系統(tǒng)無(wú)疑具有十分重大的意義;4.在Movie Lens數(shù)據(jù)集上進(jìn)行的實(shí)證表明,本文提出的基于奇異值分解和SVM的混合推薦算法確實(shí)一定程度上緩解了數(shù)據(jù)稀疏,可擴(kuò)展性差及推薦精度不高的問(wèn)題。
[Abstract]:With the arrival of the Internet 2.0 era, users of all kinds of network information data is increasing, the problem of information overload is becoming more and more serious. It is more and more difficult to quickly capture the information we need from the complicated network world. For product providers, how to integrate the information of all users and quickly tap into the potential needs of users. Pushing products of interest to users in time has become a major technical problem of precision marketing in big data era. Personalized recommendation technology as an effective means to solve information overload and an important tool emerged as the times require. Collaborative filtering recommendation technology is one of the earliest and most widely used personalized recommendation technologies in the field of electronic commerce and various social media platforms. Great success has been achieved in the practical application, but still facing the data sparse and cold start, poor scalability and other severe problems restricting the accuracy of the recommendation. Personalized recommendation has developed to the present. A large number of excellent experts and scholars have proposed a lot of different algorithm models to solve these shortcomings of traditional collaborative filtering, among which the hybrid recommendation algorithm can effectively alleviate the traditional collaborative filtering recommendation means single. Recommendation efficiency and other shortcomings have become a hot research area in the field of recommendation algorithm. The hybrid recommendation algorithm based on singular value decomposition (SVD) and support vector machine (SVM) is proposed to improve the traditional collaborative filtering algorithm. The main work is as follows: 1. Aiming at the sparsity of user-item scoring data in recommendation system, a matrix decomposition technique is proposed to maximize the extraction of effective information. Three dense singular matrices containing user preference information are obtained, which can effectively reduce the extreme sparsity of the original scoring matrix. 2. Aiming at the problem of poor scalability caused by the rapid growth of users and projects in recommendation systems, singular value decomposition (SVD) is used to extract the key features of user-project data. To reduce the singular vector dimension of users or items, compared with traditional collaborative filtering, the computational complexity of similarity matrix is reduced to a certain extent, and the problem of poor scalability is solved. 3. A hybrid recommendation algorithm based on SVD and SVM is proposed to avoid the memory loss caused by the large number of users and items in the recommendation system. It only needs to store singular matrix of user or item after singular value decomposition. The dimension of eigenvector of user or item is greatly reduced, which ensures the accuracy of recommendation and saves more storage space. This is undoubtedly of great significance to the recommendation system with vast data; 4. The empirical results on the Movie Lens dataset show that the proposed hybrid recommendation algorithm based on singular value decomposition and SVM can alleviate the data sparsity to some extent. Poor scalability and recommendation accuracy is not high problems.
【學(xué)位授予單位】：太原理工大學(xué)
【學(xué)位級(jí)別】：碩士
【學(xué)位授予年份】：2017
【分類號(hào)】：J943

【參考文獻(xiàn)】

相關(guān)期刊論文 前10條

1 洪麗平;李慶華;楊燕梅;關(guān)慶娟;;數(shù)據(jù)挖掘技術(shù)下圖書(shū)館個(gè)性化服務(wù)探析[J];福建電腦;2017年04期

2 高倩;何聚厚;;改進(jìn)的面向數(shù)據(jù)稀疏的協(xié)同過(guò)濾推薦算法[J];計(jì)算機(jī)技術(shù)與發(fā)展;2016年03期

3 蘇凡軍;唐啟桂;;SBHCF:基于奇異值分解的混合協(xié)同過(guò)濾推薦算法[J];電子科技;2016年01期

4 劉銘;吳沖;劉遠(yuǎn)超;孫承杰;;基于特征權(quán)重量化的相似度計(jì)算方法[J];計(jì)算機(jī)學(xué)報(bào);2015年07期

5 朱揚(yáng)勇;孫婧;;推薦系統(tǒng)研究進(jìn)展[J];計(jì)算機(jī)科學(xué)與探索;2015年05期

6 艾磊;趙輝;;基于知識(shí)的推薦系統(tǒng)用戶交互模型研究[J];軟件導(dǎo)刊;2015年03期

7 劉涵;梁莉莉;黃令帥;;基于分塊奇異值分解的兩級(jí)圖像去噪算法[J];自動(dòng)化學(xué)報(bào);2015年02期

8 李文海;許舒人;;基于Hadoop的電子商務(wù)推薦系統(tǒng)的設(shè)計(jì)與實(shí)現(xiàn)[J];計(jì)算機(jī)工程與設(shè)計(jì);2014年01期

9 曹洪江;傅魁;;協(xié)同過(guò)濾推薦系統(tǒng)中聚類搜索方法研究[J];計(jì)算機(jī)工程與應(yīng)用;2014年05期

10 張倩;;基于支持向量機(jī)的新經(jīng)濟(jì)增長(zhǎng)點(diǎn)選擇研究[J];統(tǒng)計(jì)與信息論壇;2012年06期

，

本文編號(hào)：1465808