本文選題：個(gè)性化推薦　切入點(diǎn)：協(xié)同過(guò)濾　出處：《華中科技大學(xué)》2016年碩士論文　論文類型：學(xué)位論文

【摘要】：在互聯(lián)網(wǎng)時(shí)代,個(gè)性化推薦系統(tǒng)得到廣泛應(yīng)用。在推薦系統(tǒng)中,推薦算法起著決定性的作用,而協(xié)同過(guò)濾算法為最為常用的一種推薦算法,采用基于用戶的協(xié)同過(guò)濾對(duì)稀疏評(píng)分矩陣進(jìn)行預(yù)填充,然后使用基于項(xiàng)目的協(xié)同過(guò)濾,對(duì)未知評(píng)分進(jìn)行預(yù)測(cè),可以在一定程度上提高預(yù)測(cè)精度。但該算法存在兩個(gè)問(wèn)題:在預(yù)填充階段,由于用戶相似性計(jì)算過(guò)程受到評(píng)分?jǐn)?shù)據(jù)稀疏的影響,導(dǎo)致得到的用戶最近鄰集不夠準(zhǔn)確,那么利用用戶最近鄰集填充得到的稠密數(shù)據(jù)精度較低;在預(yù)測(cè)階段,由于度量項(xiàng)目相似性方式不合理,導(dǎo)致得到的項(xiàng)目最近鄰集不夠準(zhǔn)確,最終導(dǎo)致算法精度下降。為了進(jìn)一步提高該推薦算法的預(yù)測(cè)精度,針對(duì)該算法存在的兩個(gè)問(wèn)題做出優(yōu)化:在預(yù)填充階段,提出了相似性矩陣最優(yōu)化模型,得到表征用戶相似性的矩陣,從而得到更準(zhǔn)確的用戶最近鄰集,該過(guò)程受數(shù)據(jù)稀疏性影響較小,然后根據(jù)所得的用戶最近鄰集對(duì)原始評(píng)分矩陣預(yù)填充,得到精度較高的稠密評(píng)分矩陣;在預(yù)測(cè)階段,將共評(píng)用戶與目標(biāo)用戶之間的相似性考慮在內(nèi),優(yōu)化了項(xiàng)目間相似性的度量方式,得到了更準(zhǔn)確的項(xiàng)目最近鄰集,然后根據(jù)所得的項(xiàng)目最近鄰集在稠密的評(píng)分矩陣基礎(chǔ)上預(yù)測(cè)未知項(xiàng)目的評(píng)分�；谏鲜鰞牲c(diǎn)優(yōu)化策略設(shè)計(jì)優(yōu)化算法,最后利用開源數(shù)據(jù)集進(jìn)行多組對(duì)比試驗(yàn),實(shí)驗(yàn)結(jié)果表明,優(yōu)化后的算法相比優(yōu)化前的算法在評(píng)分預(yù)測(cè)上有更高的預(yù)測(cè)精度,從而驗(yàn)證了優(yōu)化算法的有效性。
[Abstract]:In the Internet era, personalized recommendation system is widely used. In recommendation system, recommendation algorithm plays a decisive role, and collaborative filtering algorithm is one of the most commonly used recommendation algorithms. The sparse score matrix is pre-filled with user-based collaborative filtering, and then the unknown score is predicted by item-based collaborative filtering. The prediction accuracy can be improved to a certain extent, but there are two problems in this algorithm: in the prefill stage, due to the influence of sparse score data on the process of user similarity calculation, the nearest neighbor set is not accurate enough. Then the dense data filled with user nearest neighbor set is of low precision. In the prediction stage, because of the unreasonable method of measuring the similarity of items, the nearest neighbor set of items is not accurate enough. In order to further improve the prediction accuracy of the proposed algorithm, two problems of the algorithm are optimized: in the phase of pre-filling, the optimization model of similarity matrix is proposed. A matrix representing user similarity is obtained, and a more accurate user nearest neighbor set is obtained. The process is less affected by data sparsity, and then the original score matrix is pre-filled according to the user nearest neighbor set. The dense scoring matrix with high accuracy is obtained. In the prediction stage, the similarity between the users and the target users is taken into account, and the measurement method of the similarity between the items is optimized, and a more accurate set of nearest neighbors is obtained. Then, based on the dense score matrix, the evaluation of unknown items is predicted on the basis of the nearest neighbor set. The optimization algorithm is designed based on the above two optimization strategies. Finally, an open source data set is used to carry out multi-group comparative experiments. The experimental results show that, Compared with the algorithm before optimization, the optimized algorithm has higher prediction accuracy, which verifies the effectiveness of the optimization algorithm.
【學(xué)位授予單位】：華中科技大學(xué)
【學(xué)位級(jí)別】：碩士
【學(xué)位授予年份】：2016
【分類號(hào)】：TP391.3

【參考文獻(xiàn)】

相關(guān)期刊論文 前10條

1 Mingming Jiang;Dandan Song;Lejian Liao;Feida Zhu;;A Bayesian Recommender Model for User Rating and Review Profiling[J];Tsinghua Science and Technology;2015年06期

2 季鐸;畢臣;蔡?hào)|風(fēng);;基于類別信息優(yōu)化的潛在語(yǔ)義分析分類技術(shù)[J];中國(guó)科學(xué)技術(shù)大學(xué)學(xué)報(bào);2015年04期

3 劉慧婷;岳可誠(chéng);;可提高多樣性的基于推薦期望的top-N推薦方法[J];計(jì)算機(jī)科學(xué);2014年07期

4 賀桂和;;基于用戶偏好挖掘的電子商務(wù)協(xié)同過(guò)濾推薦算法研究[J];情報(bào)科學(xué);2013年12期

5 范雪莉;馮海泓;原猛;;基于互信息的主成分分析特征選擇算法[J];控制與決策;2013年06期

6 郝立燕;王靖;;基于填充和相似性信任因子的協(xié)同過(guò)濾推薦算法[J];計(jì)算機(jī)應(yīng)用;2013年03期

7 李華;張宇;孫俊華;;基于用戶模糊聚類的協(xié)同過(guò)濾推薦研究[J];計(jì)算機(jī)科學(xué);2012年12期

8 李宇;王宏琦;孫顯;;利用多尺度語(yǔ)義模型的復(fù)雜圖像目標(biāo)自動(dòng)提取方法[J];武漢大學(xué)學(xué)報(bào)(信息科學(xué)版);2012年09期

9 張獻(xiàn)忠;;推薦算法在網(wǎng)上商城中的應(yīng)用研究[J];福建電腦;2012年04期

10 劉慶鵬;陳明銳;;優(yōu)化稀疏數(shù)據(jù)集提高協(xié)同過(guò)濾推薦系統(tǒng)質(zhì)量的方法[J];計(jì)算機(jī)應(yīng)用;2012年04期

相關(guān)碩士學(xué)位論文 前5條

1 羅杰;基于item-user混合協(xié)同過(guò)濾推薦算法研究[D];昆明理工大學(xué);2014年

2 丁卯;基于協(xié)同過(guò)濾的推薦系統(tǒng)研究[D];河北工業(yè)大學(xué);2014年

3 白龍;融合數(shù)據(jù)檢測(cè)與用戶信任的協(xié)同過(guò)濾算法研究[D];燕山大學(xué);2010年

4 羅勝陽(yáng);協(xié)同過(guò)濾技術(shù)及其在電子商務(wù)推薦領(lǐng)域的應(yīng)用研究[D];南京理工大學(xué);2008年

5 陳曉紅;面向電子商務(wù)的個(gè)性化推薦技術(shù)研究[D];揚(yáng)州大學(xué);2007年

，

本文編號(hào)：1584563