本文選題：推薦系統(tǒng) + 隱式反饋��； 參考：《吉林大學(xué)》2017年碩士論文

【摘要】：在“互聯(lián)網(wǎng)+”的時(shí)代背景下,個(gè)性化推薦系統(tǒng)為用戶提供“私人訂制”式的推薦服務(wù)以滿足不同用戶的消費(fèi)需求。因個(gè)性化推薦系統(tǒng)具有互動(dòng)性的特點(diǎn),使其成為實(shí)現(xiàn)“互聯(lián)網(wǎng)+”新型消費(fèi)模式的重要手段。然而隨著推薦服務(wù)規(guī)模越來越大,評分?jǐn)?shù)據(jù)不足、用戶-項(xiàng)目矩陣稀疏等問題愈發(fā)凸顯,傳統(tǒng)推薦算法面臨難以突破的瓶頸。為了解決這一問題,不少研究學(xué)者開始更多地關(guān)注用戶隱式行為的分析和研究,嘗試從中挖掘用戶興趣偏好以彌補(bǔ)顯式評分?jǐn)?shù)據(jù)帶來的不足。本文針對圖書推薦系統(tǒng)的評分?jǐn)?shù)據(jù)稀疏和單類協(xié)同過濾等問題展開研究,分析系統(tǒng)中數(shù)據(jù)的特點(diǎn),將隱式反饋?zhàn)鳛榻⒑透掠脩襞d趣模型的數(shù)據(jù)來源,創(chuàng)造性地提出一種改進(jìn)算法——基于分層隱式反饋的貝葉斯排序算法,通過預(yù)測用戶對項(xiàng)目的相對喜好來得到推薦列表。本文的研究工作可概括如下:首先,文章從系統(tǒng)中存在的顯式和隱式反饋數(shù)據(jù)的特點(diǎn)與區(qū)別出發(fā),分析本系統(tǒng)選取隱式用戶反饋的原因以及隱式反饋帶來的問題。然后從基于評分和基于排序兩個(gè)方面介紹了幾種經(jīng)典的推薦算法。針對當(dāng)下排序推薦算法往往只考慮用戶有過操作行為的項(xiàng)目與未操作過的項(xiàng)目之間的偏好差別,缺乏對用戶偏好程度深度理解和區(qū)分,本文探索用戶隱式反饋所表示的用戶偏好,將隱式反饋類型按所代表用戶興趣程度進(jìn)行劃分,使用分層隱式反饋模型來表示用戶偏好。在貝葉斯個(gè)性化排序算法的基礎(chǔ)上設(shè)計(jì)了“強(qiáng)正反饋-弱正反饋-負(fù)反饋”的三層隱式反饋表示模型,為用戶生成符合用戶興趣的項(xiàng)目推薦列表。然后,經(jīng)過大量實(shí)驗(yàn)研究分析潛因子矩陣維度k值大小、推薦列表長度等參數(shù)對算法性能的影響,并從AUC、平均精度均值、平均百分比排序、歸一化折損累積增益等角度與經(jīng)典的排序推薦算法作比較,對改進(jìn)的算法進(jìn)行評估,得出本文算法所生成的推薦列表可以更準(zhǔn)確地把握用戶當(dāng)前興趣。最后,在改進(jìn)算法基礎(chǔ)上,從分析個(gè)性化圖書系統(tǒng)的功能需求入手,對系統(tǒng)進(jìn)行功能模塊劃分與設(shè)計(jì),實(shí)現(xiàn)了基于隱式反饋的個(gè)性化圖書推薦系統(tǒng)。
[Abstract]:Under the background of "Internet", personalized recommendation system provides users with "private customized" recommendation services to meet the consumer needs of different users. Because of the interactive characteristic of personalized recommendation system, it becomes an important means to realize the new consumption mode of "Internet". However, with the increasing scale of recommendation service, insufficient score data and sparse user-item matrix, the traditional recommendation algorithm is facing the bottleneck that is difficult to break through. In order to solve this problem, many researchers begin to pay more attention to the analysis and research of implicit behavior of users, and try to mine user interest preference from it in order to make up for the deficiency caused by explicit rating data. In this paper, the sparse scoring data and single class collaborative filtering of book recommendation system are studied, the characteristics of the data in the system are analyzed, and the implicit feedback is used as the data source to establish and update the user interest model. An improved algorithm, Bayesian sorting algorithm based on hierarchical implicit feedback, is proposed in this paper. The list of recommendations is obtained by predicting the relative preferences of users to items. The research work of this paper can be summarized as follows: firstly, from the characteristics and differences of explicit and implicit feedback data in the system, this paper analyzes the reasons for selecting implicit user feedback and the problems caused by implicit feedback in the system. Then, several classical recommendation algorithms are introduced from two aspects: based on score and based on sorting. In order to solve the problem, the current ranking recommendation algorithm only considers the preference difference between the items that the user has operated on and the items that have not been operated, and lacks the deep understanding and distinction of the degree of user preference. In this paper, the user preferences expressed by implicit feedback are explored. The types of implicit feedback are divided according to the degree of interest represented by the users, and the hierarchical implicit feedback model is used to express user preferences. Based on Bayesian personalized sorting algorithm, a three-layer implicit feedback representation model of "strong positive feedback, weak positive feedback and negative feedback" is designed. Then, after a lot of experiments, we analyze the effect of the parameters of the latent factor matrix dimension k value, the recommended list length and other parameters on the performance of the algorithm, and rank it from AUC, average precision mean value, average percentage, etc. The normalized loss cumulative gain is compared with the classical ranking recommendation algorithm, and the improved algorithm is evaluated. It is concluded that the list of recommendations generated by this algorithm can more accurately grasp the current interests of users. Finally, on the basis of the improved algorithm, starting with the analysis of the functional requirements of the personalized book system, the functional modules of the system are divided and designed, and the personalized book recommendation system based on implicit feedback is realized.
【學(xué)位授予單位】：吉林大學(xué)
【學(xué)位級別】：碩士
【學(xué)位授予年份】：2017
【分類號】：TP391.3

【參考文獻(xiàn)】

相關(guān)期刊論文 前10條

1 傅漢霖;顧小宇;;圖書推薦算法綜述[J];計(jì)算機(jī)時(shí)代;2016年12期

2 陸藝;曹健;;面向隱式反饋的推薦系統(tǒng)研究現(xiàn)狀與趨勢[J];計(jì)算機(jī)科學(xué);2016年04期

3 黃震華;張佳雯;田春岐;孫圣力;向陽;;基于排序?qū)W習(xí)的推薦算法研究綜述[J];軟件學(xué)報(bào);2016年03期

4 王東;陳志;岳文靜;高翔;王峰;;基于顯式與隱式反饋信息的概率矩陣分解推薦[J];計(jì)算機(jī)應(yīng)用;2015年09期

5 印鑒;王智圣;李琪;蘇偉杰;;基于大規(guī)模隱式反饋的個(gè)性化推薦[J];軟件學(xué)報(bào);2014年09期

6 胡勛;孟祥武;張玉潔;史艷翠;;一種融合項(xiàng)目特征和移動(dòng)用戶信任關(guān)系的推薦算法[J];軟件學(xué)報(bào);2014年08期

7 王鵬;景麗萍;;改進(jìn)的單類協(xié)同過濾推薦方法[J];計(jì)算機(jī)科學(xué)與探索;2014年10期

8 朱郁筱;呂琳媛;;推薦系統(tǒng)評價(jià)指標(biāo)綜述[J];電子科技大學(xué)學(xué)報(bào);2012年02期

9 張迎峰;陳超;俞能海;;基于最近鄰用戶動(dòng)態(tài)重排序的協(xié)同過濾方法[J];小型微型計(jì)算機(jī)系統(tǒng);2011年08期

10 周軍鋒,湯顯,郭景峰;一種優(yōu)化的協(xié)同過濾推薦算法[J];計(jì)算機(jī)研究與發(fā)展;2004年10期

相關(guān)博士學(xué)位論文 前2條

1 劉士琛;面向推薦系統(tǒng)的關(guān)鍵問題研究及應(yīng)用[D];中國科學(xué)技術(shù)大學(xué);2014年

2 任磊;推薦系統(tǒng)關(guān)鍵技術(shù)研究[D];華東師范大學(xué);2012年

相關(guān)碩士學(xué)位論文 前7條

1 王聰;基于用戶行為的個(gè)性化推薦算法研究[D];哈爾濱工業(yè)大學(xué);2015年

2 余露;利用矩陣分解算法建模數(shù)據(jù)稀疏環(huán)境下用戶協(xié)同行為[D];杭州師范大學(xué);2015年

3 沈鍵;電子商務(wù)的個(gè)性化協(xié)同過濾推薦算法研究[D];上海交通大學(xué);2013年

4 劉慶鵬;基于協(xié)同過濾的電子商務(wù)個(gè)性化推薦算法研究[D];海南大學(xué);2012年

5 郭少聃;數(shù)據(jù)稀疏和隱性反饋條件下用戶偏好挖掘方法[D];華中科技大學(xué);2012年

6 王金輝;基于標(biāo)簽的協(xié)同過濾稀疏性問題研究[D];中國科學(xué)技術(shù)大學(xué);2011年

7 張學(xué)勝;面向數(shù)據(jù)稀疏的協(xié)同過濾推薦算法研究[D];中國科學(xué)技術(shù)大學(xué);2011年

，

本文編號：1809386