本文選題：協(xié)同過(guò)濾　切入點(diǎn)：用戶(hù)興趣　出處：《安徽理工大學(xué)》2017年碩士論文　論文類(lèi)型：學(xué)位論文

【摘要】：面對(duì)大數(shù)據(jù)的時(shí)代,怎么從雜亂無(wú)章的信息海洋里準(zhǔn)確的推薦給用戶(hù)感興趣的信息,這將是推薦算法研究的主要任務(wù)。最為經(jīng)典的兩個(gè)推薦算法是基于內(nèi)容過(guò)濾和協(xié)同過(guò)濾推薦算法,但再經(jīng)典的推薦算法也有自己的缺點(diǎn)。數(shù)據(jù)稀疏性和冷啟動(dòng)是協(xié)同過(guò)濾推薦算法的主要問(wèn)題�；趦�(nèi)容過(guò)濾的推薦算法有一個(gè)比較嚴(yán)重的問(wèn)題,那就是新用戶(hù)問(wèn)題,因?yàn)樵撍惴ú⑽纯紤]用戶(hù)的興趣改變對(duì)推薦效果的影響。當(dāng)系統(tǒng)中新增一個(gè)用戶(hù)時(shí),新增用戶(hù)的歷史瀏覽記錄是不存在的,它將無(wú)法對(duì)新增用戶(hù)做出正確的推薦。針對(duì)這些,本文提出一種結(jié)合了用戶(hù)興趣和領(lǐng)域最近鄰的的混合推薦算法(UIDNN),用于個(gè)性化服務(wù)推薦。首先,考慮用戶(hù)的興趣偏好不是永遠(yuǎn)不變的。用戶(hù)的興趣偏好隨著時(shí)間的變化跟人類(lèi)對(duì)于基本事物的遺忘規(guī)律很類(lèi)似。引入非線(xiàn)性逐步遺忘函數(shù)求取用戶(hù)對(duì)商品項(xiàng)目的興趣度。然后根據(jù)用戶(hù)-商品屬性標(biāo)簽集合形成用戶(hù)-興趣度集合,對(duì)用戶(hù)-商品項(xiàng)目評(píng)分集合中未評(píng)價(jià)商品項(xiàng)目采用平均值法進(jìn)行填充、已評(píng)價(jià)商品項(xiàng)目進(jìn)行互補(bǔ)形成用戶(hù)-興趣度矩陣,降低了數(shù)據(jù)的稀疏性。其次,引入"屬性領(lǐng)域最近鄰"方法查找目標(biāo)用戶(hù)的最近鄰,在查找最近鄰居時(shí),根據(jù)用戶(hù)-興趣度集合去降低算法的在線(xiàn)計(jì)算量。這種做法主要是通過(guò)判斷目標(biāo)用戶(hù)的鄰居有沒(méi)有這個(gè)推薦能力,從而不去考慮那些對(duì)目標(biāo)用戶(hù)無(wú)推薦能力的用戶(hù)。預(yù)測(cè)未評(píng)價(jià)商品評(píng)分,采用用戶(hù)-興趣度集合的余弦相似度計(jì)算用戶(hù)的相似度;最后把與目標(biāo)用戶(hù)相似度大小在前N位的項(xiàng)目推薦給目標(biāo)用戶(hù)。基于這些對(duì)目標(biāo)用戶(hù)進(jìn)行推薦。通過(guò)實(shí)驗(yàn),本文提出的基于用戶(hù)興趣和領(lǐng)域最近鄰的混合推薦算法(UIDNN)跟相似度計(jì)算方法為皮爾遜相似度(Pearson)、余弦相似度(cos)兩種傳統(tǒng)的基于用戶(hù)的協(xié)同過(guò)濾推薦算法進(jìn)行比較平均絕對(duì)誤差(MAE),由實(shí)驗(yàn)結(jié)果圖可以看出,本文提出的基于用戶(hù)興趣和領(lǐng)域最近鄰的混合推薦算法(UIDNN)有較小的MAE,說(shuō)明本文提出的UIDNN算法有較高的推薦質(zhì)量。
[Abstract]:In the face of big data's time, how to accurately recommend information of interest to users from a messy ocean of information, This will be the main task in the research of recommendation algorithms. The two most classical recommendation algorithms are based on content filtering and collaborative filtering recommendation algorithms. However, the classical recommendation algorithm also has its own shortcomings. Data sparsity and cold start are the main problems of collaborative filtering recommendation algorithm. There is a serious problem in the content filtering recommendation algorithm, that is, the problem of new users. Because the algorithm does not take into account the influence of the user's interest change on the recommendation effect. When a new user is added to the system, the historical browsing record of the new user does not exist, and it will not be able to make the correct recommendation to the new user. In this paper, we propose a hybrid recommendation algorithm, which combines the interests of users and the nearest neighbor of the domain, for personalized service recommendation. The change of user's interest preference over time is very similar to the law of human's forgetting of basic things. The nonlinear stepwise forgetting function is introduced to find the user's item of merchandise. Interest. Then form a user-interest set based on the user-commodity attribute label set, The average value method is used to fill the unevaluated items in the user-commodity item score set. The evaluated commodity items complement each other to form the user-interest matrix, which reduces the sparsity of the data. The nearest neighbor of the property domain method is introduced to find the nearest neighbor of the target user. Based on the user-interest set to reduce the online computation of the algorithm. This approach is mainly by judging whether the neighbor of the target user has the ability to recommend. Therefore, the users who have no recommendation ability to the target users are not considered. The users' similarity is calculated by using the cosine similarity of the user-interest set. Finally, we recommend the items with the first N bit similarity to the target users. Based on these, we recommend the target users. This paper proposes a hybrid recommendation algorithm based on user interest and domain nearest neighbor (UIDNN) and its similarity calculation methods are Pearsonian (cosine similarity) and Pearsonian (Pearsonian), two traditional user-based collaborative filtering and recommendation algorithms are compared. The mean absolute error can be seen from the diagram of the experimental results. The proposed hybrid recommendation algorithm based on user interest and domain nearest neighbor has a small mae, which shows that the proposed UIDNN algorithm has high recommendation quality.
【學(xué)位授予單位】：安徽理工大學(xué)
【學(xué)位級(jí)別】：碩士
【學(xué)位授予年份】：2017
【分類(lèi)號(hào)】：TP391.3

【參考文獻(xiàn)】

相關(guān)期刊論文 前10條

1 郭弘毅;劉功申;蘇波;孟魁;;融合社區(qū)結(jié)構(gòu)和興趣聚類(lèi)的協(xié)同過(guò)濾推薦算法[J];計(jì)算機(jī)研究與發(fā)展;2016年08期

2 胡德敏;龔燕;;基于譜聚類(lèi)和擴(kuò)展樸素貝葉斯的混合推薦算法[J];計(jì)算機(jī)應(yīng)用研究;2016年12期

3 王桐;曲桂雪;;基于柯西分布量子粒子群的混合推薦算法[J];中南大學(xué)學(xué)報(bào)(自然科學(xué)版);2015年08期

4 宋文君;郭強(qiáng);劉建國(guó);;一種改進(jìn)的混合推薦算法[J];上海理工大學(xué)學(xué)報(bào);2015年04期

5 謝彬;唐健常;唐新懷;;基于排序?qū)W習(xí)的混合推薦算法[J];黑龍江科技大學(xué)學(xué)報(bào);2015年04期

6 袁林;虞飛華;;在線(xiàn)商品個(gè)性化混合推薦算法研究[J];浙江樹(shù)人大學(xué)學(xué)報(bào)(自然科學(xué)版);2015年02期

7 郭均鵬;王啟鵬;寧?kù)o;李嬡嬡;;基于符號(hào)數(shù)據(jù)與非負(fù)矩陣分解法的混合推薦算法[J];系統(tǒng)管理學(xué)報(bào);2015年03期

8 景民昌;;從ACM RecSys' 2014國(guó)際會(huì)議看推薦系統(tǒng)的熱點(diǎn)和發(fā)展[J];現(xiàn)代情報(bào);2015年04期

9 翟爍;;基于用戶(hù)興趣和雙重聚類(lèi)融合的協(xié)同過(guò)濾算法的優(yōu)化研究[J];無(wú)線(xiàn)互聯(lián)科技;2015年05期

10 李瑞敏;林鴻飛;閆俊;;基于用戶(hù)-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

，

本文編號(hào)：1652867