本文關(guān)鍵詞： 協(xié)同過(guò)濾 用戶評(píng)論 主題模型 融合策略　出處：《華南理工大學(xué)》2016年碩士論文　論文類型：學(xué)位論文

【摘要】：隨著互聯(lián)網(wǎng)的迅猛發(fā)展,網(wǎng)絡(luò)上與商品信息相關(guān)的數(shù)據(jù)量急劇增長(zhǎng),商品發(fā)展呈現(xiàn)多樣化、品種多、類目繁雜等特點(diǎn),互聯(lián)網(wǎng)開(kāi)始進(jìn)入大數(shù)據(jù)時(shí)代,而由于“信息過(guò)載”問(wèn)題的存在,用戶無(wú)法快速、準(zhǔn)確地定位到自己感興趣的產(chǎn)品。在此背景下,個(gè)性化推薦系統(tǒng)應(yīng)運(yùn)而生,它通過(guò)獲取用戶個(gè)性化的需求和特征,在合適的場(chǎng)景給用戶推送合適的服務(wù),引導(dǎo)用戶便捷地尋找到所需的信息,從而很好地解決“信息過(guò)載”的問(wèn)題。個(gè)性化推薦技術(shù)廣泛應(yīng)用在電子商務(wù)、廣告投放、移動(dòng)平臺(tái)等領(lǐng)域上,其中在諸多實(shí)現(xiàn)推薦的算法中,協(xié)同過(guò)濾的推薦算法得到的研究最多、應(yīng)用最為廣泛。但考慮到該算法面臨的數(shù)據(jù)稀疏問(wèn)題,以及其僅僅關(guān)注用戶發(fā)表對(duì)商品的評(píng)分,忽略了用戶發(fā)布的具有高價(jià)值的商品評(píng)論信息,本文提出一種融合用戶評(píng)論的協(xié)同過(guò)濾推薦算法,在傳統(tǒng)的評(píng)分?jǐn)?shù)據(jù)上融合用戶評(píng)論文本信息,通過(guò)應(yīng)用LDA(Latent Dirichlet Allocation)主題模型及Rocchio算法挖掘用戶發(fā)表的評(píng)論文本信息,并考慮到用戶對(duì)顯著主題的關(guān)注差異,實(shí)現(xiàn)對(duì)用戶偏好建模,在此基礎(chǔ)上提出相似度融合和評(píng)分融合兩種融合策略以及靜態(tài)加權(quán)和動(dòng)態(tài)加權(quán)兩種加權(quán)策略實(shí)現(xiàn)評(píng)論文本和評(píng)分?jǐn)?shù)據(jù)的結(jié)合,得到最終的綜合推薦結(jié)果。由于在對(duì)用戶評(píng)論文本進(jìn)行建模時(shí)使用的是用戶所有評(píng)論文本,不再是僅僅利用共同評(píng)分項(xiàng)目的數(shù)據(jù),因而能夠極大地緩和了評(píng)分?jǐn)?shù)據(jù)稀疏的問(wèn)題,同時(shí)將傳統(tǒng)協(xié)同過(guò)濾和用戶文本主題偏好信息相結(jié)合可計(jì)算得到更為精準(zhǔn)的用戶近鄰,為用戶產(chǎn)生更準(zhǔn)確的推薦。最后,針對(duì)本文提出的融合算法,選取公開(kāi)中、英文數(shù)據(jù)集及相應(yīng)效果評(píng)估指標(biāo),設(shè)計(jì)對(duì)比實(shí)驗(yàn)驗(yàn)證融合算法的有效性。實(shí)驗(yàn)結(jié)果表明:本文提出的融合算法能顯著提高傳統(tǒng)協(xié)同過(guò)濾算法的推薦效果,并且在提升效果上相似度融合策略比評(píng)分融合策略優(yōu)秀,動(dòng)態(tài)加權(quán)策略比靜態(tài)加權(quán)策略更能顯著地提高推薦效果,同時(shí)抽取顯著的LDA主題再進(jìn)行加權(quán)融合的思路可進(jìn)一步提高推薦效果。
[Abstract]:With the rapid development of the Internet, the amount of data related to the commodity information on the network has increased dramatically, the commodity development is diversified, the variety, the category is complicated and so on, the Internet begins to enter the big data era. However, due to the existence of "information overload", users can not quickly and accurately locate the products they are interested in. Under this background, personalized recommendation system emerges as the times require. It provides users with personalized needs and features, pushes the right services to the users in the right scene, and guides the users to find the information they need conveniently. Personalized recommendation technology is widely used in e-commerce, advertising, mobile platform and other fields, among which in many algorithms to implement recommendations. Collaborative filtering recommendation algorithm is the most widely studied and widely used. However, considering the data sparsity problem faced by the algorithm, and it only pays attention to the rating of the product published by the user. Ignoring the high value commodity comment information released by users, this paper proposes a collaborative filtering recommendation algorithm which integrates user comments, and integrates user comment text information on traditional rating data. By applying LDA(Latent Dirichlet allocation) topic model and Rocchio algorithm to mine the comment text information published by users. Considering the difference of the user's attention to the obvious theme, the model of user preference is realized. On this basis, two fusion strategies, similarity fusion and score fusion, as well as static weighting and dynamic weighting, are proposed to realize the combination of comment text and rating data. Get the final comprehensive recommendation result. Because the user comments text modeling is all user comments text, it is no longer just using the data of common rating items. Therefore, the problem of sparse scoring data can be greatly alleviated. At the same time, a more accurate user neighbor can be obtained by combining traditional collaborative filtering with user text topic preference information. Finally, for the fusion algorithm proposed in this paper, the open Chinese, English data sets and the corresponding evaluation indicators are selected. Experimental results show that the proposed fusion algorithm can significantly improve the recommendation effect of the traditional collaborative filtering algorithm. And the similarity fusion strategy is better than the score fusion strategy in improving the effect, and the dynamic weighting strategy can significantly improve the recommendation effect than the static weighting strategy. At the same time, the idea of extracting significant LDA themes and weighted fusion can further improve the recommendation effect.
【學(xué)位授予單位】：華南理工大學(xué)
【學(xué)位級(jí)別】：碩士
【學(xué)位授予年份】：2016
【分類號(hào)】：F724.6

【參考文獻(xiàn)】

相關(guān)期刊論文 前7條

1 蔡淑琴;袁乾;周鵬;;基于社會(huì)網(wǎng)絡(luò)關(guān)系的微博個(gè)性化推薦模型[J];情報(bào)學(xué)報(bào);2014年05期

2 鄧曉懿;金淳;韓慶平;j口良之;;基于情境聚類和用戶評(píng)級(jí)的協(xié)同過(guò)濾推薦模型[J];系統(tǒng)工程理論與實(shí)踐;2013年11期

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

4 郁雪;李敏強(qiáng);;基于PCA-SOM的混合協(xié)同過(guò)濾模型[J];系統(tǒng)工程理論與實(shí)踐;2010年10期

5 許海玲;吳瀟;李曉東;閻保平;;互聯(lián)網(wǎng)推薦系統(tǒng)比較研究[J];軟件學(xué)報(bào);2009年02期

6 劉建國(guó);周濤;汪秉宏;;個(gè)性化推薦系統(tǒng)的研究進(jìn)展[J];自然科學(xué)進(jìn)展;2009年01期

7 鄧愛(ài)林,朱揚(yáng)勇,施伯樂(lè);基于項(xiàng)目評(píng)分預(yù)測(cè)的協(xié)同過(guò)濾推薦算法[J];軟件學(xué)報(bào);2003年09期

，

本文編號(hào)：1486774