基于神經(jīng)網(wǎng)絡(luò)集成和用戶偏好模型的協(xié)同過(guò)濾推薦算法研究
[Abstract]:With the development of Internet technology and the vigorous growth of electronic commerce, the network data information is growing at the exponential level, users have to spend a lot of time searching for the information and goods they want, and people have entered an era of "information overload". Recommendation system emerges as the times require, its main task is to actively push the resources it may need for users from a large number of resources, and alleviate the pressure of information retrieval. In the current application, the cooperative filtering algorithm has made the greatest achievement, but it is also inevitable to encounter many obstacles in the development. The problem of data sparse is an important problem that hinders its development. In order to solve the problem of data sparsity, from the point of view of user interest, this paper constructs a user preference model to predict the score of ungraded items from the point of view of user interest, and fills the data into the user scoring matrix. However, there are fuzziness and uncertainty in the description of user preference, which brings some difficulties to user preference modeling. It is necessary to introduce machine learning method to construct user preference model. Neural network integration algorithm has good generalization ability, and it is a hot research topic in the field of machine learning, which can be used to simulate the preferences of users. However, in the face of the complexity of user preferences, neural network integration algorithms will also have some shortcomings. In view of this situation, this paper first puts forward its own improved idea for the traditional neural network integration algorithm, and proposes a negative correlation neural network integration algorithm based on differential evolution, which improves the generalization ability of the neural network integration algorithm. Secondly, using the improved algorithm, combined with the existing user data, the user preference model is constructed. Finally, the constructed preference model is used to predict the score of ungraded items, fill the user scoring matrix, and improve the calculation of similarity in order to solve the problem of overfilling. The basic idea of negative correlation neural network integration algorithm based on differential evolution is: in order to increase the difference of integrated individuals, negative correlation learning method is introduced to train member networks in parallel, and in the generation of conclusions, the weight coefficients of member networks are optimized by using the good optimization ability of differential evolution algorithm. Through the experimental simulation and comparing it with other algorithms, the results show that the algorithm performs better both in terms of generalization performance and robustness. The basic idea of user preference model based on differential evolution neural network integration is: make full use of project feature attributes, construct project feature vector, construct user preference model through the mapping of project feature vector and user preference, and use the proposed differential evolution negative correlation neural network integration algorithm to simulate users' interests and hobbies. The experimental results show that the proposed differential evolution negative correlation neural network integration algorithm can simulate the preferences of users and predict the score of ungraded items. The basic idea of collaborative filtering recommendation algorithm based on user preference model is to use the constructed preference model to predict the score of ungraded items, fill in the user scoring matrix, and form a pseudo-user scoring matrix. When the pseudo-user rating matrix is used to calculate the similarity of users, only part of the items are selected to calculate the possible overfilling problem. Through the test of MovieLens dataset, this algorithm has better performance than the traditional collaborative filtering recommendation algorithm.
【學(xué)位授予單位】:華中師范大學(xué)
【學(xué)位級(jí)別】:碩士
【學(xué)位授予年份】:2016
【分類號(hào)】:TP391.3
【相似文獻(xiàn)】
相關(guān)期刊論文 前10條
1 沈掌泉,孔繁勝;基于廣義回歸網(wǎng)絡(luò)的動(dòng)態(tài)權(quán)重回歸型神經(jīng)網(wǎng)絡(luò)集成方法研究[J];計(jì)算機(jī)應(yīng)用研究;2005年12期
2 王正群;陳世福;陳兆乾;;帶偏置的選擇性神經(jīng)網(wǎng)絡(luò)集成構(gòu)造方法[J];計(jì)算機(jī)科學(xué);2005年07期
3 沈掌泉,孔繁勝;基于個(gè)體選擇的動(dòng)態(tài)權(quán)重神經(jīng)網(wǎng)絡(luò)集成方法研究[J];計(jì)算機(jī)工程與應(yīng)用;2005年12期
4 鞏文科;李長(zhǎng)河;石爭(zhēng)浩;趙潔;;基于樣本重構(gòu)的神經(jīng)網(wǎng)絡(luò)集成學(xué)習(xí)方法[J];計(jì)算機(jī)應(yīng)用;2006年06期
5 文習(xí)明;;基于神經(jīng)網(wǎng)絡(luò)集成的數(shù)據(jù)分析[J];現(xiàn)代計(jì)算機(jī);2006年05期
6 徐敏;;神經(jīng)網(wǎng)絡(luò)集成在圖書(shū)剔舊分類中的應(yīng)用[J];計(jì)算機(jī)工程;2006年20期
7 張曉陽(yáng);徐敏;施化吉;李星毅;;一種新型的神經(jīng)網(wǎng)絡(luò)集成模型[J];計(jì)算機(jī)工程與應(yīng)用;2006年34期
8 樂(lè)曉蓉;王正群;郭亞琴;侯艷平;;基于差異性度量的選擇性神經(jīng)網(wǎng)絡(luò)集成[J];揚(yáng)州大學(xué)學(xué)報(bào)(自然科學(xué)版);2007年02期
9 於時(shí)才;陳涓;馬寧;;一種提高神經(jīng)網(wǎng)絡(luò)集成系統(tǒng)泛化能力的方法[J];微電子學(xué)與計(jì)算機(jī);2009年04期
10 趙勝穎;高廣春;;基于蟻群算法的選擇性神經(jīng)網(wǎng)絡(luò)集成方法[J];浙江大學(xué)學(xué)報(bào)(工學(xué)版);2009年09期
相關(guān)會(huì)議論文 前5條
1 安金霞;朱紀(jì)洪;袁夏明;;基于神經(jīng)網(wǎng)絡(luò)知識(shí)庫(kù)的多神經(jīng)網(wǎng)絡(luò)集成方法[A];2009年中國(guó)智能自動(dòng)化會(huì)議論文集(第六分冊(cè))[中南大學(xué)學(xué)報(bào)(增刊)][C];2009年
2 楊育榮;左瑞芹;王科俊;;基于粒子群優(yōu)化的神經(jīng)網(wǎng)絡(luò)集成研究[A];系統(tǒng)仿真技術(shù)及其應(yīng)用(第7卷)——'2005系統(tǒng)仿真技術(shù)及其應(yīng)用學(xué)術(shù)交流會(huì)論文選編[C];2005年
3 吳月明;王益群;李莉;;基于神經(jīng)網(wǎng)絡(luò)集成的可靠性分布模型的智能識(shí)別[A];2006年全國(guó)機(jī)械可靠性學(xué)術(shù)交流會(huì)論文集[C];2006年
4 張正道;胡壽松;;基于神經(jīng)網(wǎng)絡(luò)免疫集成的非線性時(shí)間序列故障預(yù)報(bào)[A];江蘇省自動(dòng)化學(xué)會(huì)七屆四次理事會(huì)暨2004學(xué)術(shù)年會(huì)青年學(xué)者論壇論文集[C];2004年
5 杜曉鳳;丁友東;;FloatBag選擇性神經(jīng)網(wǎng)絡(luò)集成及其在人臉檢測(cè)中的應(yīng)用[A];第十二屆全國(guó)信號(hào)處理學(xué)術(shù)年會(huì)(CCSP-2005)論文集[C];2005年
相關(guān)博士學(xué)位論文 前8條
1 王征宇;神經(jīng)網(wǎng)絡(luò)集成分類方法及其在并行計(jì)算環(huán)境中的應(yīng)用研究[D];華南理工大學(xué);2015年
2 沈掌泉;神經(jīng)網(wǎng)絡(luò)集成技術(shù)及其在土壤學(xué)中應(yīng)用的研究[D];浙江大學(xué);2005年
3 劉悅;神經(jīng)網(wǎng)絡(luò)集成及其在地震預(yù)報(bào)中的應(yīng)用研究[D];上海大學(xué);2005年
4 林民龍;基于神經(jīng)網(wǎng)絡(luò)集成的增量式學(xué)習(xí)[D];中國(guó)科學(xué)技術(shù)大學(xué);2012年
5 張東波;粗集神經(jīng)網(wǎng)絡(luò)集成方法及其在模式識(shí)別中的應(yīng)用[D];湖南大學(xué);2007年
6 高敬陽(yáng);神經(jīng)網(wǎng)絡(luò)集成BOOSTING類算法研究[D];北京化工大學(xué);2012年
7 戴群;基于ICBP模型的混合神經(jīng)網(wǎng)絡(luò)與多樣化集成方法的研究[D];南京航空航天大學(xué);2009年
8 陳敏之;文胸作用下女體胸部形態(tài)變化效果分析及其模擬研究[D];東華大學(xué);2012年
相關(guān)碩士學(xué)位論文 前10條
1 樂(lè)曉蓉;神經(jīng)網(wǎng)絡(luò)集成算法設(shè)計(jì)及分析[D];揚(yáng)州大學(xué);2007年
2 張藝馨;基于神經(jīng)網(wǎng)絡(luò)集成算子的混合模型的研究與應(yīng)用[D];蘭州大學(xué);2015年
3 李汶虹;基于神經(jīng)網(wǎng)絡(luò)集成學(xué)習(xí)的SAR目標(biāo)識(shí)別方法研究[D];電子科技大學(xué);2015年
4 楊鳳萍;基于神經(jīng)網(wǎng)絡(luò)集成和用戶偏好模型的協(xié)同過(guò)濾推薦算法研究[D];華中師范大學(xué);2016年
5 陳涓;神經(jīng)網(wǎng)絡(luò)集成的泛化能力研究及其應(yīng)用[D];蘭州理工大學(xué);2009年
6 劉何秀;神經(jīng)網(wǎng)絡(luò)集成算法的研究[D];中國(guó)海洋大學(xué);2009年
7 徐瑜;基于主動(dòng)差異學(xué)習(xí)神經(jīng)網(wǎng)絡(luò)集成的電力變壓器故障診斷方法研究[D];湘潭大學(xué);2010年
8 李巖;神經(jīng)網(wǎng)絡(luò)集成及其在分類和回歸問(wèn)題中的應(yīng)用研究[D];華北電力大學(xué)(河北);2010年
9 張冬威;基于網(wǎng)絡(luò)聚類選擇的神經(jīng)網(wǎng)絡(luò)集成方法及其在農(nóng)業(yè)中的應(yīng)用[D];吉林大學(xué);2011年
10 徐敏;基于神經(jīng)網(wǎng)絡(luò)集成的圖書(shū)剔舊專家系統(tǒng)的研究[D];江蘇大學(xué);2006年
,本文編號(hào):2505015
本文鏈接:http://sikaile.net/jingjilunwen/dianzishangwulunwen/2505015.html