基于QoS反向預測的服務推薦
發(fā)布時間:2018-08-25 20:04
【摘要】:隨著云計算的發(fā)展,互聯(lián)網(wǎng)上涌現(xiàn)出越來越多的功能相同但服務質(zhì)量(QoS)不同的Web服務.基于服務質(zhì)量的服務推薦,旨在從這些等功能服務中挑選出滿足用戶服務質(zhì)量需求的服務,已成為服務計算領域的一個熱門課題.由于極少有用戶曾調(diào)用過所有候選服務,推薦系統(tǒng)將面臨服務質(zhì)量缺失的問題,因此,基于協(xié)同過濾的思想,提出一種服務質(zhì)量預測算法RST.與以往算法相比,RST算法利用反向預測機制解決數(shù)據(jù)稀疏問題,提高了預測準確度.此外,RST算法基于用戶對推薦結果的反饋,自動建立與維護信任度模型,可動態(tài)改善預測效果.最后,基于真實的數(shù)據(jù)集,驗證RST預測算法的效果,并衡量各參數(shù)對預測結果的影響.
[Abstract]:With the development of cloud computing, more and more Web services with the same function but different quality of service (QoS) emerge on the Internet. Service recommendation based on quality of Service (QoS), which aims to select services from these functional services to meet the needs of users, has become a hot topic in the field of service computing. Because very few users have ever called all candidate services, the recommendation system will face the problem of missing quality of service. Therefore, based on the idea of collaborative filtering, a quality of service prediction algorithm RST. is proposed. Compared with the previous algorithms, the RST algorithm uses reverse prediction mechanism to solve the problem of data sparsity, and improves the accuracy of prediction. In addition, based on the user feedback to the recommended results, the RST algorithm automatically establishes and maintains the trust model, which can dynamically improve the prediction results. Finally, based on the real data set, the effectiveness of the RST prediction algorithm is verified, and the influence of various parameters on the prediction results is measured.
【作者單位】: 浙江大學計算機學院;
【基金】:國家科技支撐計劃項目(2011BAH16B04)資助 國家自然科學基金項目(61173176)資助 浙江省科技項目(2008C03007)資助 國家“八六三”高技術研究發(fā)展計劃項目(2011AA010501)資助
【分類號】:TP393.09
[Abstract]:With the development of cloud computing, more and more Web services with the same function but different quality of service (QoS) emerge on the Internet. Service recommendation based on quality of Service (QoS), which aims to select services from these functional services to meet the needs of users, has become a hot topic in the field of service computing. Because very few users have ever called all candidate services, the recommendation system will face the problem of missing quality of service. Therefore, based on the idea of collaborative filtering, a quality of service prediction algorithm RST. is proposed. Compared with the previous algorithms, the RST algorithm uses reverse prediction mechanism to solve the problem of data sparsity, and improves the accuracy of prediction. In addition, based on the user feedback to the recommended results, the RST algorithm automatically establishes and maintains the trust model, which can dynamically improve the prediction results. Finally, based on the real data set, the effectiveness of the RST prediction algorithm is verified, and the influence of various parameters on the prediction results is measured.
【作者單位】: 浙江大學計算機學院;
【基金】:國家科技支撐計劃項目(2011BAH16B04)資助 國家自然科學基金項目(61173176)資助 浙江省科技項目(2008C03007)資助 國家“八六三”高技術研究發(fā)展計劃項目(2011AA010501)資助
【分類號】:TP393.09
【相似文獻】
相關期刊論文 前10條
1 蔡浩;賈宇波;黃成偉;;結合用戶信任模型的協(xié)同過濾推薦方法研究[J];計算機工程與應用;2010年35期
2 陳琦;;基于用戶反饋QoS的動態(tài)Web服務發(fā)現(xiàn)方法[J];計算機工程與設計;2011年04期
3 袁玉倩;胡曉惠;;基于QoS真實性與工作流模型的Web服務選擇[J];北京航空航天大學學報;2011年04期
4 王玲玲;張為群;劉哲;;一種改進的基于需求上下文的Web服務信任評估方法研究[J];計算機科學;2011年03期
5 薛開平;洪佩琳;李津生;黃冠堯;;P2P系統(tǒng)中基于服務反饋機制的信任模型研究[J];電路與系統(tǒng)學報;2007年05期
6 黃開枝;賀曉s,
本文編號:2203961
本文鏈接:http://sikaile.net/guanlilunwen/ydhl/2203961.html
最近更新
教材專著