英语翻译
英语翻译
随着这些年互联网的快速发展,大量的Web服务出现在网络中.目前,很多的大型电子商务系统,都或多或少地使用了各种形式的Qos预测系统.所谓的预测系统就是以消费者的消费历史为基础,通过相似度计算向消费者推荐消费者感兴趣的消费品.Web服务预测系统是建立在海量数据的基础上的一种高级商务智能平台,它可以帮助消费者节约时间从而更好、更快的从大量的相似物品中选择适合自己的东西.所以,Web服务Qos预测方法的技术研究的意义很大.
本文先介绍了预测系统的研究背景,并提出了研究该课题的意义.对于Web服务和Qos的概念、模型、技术和发展趋势等一些相关内容作了简单分析.对现有的预测系统分析可以看出预测系统主要是由数据和算法这两个重要部分组成,所以还对数据的获取做了详细的介绍.算法是预测系统的核心,本文重点介绍了算术平均算法,基于项目的协同过滤算法和欧几里德距离相似算法的算法思想和公式并对存在的一些问题作了简单分析.最后是预测算法的实现,通过matlab写出三种算法并进行仿真实验.
本文利用Matlab对Guelph大学的Eyhab Al-Masri在网上收集到的Qos数据进行仿真.通过仿真实验可以看出,欧几里德距离相似算法对于算术平均算法和基于项目评分的协同过滤算法在预测结果的准确性方面有显著的提高.
With the rapid development of the Internet, a large number of Web services appear on the network. At present, many large e-commerce systems, are more or less use of various forms of Qos prediction system. Prediction system is called to the consumer's history as the foundation, through the calculation of similarity to recommend the consumer interest to consumer goods. The Web service prediction system is based on mass data is a kind of advanced business intelligence platform, it can help consumers save time to better, faster from a large number of similar items to choose their own things. So, technology research and prediction methods of Web service Qos of great significance.
This paper first introduces the research background prediction system, and puts forward the research significance. For concept, Web services and Qos model, technology and development trend as well as some related content has made the simple analysis. The prediction system of the existing analysis shows that the prediction system is mainly composed of the two important part of data and algorithms, so the data acquisition was introduced in detail. The algorithm is to predict the core of the system, this paper focuses on the arithmetical average algorithm, the item-based collaborative filtering algorithm and Euclidean distance similarity algorithm idea and the formula and a simple analysis of the existing problems based on the. The last is predicted through MATLAB algorithm, write down three kinds of algorithm and simulation experiments.
Simulation of Qos data using the Matlab of Guelph University Eyhab Al-Masri collected online. The simulation result shows that, there is significantly improved Euclidean distance similarity collaborative filtering algorithm for the arithmetical average algorithm and based on item rating accuracy in predicting the outcome of hand.
求采纳