欢迎您！东篱公司
退出
-
申报数据库
-
立项数据库
-
成果数据库
-
项目获奖数据库
中文摘要：
论文引用网络是一个动态变化的网络,不断有新的论文加入引用网络中.传统的论文评 价标准如引用次数、PageRank值等＂终身评价标准＂存在排挤新结点的问题,如何在海量论文中寻找有 价值、被持续关注的论文,成为人们感兴趣的问题. Sayyadi提出了FutureRank算法,该算法通过预测论文未来＂一段时间＂的被引次数排名和PageRank值排 名来达到这一目的.但FutureRank算法需提前计算PageRank值,要耗费大量运算时间.据此,我们尝 试在不计算论文现有PageRank值的条件下,从论文的撰写者以及引用者的权威值的角度来预测论文未来 的被引次数排名和PageRank值排名.实验结果表明,我们的算法与FutureRank相比,不但缩短了运算时间,而且提高了预测准确率.
英文摘要：
Citation network is a dynamic network and new papers are added to it every day. The traditional literature evaluation criteria like citations number and PageRank are unfair to the new node. How to retrieve the valuable papers of continuous concerns has become an interesting focus. To solve this problem, Hassan Sayyadi proposed the FutureRank algorithm, but it needs to calculate the PageRank value, which takes a lot of time. Accordingly, we proposed a paper value prediction algorithm without computing the PageRank value. We predict paperts rank of citations number and PageRank value in the future by writerst authority value and citer~s authority value. Experimental results show that as compared with FutureRank, our algorithm not only shortens the computing time but also improves the forecast accuracy.
同期刊论文项目
同项目期刊论文
Reduced Ordered Binary Decision Diagram with Implied Literals: A New Knowledge Compilation Approach.
A new hybrid method based on local fisher discriminant analysis and support vector machines for hepa
A support vector machine classifier with rough set based feature selection for breast cancer diagnos
期刊信息
位置：