图书情报工作

图书情报工作 >

2013 , Vol. 57 >Issue 01: 136 - 141

DOI: https://doi.org/10.7536/j.issn.0252-3116.2013.01.024

知识组织

基于传递闭包的知识发现方法和Swanson知识发现方法的比较研究——以癌药物靶点为例

  • 杨渊 ,
  • 李扬 ,
  • 孙晓北 ,
  • 高柳滨 ,
  • 池慧
展开
  • 1. 中国医学科学院医学信息研究所;
    2. 中国科学院上海药物研究所
杨渊,中国医学科学院医学信息研究所研究实习员;李扬,中国医学科学院医学信息研究所助理研究员;孙晓北,中国医学科学院医学信息研究所助理研究员;高柳滨,中国科学院上海药物研究所研究馆员

收稿日期: 2012-09-12

  修回日期: 2012-11-21

  网络出版日期: 2013-01-05

收起

Comparative Study on Knowledge Discovery Methods of Transitive Closure-based and Swanson: Taking Cancer Drug Target for an Example

  • Yang Yuan ,
  • Li Yang ,
  • Sun Xiaobei ,
  • Gao Liubin ,
  • Chi Hui
Expand
  • 1. Institute of Medical Information, Chinese Academy of Medical Sciences, Beijing 100020;
    2. Shanghai Institute of Materia Medica, Chinese Academy of Sciences, Shanghai 201203

Received date: 2012-09-12

  Revised date: 2012-11-21

  Online published: 2013-01-05

Fold

摘要

以探寻癌药物靶点间的潜在关联为例,对基于传递闭包的知识发现方法和Swanson的一阶知识发现方法进行了比较,结果得到,运用传递闭包的方法获得知识假设,可能发现更多有意义的新关联,且得到较高查全率的同时,并没有牺牲过多的查准率。

关键词： 传递闭包; 知识发现; Warshall算法; 向量空间模型; 药物靶点

本文引用格式

杨渊 , 李扬 , 孙晓北 , 高柳滨 , 池慧 . 基于传递闭包的知识发现方法和Swanson知识发现方法的比较研究——以癌药物靶点为例[J]. 图书情报工作, 2013 , 57(01) : 136 -141 . DOI: 10.7536/j.issn.0252-3116.2013.01.024

Abstract

The mechanism of method to find the transitive closure in discrete mathematics and Swanson’s method to discovery disjoint literature knowledge is similar, but use the former to discovery disjoint literatures can develop the one step to multi steps at one time to get more potential relevance. Taking the cancer drug target discovery for an example, this paper compares the two methods. The results show that using the transitive closure method to acquire knowledge assumptions may find more meaningful association; it can get a higher recall rate, while not sacrifice too much precision.

Key words： transitive closure; knowledge discovery; Warshall’s algorithm; vector space model; drug target

参考文献

[1] Swanson D R. Undiscovered public knowledge[J].Library Quarterly,1986,56(2):103-118.



[2] Lindsay R K, Gordon M D. Literature based discovery by lexical statistics[J]. Journal of American Society for Information Science, 1999(49):674-685.



[3] Gordon M D, Dumais S.Using latent semantic indexing for literature based discovery[J]. Journal of American Society for information Science,1998,4(8): 674-685.



[4] Wren J D. Extending the mutual information measure to rank inferred literature relationship[J]. BMC Bioinformatics, 2004(5):145.



[5] 张云秋,冷伏海.基于文献内聚度的非相关文献知识发现排序方法研究[J].现代图书情报技术,2009(6):50-54.



[6] 张云秋,于双成.基于MeSH加权的非相关文献知识发现排序方法研究[J].情报理论与实践,2009,32(7):113-115.



[7] 张云秋,郭柯磊. 基于双向词频统计的非相关文献知识发现排序方法研究[J]. 情报科学,2009,27(8):1240-1244.



[8] 张云秋,冷伏海.非相关文献知识发现初始集过滤方法的试验研究[J].图书情报工作,2009, 53(16):116-119.



[9] Stegmann J, Grohmann G. Hypothesis generation guided by co-word clustering[J]. Scientometrics,2003,56(1): 111-135.



[10] Huang Wei, Nakamoril Y, Wang Shouyang, et al. Mining scientific literature to predict new relationships[J]. Intelligent Data Analysis,2005, 9(2):219-234.



[11] Yetisgen-Yildiz M, Pratt W. Using statistical and knowledge-based approaches for literature based discovery[J]. Journal of Biomedical Informatics,2006,39(6):600-611.



[12] Hristovski D, Stare J, Peterlin B, et al. Supporting discovery in medicine by association rule mining of bibliographic database[C]//Zighed D A, Komorowski H J, Zytkow J M. Principles of Data Mining and Knowledge Discovery, 4th European Conference, PKDD 2000, Lyon, France, September 13-16, 2000.Heidelberg:Springer,2000.



[13] 陈显强.二元关系的传递性和传递闭包探讨[J].数学的实践与认识,2004,34(9): 135-137.



[14] 杨渊,高柳滨. 基于传递闭包方法的非相关文献知识发现探索——以癌药物靶点为例[J].现代图书情报技术,2010(12):52-57.



[15] [EB/OL].[2012-09-10].https://www.thomson-pharma.com/portal/page/portal/Search/BROWSE%20TARGETS?_dummy=y&termType=Targets.

Options
文章导航

/

〈 〉