传统搜索引擎的搜索结果采用的是以一维列表的形式展现,随着文献数据的急剧增多,用户对于搜索结果的辨识和分析速度也在急剧下降。为了弥补传统工具的这一不足,本文开发了“语义图”(semanticmap,SMAP),此工具对文献数据进行数据挖掘和可视化,包括关联匹配和聚类,将搜索结果以二维矩阵图的形式展示出来,方便用户理解数据之间的内部联系,并帮助用户迅速从整体上把握搜索结果。最后以蛋白质组学文献分析过程为例具体展示了此工具的应用。
The results of traditional search engine technologies are currently shown as one-dimensional list. With the explosion of publication data, users get rapid slowdown on the cognition of the results. Hence we were motivated to develop a novel method of literature data mining and visualization called semantic map(SMAP) , utilizing data mining tech- niques and visualization, including association matching and clustering, etc. A semantic map is shown as a two-dimensional array image to reveal the inner relations of data, and could help users quickly grasp the entire results from whole data. Finally, literature data of proteomics was analyzed, which was a good example of applying SMAP.