文档介绍：如何写好一篇SCI论文的
SCI高被引论文题名券
万跃华
浙江工业大学图书馆
THOMSON
利用SCI数据库中高被引
论文的题名类型进行分析,
帮助读者写好英文题名
THOMSON
题名应以最少的数量的单词来充分表述论文的内容。
题名最好由最能反映论文核心内容的主题词来扩展,要注意采用正确的单
词顺序,形容词应与其所修饰的名词紧密相邻
题名的作用主要有两个方面
(1)吸引读者。一般的读者是根据题名来考虑是否需要阅读摘要或全文。
(2)帮助文献追踪或检索。文献检索系统多以题名中的主题词作为线索,
主题词必须要准确地反映论文的核心内容,否则就有可能产生漏检。不恰当
的题名很可能会导致论文“丢失”,从而不能被潜在的读者获取。
在设计题名时,作者应思考一下“我如何检索这类信息”
为了方便二次检索,题名中应避免使用化学式、上下角标、特殊符号(数
字符号、希腊字母等)、公式、不常用的专业术语和非英语词汇(包括拉
丁语)等。部分文体指南和作者须知中还特别规定题名中不得使用专利名
化工产品、药品、材料或仪器的公司名、特殊商业标记或商标
剩用S數提库中高被引铪的题名类塑行台,鼙助自己写好萁文题
THOMSON
bat Reader -[
i, pdfl
Acrobat Reader-D万联华发表的论文6W38e28676pm
已回网
路面N4卜H◆◆·,囫
24巴国岛
眉题
Video Program Clustering Indexing Based on Face Recognition Hybrid Model 741
transformed into a 2530 gray image(the transformed image of eyes area is shom i
from below to top using 10 70 image and every apparatus of face can be found (e
rea found by scan is shown in the middle of Fig 1b). Moreover, every apparat
the below of Fig. Ib). The training face samples may be transacted also. This gains the
transformed images of every apparatus. Then the independent basis feature can be
ught. Five SVM are created through training those independent basi
reconstructed which transfers from left to right and has five states. Those five states in
HMM are corresponding to five face apparatus. The five states and transfer
probability between those states forms the develop restriction to face every apparatus
he issues of the mixed model construction, face identifying based on the mixed
model and face-clustering indexing are discussed be low.
The mixed Identifying model of Face based on svm and hmm
SVM brought forward by vapink is a statistic
theory based on Structural Risk
Minimization(SRM)[7-8]. It can be used to analyze the sort and regression problems
The decision-making hyperplane is formed based on SVM theory, which makes the
米口
组的形式
ISI Web of Knowledge
S题名通常由名词性短语构成,其举的司
Web of Sciencee
多以分词或动名词形式出现。
偷叫pp温p鄙四
ME THODS IN ENZ