期刊文献+
共找到3篇文章
< 1 >
每页显示 20 50 100
基于网格和密度的随机样例的聚类算法 被引量:2
1
作者 孙志伟 赵政 王红梅 《天津大学学报》 EI CAS CSCD 北大核心 2006年第5期621-626,共6页
为提高密度聚类算法效率并处理非空间属性约束,提出了基于网格和密度的聚类算法(GDRS).它使用网格区域表示点的邻域,非空间属性被分为数值和字符类型.首先通过网格方法找到能准确反映数据空间几何特征的参考点;然后随机选择没有分类... 为提高密度聚类算法效率并处理非空间属性约束,提出了基于网格和密度的聚类算法(GDRS).它使用网格区域表示点的邻域,非空间属性被分为数值和字符类型.首先通过网格方法找到能准确反映数据空间几何特征的参考点;然后随机选择没有分类的参考点,并测试其邻域的稀疏状况、与其他聚类的关系以及非空间属性的约束来决定加入、合并聚类或形成新的聚类;最后把参考点映射回数据.把此算法和DBSCAN及DBRS算法进行了理论比较,并使用合成和真实数据集对GDRS和DBSCAN进行了对比.实验表明,GDRS具有密度算法的优点,即可发现各种形状的聚类并能屏蔽噪声点,且执行效率明显优于密度算法. 展开更多
关键词 数据挖掘 聚类算法 密度 网格 参考点 随机样例 约束
下载PDF
从有分类噪声的随机样例中学习k-判定表
2
作者 尹激雷 朱洪 《计算机学报》 EI CSCD 北大核心 1994年第1期16-22,共7页
在本文中,我们解决了Rivest在[4]中提出的一个悬而未决的问题:证明了在Valiant可学习模型下,从带有分类噪声的随机样例中可学习k-判定表.
关键词 分类 噪声 随机样例 k-判定表
下载PDF
Analysis of Prognostic Factors of Esophageal and Gastric Cardiac Carcinoma Patients after Radical Surgery Using Cox Proportional Hazard Model-A Random Sampling Study from the Fourth Hospital of Hebei Medical University during the Period of 1996-2004
3
作者 Wei Liu Xishan Hao +12 位作者 Qian Fan Peizhong Wang Haixin Li Linan Song Shijie Wang Ying Jin Yong Chen Liyun Guan Yumin Ping Xianli Meng Rui Wang Junfeng Liu Xiaoling Wang 《Clinical oncology and cancer researeh》 CAS CSCD 2009年第4期290-295,共6页
OBJECTIVE To retrospectively analyze clinical data of patientsfrom our hospital who underwent radical surgery for esophagealcarcinoma and for adenocarcinoma of the gastric cardia,as well asto investigate prognostic fa... OBJECTIVE To retrospectively analyze clinical data of patientsfrom our hospital who underwent radical surgery for esophagealcarcinoma and for adenocarcinoma of the gastric cardia,as well asto investigate prognostic factors affecting the long-term survival ofthe patients.METHODS Data from the patients eligible for our study,admitted to the 4th Hospital of Hebei Medical University fromJanuary 1996 to December 2004,were randomized,and 12distinctive clinicopathologic factors influencing the survival rateof those who underwent radical surgery for esophageal carcinomaor carcinoma of the gastric cardia were collected.Univariate andmultivariate analysis of these individual variables were performedusing the Cox proportional hazard model.RESULTS It was shown by univariate analysis that age,tumorsize,pathologic type,lymph node status,TNM staging,depthof infiltration and encroachment into local organs,etc.,were thefactors that markedly influenced the prognosis of patients(P<0.01).Multivariate analysis showed that pathologic type,numberof the lymph node metastases,involvement of local organs,andTNM staging were independent prognostic factors(P<0.05).CONCLUSION The independent factors influencing theprognosis of patients with esophageal cancer and carcinoma ofthe gastric cardia include pathologic type,number of lymph nodemetastases,involvement of local organs and TNM staging.Themain prognostic factors affecting the patient's survival are patientage,tumor size and depth of infiltration.In addition,patients withinvolvement of the local organs have a worse prognosis,and theyshould be closely followed up. 展开更多
关键词 esophageal carcinoma carcinoma of gastriccardia Cox model prognosis.
下载PDF
上一页 1 下一页 到第
使用帮助 返回顶部