小样本量大肠癌数据对BP神经网络准确度的影响探讨 |
投稿时间:2018-03-23 点此下载全文 |
引用本文:刘秀峰,刘芬,李金城.小样本量大肠癌数据对BP神经网络准确度的影响探讨[J].医学信息学杂志,2018,39(8):62-66 |
摘要点击次数: |
全文下载次数: |
|
基金项目:广州中医药大学薪火计划(项目编号:XH20160105)。 |
|
中文摘要:采用二进制编码对数据进行量化,建立基于BP神经网络的大肠癌证型分类模型并分析其优势,探索8个证型的样本量对BP神经网络准确度的影响,通过删减较少样本的大肠癌证型得出样本量与准确度的量化关系。 |
中文关键词:大肠癌 小样本量 BP神经网络 准确度 |
|
Discussion on the Impact of Colorectal Cancer Data of a Small Sample Size on BP Neural Network Accuracy |
|
|
Abstract:The paper quantifies data by using binary coding, builds the syndrome type classification model of colorectal cancer based on BP neural network, analyzes its advantages, and explores the impact of the sample size of 8 syndrome types on Back Propagation (BP) neural network accuracy. By ruling out colorectal cancer syndrome types of a small number of samples, it gains the quantified relation between sample size and accuracy. |
keywords:Colorectal cancer Small sample BP neural network Accuracy |
查看全文 查看/发表评论 下载PDF阅读器 |