如何评价分类模型性能?(足球荔枝)
2017-02-28 21:36
274 查看
【1】评价分类模型,我们一般从两个点出发
1.通过指标来评估该模型是否适合对预测对象? 评价指标主要有:1)Precision;2)Recall;3)F-score;4)Accuracy;5)ROC;6)AUC[1]
2.通过计算预测模型所产生的模拟值与历史实际值拟合程度的优劣来估计该模型的预测值的拟合效果。
指标有:预测精度和预测准确度,两者是不同的概念。 [2]
准确率accuracy 和 精度precision 的不同: [3]
(1)Accuracy is how close a measured value is to the actual (true) value.
准确率是测量值与实际(真)值的接近程度。
(2)Precision is how close the measured values are to each other.
精度是测量值彼此接近的程度。
So, if you are playing soccer and you always hit the left goal post instead of scoring, then you are not accurate, but you are precise!
对于 预测精度和预测准确度的区分,你也可以参见 wikipedia:[4] Accuracy
and precision
Accuracy is the proximity of measurement results to the true value; precision, the
repeatability, or reproducibility of the measurement. (准确度是测量结果与真实值的接近程度,而精度表示重复性,测量值是否集中)
According to ISO 5725-1,[5] the
general term "accuracy" is used to describe the closeness of a measurement to the true value. When the term is applied to sets of measurements of the same measurand,
it involves a component of random error and a component of systematic error. In this case trueness is the closeness of the mean of a set of measurement results to the actual
(true) value and precision is the closeness of agreement among a set of results.
References
[1] 分类算法中常用的评价指标
[2] 预测精度
[3] Accuracy and Precision(www.mathsisfun.com)
[4] Accuracy and Precision( wikipedia)
1.通过指标来评估该模型是否适合对预测对象? 评价指标主要有:1)Precision;2)Recall;3)F-score;4)Accuracy;5)ROC;6)AUC[1]
2.通过计算预测模型所产生的模拟值与历史实际值拟合程度的优劣来估计该模型的预测值的拟合效果。
指标有:预测精度和预测准确度,两者是不同的概念。 [2]
准确率accuracy 和 精度precision 的不同: [3]
(1)Accuracy is how close a measured value is to the actual (true) value.
准确率是测量值与实际(真)值的接近程度。
(2)Precision is how close the measured values are to each other.
精度是测量值彼此接近的程度。
Examples of Precision and Accuracy:
Low Accuracy High Precision | High Accuracy Low Precision | High Accuracy High Precision |
对于 预测精度和预测准确度的区分,你也可以参见 wikipedia:[4] Accuracy
and precision
Accuracy is the proximity of measurement results to the true value; precision, the
repeatability, or reproducibility of the measurement. (准确度是测量结果与真实值的接近程度,而精度表示重复性,测量值是否集中)
According to ISO 5725-1,[5] the
general term "accuracy" is used to describe the closeness of a measurement to the true value. When the term is applied to sets of measurements of the same measurand,
it involves a component of random error and a component of systematic error. In this case trueness is the closeness of the mean of a set of measurement results to the actual
(true) value and precision is the closeness of agreement among a set of results.
References
[1] 分类算法中常用的评价指标
[2] 预测精度
[3] Accuracy and Precision(www.mathsisfun.com)
[4] Accuracy and Precision( wikipedia)
相关文章推荐
- 二分类模型性能评价(R语言,logistic回归,ROC曲线,lift曲线,lorenz曲线)
- 二分类模型性能评价(R语言,logistic回归,ROC曲线,lift曲线,lorenz曲线)
- QQ餐厅与系统性能模型(续):如何评价系统的性能
- 二分类模型性能评价 2.0(ROC曲线,lift曲线,lorenz曲线)
- 分类模型的性能评估——以SAS Logistic回归为例(1): 混淆矩阵
- 分类模型的性能评估——以SAS Logistic回归为例(2): ROC和AUC
- 分类模型的性能评估——以SAS Logistic回归为例(1): 混淆矩阵
- 孰优孰劣 如何评断?——搜索引擎性能评价的方法
- 如何评价一个嵌入式OS的性能
- 使用Weka进行数据挖掘(Weka教程七)Weka分类/预测模型构建与评价
- 视频客观质量评价模型性能的评估方法
- R语言︱机器学习模型评价指标+(转)模型出错的四大原因及如何纠错
- 分类模型的性能评估——以SAS Logistic回归为例(2): ROC和AUC
- 前端性能评价模型自动分析方法研究
- PHP之父评价Facebook的HipHop项目:谈如何找出PHP的网站的性能瓶颈
- 如何利用AdaBoost提高分类性能
- 分类模型的性能评估——以SAS Logistic回归为例(1): 混淆矩阵
- 如何提高模型的性能
- 浏览器性能对比测试方法与评价模型研究
- 分类模型的性能评估——以SAS Logistic回归为例(2): ROC和AUC