简体   繁体   English

如何对在给定数据中挑选出的功能列表进行排名,并对功能进行排名?

[英]How do I rank a list of features that I have picked out in the given data and rank the features?

I am currently doing ML on house prediction dataset. 我目前正在房屋预测数据集上执行ML。 > dataset > 数据集

How do I select 10 best features in my dataset for aiding in building a model for predicting SalePrice and how to rank the 10 features ? 如何在数据集中选择10个最佳功能以帮助构建预测SalePrice的模型以及如何对这10个功能进行排名?

You can use a feature ranking algorithm like "ReliefF"or use a dimensionality reduction technique like PCA, which will transform all of your features into your desired number of features. 您可以使用“ ReliefF”之类的特征分级算法,或者使用诸如PCA之类的降维技术,该技术会将您所有的特征转换为所需数量的特征。

This is also a good link for learning more about feature selection: https://machinelearningmastery.com/feature-selection-machine-learning-python/ 这也是学习更多有关功能选择的好链接: https : //machinelearningmastery.com/feature-selection-machine-learning-python/

You might also need to convert strings to numbers in features where the values are alphanumeric in nature. 您可能还需要在特征值本质上为字母数字的功能中将字符串转换为数字。

声明:本站的技术帖子网页,遵循CC BY-SA 4.0协议,如果您需要转载,请注明本站网址或者原文地址。任何问题请咨询:yoyou2525@163.com.

 
粤ICP备18138465号  © 2020-2024 STACKOOM.COM