什麼是 Feature Engineering ?

Machine Learning机器学习 是用输入的 feature特征 经过机器学习的算法来 得到 predictive models预测模型。
Feature engineering特征工程 的目的就是选择、找到一个更好的Feature为的是predictive models预测模型更加准确。
我们先解释一下Feature。
任何一组选定的数据列(我想表达的是Attribute属性)都可以作为Feature。我举一个脑洞的例子,想预测人性别是男是女,我们会更乐意试着选择一个人的姓名作为Feature而不是他/她的手机号。因为名字直觉上更能像是可能分辨出人性别的一种信息。
所以说选择Feature是个大难题!
罗列一下Wikipedia维基百科里写的Feature engineering步骤:
1. Brainstorming Or Testing features 头脑风暴或测试的方法挑选features
2. Deciding what features to create 决定想去创建什么features
3.Creating features 创建features
4. Checking how the features work with your model 检查这些features在现有模型中的输出情况
5. Improving your features if needed 如有需要改进feature
6. Go back to brainstorming/creating more features until the work is done. 回到第一步头脑风暴,以及考录是否需要添加更多的features,直到认为足够为止。
若有错误或不清楚之处 望指正
Ref.
【什麼是 Feature Engineering ?】 Wikipedia:Feature engineering

■网友
feature engineering is the process of transforming raw data into features that better represent the underlying problem to the predictive models, resulting in improved model accuracy an unseen data.Feature engineering turn your inputs into things the algorithm can understand.From Mindorks


    推荐阅读