1. 初识机器学习
1. 机器学习的次要算法(Machine learning algorithm)分类:
-
监督学习(Supervised learning):给予学习算法示例,包含正确的答案 (learns from being given right answers):
Regression(回归): 试图预测一个数字,从可能的到有限多个可能的后果中失去预测(Predict a number,infinitely many possible outputs)*** 房价回归预测 ***
Classification(分类):预测分类,从一组无限的可能的输入后果中进行分类(Predict categories,small number of possible outputs)
在医学诊断肿瘤的算法中,通过和肿块大小和病患年龄的关系,预测肿瘤的性质,通过数据拟合一条分界线,以便对后果做出预测
- 无监督学习(Unsupervised learning): 从一组无标签的数据中找到咱们感兴趣的数据据(Find something interesting in unlabeled data,Data only come with inputs x,but not output y,algorithm has to find structure in the data)
clustering(聚类算法):将未标记的数据搁置到不同的集群中 (**Group similar data
points together**)
Dimensionality Reduction(降维算法):
-
强化学习(Reinforcement learning):待续
- 无监督学习(Unsupervised learning): 从一组无标签的数据中找到咱们感兴趣的数据据(Find something interesting in unlabeled data,Data only come with inputs x,but not output y,algorithm has to find structure in the data)