Homework_Week8_Coursera【Machine Learning】AndrewNg、Principal Component Analysis
- 第一题Consider the following 2D dataset:Which of the following figures correspond to possible values that PCA may return for u...
- 第二题 Which of the following is a reasonable way to select the number of principal components k? (Recall that n is the dimensionality of the input data and mm is the number of input examples.)
- 第三题 Suppose someone tells you that they ran PCA in such a way that "95% of the variance was retained." What is an equivalent statement to this?
- 第四题 Which of the following statements are true? Check all that apply
- 第五题 Which of the following are recommended applications of PCA? Select all that apply.
第一题Consider the following 2D dataset:Which of the following figures correspond to possible values that PCA may return for u…
解析
Which of the following figures correspond to possible values that PCA may return for u^{(1)}u
(1)
(the first eigenvector / first principal component)?
很明显,求高纬度在低纬上误差最小的表示方法,那么只有AB两种方向能表示梯度方向
答案
AB
第二题 Which of the following is a reasonable way to select the number of principal components k? (Recall that n is the dimensionality of the input data and mm is the number of input examples.)
解析
这里的k不是kmeans中的聚类数量。要注意
答案
D
第三题 Suppose someone tells you that they ran PCA in such a way that “95% of the variance was retained.” What is an equivalent statement to this?
解析
答案
D
第四题 Which of the following statements are true? Check all that apply
解析
A说 automatically错误、
B 给到了输入、PCA把它压缩到更低维度的数据,正确
C 如果输入特征是不同的规模,那么在应用PCA之前做以恶个特征扫描是个不错的选择、正确
D PCA只可以用于减少数据1维度,明显错误。
答案
BC
第五题 Which of the following are recommended applications of PCA? Select all that apply.
解析
推荐给PCA应用
A说给一个学习算法更多特征去学习、错误/明显减少了学习的维度。所以错
B说数据压缩可以减少输入数据X的维度、用于监督学习算法,所以监督学习算法会跑的更快、确实没错
C降维数据至2-3维就可以被画出、没错
D说自动把样本分组、这PCA应该算是准备动作,pca只是做了数据降维,并没有分组
答案
BC