Oracle Interview Questions | Dimentionality
Question
What is meant by ‘curse of dimensionality’? How can we solve it?
in progress
0
Machine Learning
55 years
2 Answers
898 views
Great Grand Master 0
Answers ( 2 )
Curse of Dimensionality occurs when there are too many features in your data set.
For a Machine Learning algorithm to perform well, your data set must at least have 10
observations behind every single feature. When the ratio of observations to features is
not adequate, you run the risk of overfitting the model. Also, with too man features,
observations become harder to cluster.
To avoid such issues, we can dimensionality reduction techniques like PCA( Principal Components Analysis).
PCA combines highly correlated features into a single feature and helps in reducing the dimesions.
Curse of dimensionality occurs when there are too many feature like 100’s of features. When we have suc large number of features then it becomes difficult to visulaise the data, find correlations, fit the data.
This can be removed if we reduce features by using Principal Component Analysis (PCA), Linear Discriminant Analysis (LDA) or Autoencoder.