Principal Component analysis (PCA)
- It is a linear Dimensionality reduction technique.
- It tries to preserve the global structure of the data.
- It does not work well as compared to t-SNE.
- It does not involve Hyperparameters.
- It gets highly affected by outliers.
- PCA is a deterministic algorithm
- It works by rotating the vectors for preserving variance.
- We can find decide on how much variance to preserve using eigen values.
t-distributed stochastic neighbourhood embedding (t-SNE)
- It is a non-linear Dimensionality reduction technique.
- It tries to preserve the local structure(cluster) of data.
- It is one of the best dimensionality reduction technique.
- It involves Hyperparameters such as perplexity, learning rate and number of steps.
- It can handle outliers
- It is a non-deterministic or randomised algorithm.
- It works by minimising the distance between the point in a guassian.
- We cannot preserve variance instead we can preserve distance using hyperparameters.
#t-SNE #PCA #AI #ML #Probyto #ProbytoAI
Subscribe and follow us for latest news in Data Science and Machine learning and stay updated!