Exploratory Data Analysis and apply TSNE On DonorsChoose data set features:
pipeline of EDA: 1. Reading Data,
2. Data Analysis(Univariate Analysis for each feature and ploting with summary ),
3. Text preprocessing - Text Data,
4.Preparing data models - Vectorizing categorical data(one hot encoding),
5.Vectorizing Text data ( Bag of words, TFIDF, avg W2v, TFIDF weighted W2V ),
6.Vectorizing Numerical features ( StandardScaler ),
7.Merging all the above features (Categorical + Text + numarical ) - using hstck method,
8. Apply TSNE,
9. Overall Summary.
Thank you.