I have a students data set and need to predict their conversion in a college . It is an unsupervised data , which techniques can i use to predict the data and get probabilities of yes and No
you can use k-means clustering to model each observation to one of 2 states. Assuming it’s clear which state to label ‘yes’ and ‘no’, you can then do a regression model to see what factors best describe the state you care about.
KNN and Logistics Regression
You can use gans and train model on it by manually labelling the train data .
Then run your model and prdeict using generator
Hope you know the architecture of gan
Use KMean, hierarchical clustering
You have to label the data first (with YES/NO), this you can do it by collecting the historical conversions. Once your training data is ready, you can use any classification model to learn from that data and predict.