supervised machine learning algorithum are trained using label data |
unsupervised machine learning alorithum are traind using unlabeled data |
Model takes direct feedback to check whether it is predicting correct output or not |
Model does not take any feedback |
Model predicts the output |
Model does not take any feedback |
Input data provided to model along with output |
Only input data is provide to the model |
GOAL : To train model so that it can predict the output when it given new data |
GOAL: To find the hidden pattern & useful insights from the unknown dataset |
Need supervision to train the model |
Does not need supervision to train model |
Used for those cases where we know the input as well as corresponding output data |
Used for those cases where we know the input data and no corresponding output data |
Catogerical in regression and classification |
Classification in clustering and association |
Produce accurate result |
Model may give less accurate result as compared to supervised learning |
SL is not close to AI as in this case, we first train the model for each data & then only it can predict the correct output |
More close to true AI as it learn similarity as a child learns daily routine things by his experinces |
Include algorithum : Regression, Decision Tree , SVM, Multi-class Classification , K-Neighbour , Logistic Regression.,etc |
Include algorithum : Clustering, KNN, Apriori Algorithum.,etc |