Inscrivez-vous ou connectez-vous pour rejoindre votre communauté professionnelle.
Supervised Learning is most widely used method of machine learning today. In this you try to feed in data from previous results and try to predict future results based on same. Couple of examples which will help you understand:
1. Prediction of housing prices in certain locality based on data provided for past few years.
2. Prediction if tumor of certain size is cancerous or not based on many sames collected in past.
3. Weather predictions / image recognition like Cats pictures / likes - dislike etc.
In all above examples you feed in lots of data to machines which is actual data from past (this is called test set) and based on this you try to predict what can be future or new data (dev set)
Obviously there are algorithms to fine tune our learning methods by machines come under neural networks etc...
Unsupervised learning is more about grouping of similar / identical items rather than predicting...
Hope this helps.
Supervised Machine learning is when the output of the data set is provided, the machine will run an algorithm (E.G neural networking), to try and get that result form the provided input data set, the result will get better over clusters as if you got your output in 50 clusters you might get a better output in 100 clusters.
