data collection

[SamuPre]

How many data points are necessary to build a reliable machine learning model (with mlc)? 

Is there a point where adding more data no longer significantly improves the model's performance?

[Federica Bossi]

Hi @SamuPre ,

There is no rule. ML theory says to divide the dataset into three parts (training, validation and test) and the model is built with the training data, maximising the accuracy on validation and then seeing how it performs on the test (data the model has never seen).