Why do we use sampling theory in machine learning | Data Science Interview Questions and Answers

Thinking Neuron · Advanced ·📐 ML Fundamentals ·4y ago

About this lesson

https://thinkingneuron.com/what-is-the-use-of-sampling-theory-in-data-science/ Sampling means choosing random rows from a dataset. Sampling theory says, If you select the rows randomly then the selected subset of the data represents the whole data. What is Training data? A major part (usually 70%) of data is randomly selected from the full data. This chunk of data is used to train the predictive model. These are the examples that the predictive model uses to learn the patterns. What is Testing data? That part (rest 30% ) of full data which was NOT selected in Training data. This chunk of data is used to TEST the predictive model for its performance. These are the examples that are UNSEEN by the predictive model. Hence, these are used to test the accuracy by comparing the predicted values with the original values.

Original Description

https://thinkingneuron.com/what-is-the-use-of-sampling-theory-in-data-science/ Sampling means choosing random rows from a dataset. Sampling theory says, If you select the rows randomly then the selected subset of the data represents the whole data. What is Training data? A major part (usually 70%) of data is randomly selected from the full data. This chunk of data is used to train the predictive model. These are the examples that the predictive model uses to learn the patterns. What is Testing data? That part (rest 30% ) of full data which was NOT selected in Training data. This chunk of data is used to TEST the predictive model for its performance. These are the examples that are UNSEEN by the predictive model. Hence, these are used to test the accuracy by comparing the predicted values with the original values.
Watch on YouTube ↗ (saves to browser)
Sign in to unlock AI tutor explanation · ⚡30

Related Reads

Up next
Arrays vs Lists: What AI Actually Prefers | Common Tech Interview Questions
SCALER
Watch →