• Home
  • /
  • Education
  • /
  • Data Science Bootcamp Supervised Learning benefits

Data Science Bootcamp Supervised Learning benefits

Learning is a Type of Technique in which you state that given a data in the past, that there are attributes you have something called a tag.

Learning as a child

So, for you, like you was a kid Learning how to identify various sorts of fruits. This fruit is visually looked at by you and you know what an apple looks like. You form a perception around it. And someone taught you that is an apple. Similar is the case with other fruits orange, for instance a banana and so forth. So this visual perception you have learnt as a child and the other help you have got from someone else told you this visual perception of yours is an apple. This is what is referred to as learning.

data science bootcamp singapore

The input to Supervised Learning

There’s an input attribute to your Perception that is more around the color, shape and the arrangement of the fruit and someone else telling you that this type of a thing is something known as an apple. Over a time period, in spite of the sort of form and color and textures of different kinds of apple, you will have the ability to recognize that this is an apple. So, no matter how different tricks you do, regardless of how nature plays out later on too in coming out with new kinds of apples, your perception is extremely strong when it comes to identifying an apple because someone has coached you on that. And this is what happens in a machine learning model.

Training and Accuracy needed

You train yourself for data science bootcamp singapore about anything and based upon which you have got this tag and a label is what tells you that this is an apple. Remember here that because we are training someone on what that thing is that you should be very cautious that if you curate a data set for a supervised machine learning algorithm that your information should be 100% right. Even in the event that you miss out on 10% of data place where you feel the tagging is not right, anticipate that 10% as a mistake in output too. Your model is like your data in terms that are easy.