Supervised learning in machine learning refers to a category of machine learning. In the previous section we have discussed many categories.
We have seen one key feature of machine learning is to train a machine based on a training dataset. Right?
However, for doing that we need an algorithm, of course.
As a result the machine can classify the data and predict an accurate outcome.
Basically in supervised learning, we have a set of labeled dataset that trains the algorithm so that it can classify data and predict an outcome.
Therefore first we need training data. Second we need to feed the data to the algorithm.
However, we use labeled dataset and the algorithm helps us to build the model.
We have discussed classification in the previous section.
Although not in detail yet it will give us an idea of how we can use supervised learning for the classification.
But classification is one of the examples. Another example is regression.
If spam filters are one of the examples of classifications, for regression we need other types of examples.
The classification algorithm uses many email examples to classify the data and then predict what is spam email and what is not.
But to do that we need human intervention always. Because users move the emails to spam folders and based on that dataset the algorithm trains itself to classify the spam emails.
On the contrary, consider an old car. The age might become one of the features that a set of dataset consists of.
It’s because as a car gets older, its mileage drops down, right?
Therefore, we can call this feature like age, or mileage as a predictor.
Why?
Because it helps us to predict the price.
This is a good example of regression. We have a dataset with labels and predictors.
Now whether it’s classification or regression, we find a common pattern.
Datasets act as a teacher that provides supervision to the algorithms. Right?
If you are a complete beginner your journey to learn TensorFlow might start from here.
For the TensorFlow beginners we have a dedicated category – TensorFlow for Beginners.
But besides that, you may need to learn several other machine learning and data science libraries.
As a result, you may check these categories as well – NumPy, Pandas, Matplotlib.
However, without learning Python, you cannot learn the usages of these libraries. Why? Because they all use Python as the Programming language.
Therefore please learn Python at the very beginning and start learning TensorFlow.
And, finally please check our Mathematics, Discrete Mathematics and Data Structures categories specially. We have tried to discuss from basic to intermediate level so that you can pick up the core ideas of TensorFlow.
Don’t worry, as we progress, we will discuss regression in a separate section.
Before that let’s try to understand the basic concept of supervised learning first.
So far we have learned a few important facts about supervised learning in machine learning.
There are inputs and outputs.
Whether a user moves a few emails to the spam folders, or a car is old and cannot give a good mileage, these are all treated as inputs.
We may call them predictors, label them and finally the algorithms learn from these inputs and based on this learning it either classifies them, or predicts some outcome.
It also answers another key question. Why do we need machine learning algorithms as a whole?
We cannot create datasets manually. It’s not only tedious but it is very time consuming as long as we think of big data.
In some cases, collecting data is easy, 😂 but there are cases when it involves difficulties 😡.
Leave a Reply