The first step to Machine learning…

Simply, Machine learning is a subset of Artificial intelligence(AI) and data science, which uses data and algorithms to imitate human learn.

Let’s see, how we can imitate human learning using our machines.

In traditional programming, we feed some data and program logic, run it on the machine and get output. But in machine learning, we feed data and output, run it on the machine ( we call this process as training) during training, and create its own program and evaluate this programmer using testing. So this generated program is called a Model.

How do we create our own MODEL?

Well… To build our own model, we must first discover a sufficient data set and then select an appropriate machine-learning technique. The algorithms we use are determined by the type of data we have and the task we are attempting to automate. After that, we’ll clean up our data and divide it into two sections: training and testing ( I will discuss more data pre-processing on my future blogs). We utilize the training set to train our algorithm and the test set to evaluate our model.

Machine Learning algorithms types…..

Apart from that machine-learning algorithm implementations are classified into three major categories, Depending on the nature of the learning system which is as follows:

Supervised learning

This approach is similar to human learning under a teacher. The teacher gives good examples for students to memories. Then the students derive general rules from these specific examples. In supervised learning, the model learns from example data and associated target response that can consist of numeric values or string labels, such as class or tags, later the model predict the correct response when new data comes according to the given target response.

Unsupervised learning

This type of algorithm learns from examples without any target response, leaving the algorithm to determine the data patterns independently. This type of algorithm tends to restructure the data into something else, such as new features that may represent a class or a new series of un-correlated values.

Reinforcement learning

This is similar to how we learn a new skill with limited understanding so that we can gain experience while performing the activity. Like this, we need to provide some example data to an algorithm with a lack of labels as in unsupervised learning. Then we can accompany an example with positive or negative feedback according to the solution we made.

here is a brief idea of the Ml learning approaches As well as The choice of algorithms depends on what type of data do we have and what kind of task we are trying to automate.

In the next blogs, I am to discuss these learning types and the Ml algorithms in more detail.

Thank you for the reading, hope you get some idea about Machine Learning.