Hi, Hash Noders!

A lot has been going around Machine learning- its use cases, can this replace humans? Is it injurious?

Blah Blah Blah!

But very few are aware of its two components ie supervised and unsupervised learning, even some of whom know aren’t aware of their differences.

We at Purple Olive Labs , built some interesting features using ML, which include predicting stock market movements using huge dataset.

Now I 'll get rich!!

I have been working on ML for a decade, did a lot of research, and got some fascinating results. Here I would discuss some, but not everything.

Time constraints you know!!

What is Supervised Learning

An easy way to begin grasping the concept of supervised learning is by looking directly at the words that make it up. Supervised means to observe and direct the execution of(a task, project, or activity).

Obviously, we aren’t going to supervise a person? Instead, we ll be supervising a machine learning model that might be able to produce classification regions as we see here.

So how do we supervise a machine learning model?

We do this by teaching the model.

That is, we load the model with knowledge so that we can have it predict future instances. But this leads to a new question, which is, how exactly do we teach a model? Generally speaking, the model is trained on a labeled dataset, so it can predict the outcome of out-of-sample data.

We teach the model by training it with some data from a labeled dataset. It’s important to note that data is labeled. And what does a labeled dataset look like? Well it can look something like this

This example is just taken from the Iris dataset which is a famous dataset used for machine learning.

Let's start by classifying some components of this table. The names up here- Sepal length, Petal Length, Petal width, and Species are called Attributes. The columns are called features, which include the data. If you look single data point on the plot, it 'll have all these attributes. That would make a row on this chart, or an observation. Looking directly at the value of data, you can have 2 kinds. The first is numeric, the most commonly used data when dealing with machine learning.

The second is categorical? That is, it's non-numeric, because it contains character, because it contains characters rather than numbers. In this case, it is categorical because this dataset is made for classification. Usually, a dataset like this will be put into a .csv file or comma-separated value file. This file separates observations by new lines, and attributes by comma (hence comma separated)

There are 2 types of supervised learning- classification and regression.

Since we know the meaning of supervised learning, what do you think unsupervised learning means?

Unsupervised learning

Its exactly as it sounds. We do not supervise the model, but we let the model on its own to discover information that may not be visible to the human eye. They use machine learning algorithms that draw conclusions on UNLABELED data.

They have more difficult algorithms than supervised learning since we know little to no information about the data, or the outcomes that are to be expected. With unsupervised learning, we are looking to find things such as groups/clusters, perform density estimation, and dimensionality reduction.

Whereas in supervised learning we know what data we are dealing with since it labeled data. Moreover, unsupervised learning has fewer tests and models to ensure the outcome is accurate. Due to this unsupervised learning creates a less controllable environment, as a machine is creating an outcome for us.

One of the best examples of unsupervised learning is Kohonen's self-organizing maps Its main characteristics are: It transforms an incoming signal pattern of arbitrary dimension into one or 2-dimensional map and performs this transformation adaptively The network represents a feedforward structure with a single computational layer consisting of neurons arranged in rows and columns. At each stage of representation, each input signal is kept in its proper context and, Neurons dealing with closely related pieces of information are close together and they communicate through synaptic connections.