- Machine Learning
- Cyber Forensics
- Full Stack Developer
- Network Design
Answer – Option (A) Machine Learning
Machine Learning is the correct answer to the question “Which of the following is the branch of artificial intelligence”. There are many branches also known as sub-areas of Artificial Intelligence such as Machine Learning, Computer Vision, Natural language processing, and others. Let’s get a better understanding of the answer by learning more about Machine Learning.
What is Machine Learning?
Machine learning instructs a computer on how to perform a specific task without explicitly programming it to perform that said task. Instead, you can directly feed data into an algorithm to slowly but surely improve the results with experience similar to how we as a human learn.
You can now find predictive models that are placed in many of the products we use every day. In short, they perform two basic jobs, one is to classify data and the other is to make predictions about future outcomes like will the stock will go up, etc.
Python is the language of choice among data scientists but “R” and “Julia” are also popular options and there are many supporting frameworks out there to make the whole process easy.
Steps Involved In Machine Learning
The first step in the process is to get the data you need and clean up that data. Additionally, you will need a lot of data because the better the data represents your problem the better the results you get. The data needs to have some kind of signal to be valuable to the algorithm for making predictions.
Data scientists do a task called feature engineering. This is to transform all the raw data into features that better represent what the underlying problems are.
The next step is to separate the data into a training set and a testing set. The training data goes into an algorithm to build a model, and then that exact testing data is used to see as well as check the accuracy or error of the model.
After that, the next step is to choose an algorithm which might be a simple statistical model like linear or logistic regression or a decision tree. A decision tree sets different weights to features in the data.
You can also get fancy with a convolutional neural network which is an algorithm that also sets weights to features but also takes the input data and creates additional features automatically. This is very useful for data sets that contain things like images or natural language where manual feature engineering is almost impossible.
All of these algorithms learn to get better by comparing between their own predictions to an error function. If you give it a classification problem like Is this animal a cat or a dog then the error function might be “accuracy”. And if it is a regression problem like how much will a loaf of bread cost next year then it might be a “mean absolute error”.
The end result of the machine learning process is a “model” which in simple terms is just a file that takes some input data in the same shape that you trained it for and then spits out a prediction that tries to reduce the error that it was optimized for. You can then use it on an actual device as well as deploy it into the cloud to build a real world product out of it.