In the beginning, God created heaven, earth, and natural intelligence…
Let us imagine a world where cars drive themselves, a world where computers predict your stocks, a world where your phone understands your mood, a world where your phone and computer can keep up chatting with your friends in your absence. This kind of world is a Machine Learning driven world.
The first ever definition of Machine Learning was by Arthur Samuel where he explained machine learning as the ability of computers to perform intelligent tasks without being explicitly programmed. ML’S exploration became a great deal decades ago and got its magnanimous outburst since the inception of Data Driven Disruption.
However, Machine Learning is a subfield of Artificial Intelligence as well as other data driven fields such as computer vision, IOT, deep learning and so on. Machine Learning can be categorized into two main learning categories, which are Supervised Learning, and Unsupervised Learning.
Let’s take a brief overview of the categories before we dive straight into our first lesson in machine learning.
1. Supervised Learning: This is a type of learning where a computer algorithm is fed with inputs (training sets) and expected outputs (results), so that it can learn the correlation between the two-dataset variables such that it understands the relationship and can give accurate results on its own. Imagine a teacher teaching a boy and she teaches him the multiplication table, the point where she stops teaching the boy is when she gets a satisfactory level of performance from the boy. The same way learning stops when the algorithm achieves a satisfactory performance level.
There are two main categories of Supervised Learning, which are:
· Regression: This are always implemented when the problem is a regression problem such that the output(result) is an expected variable such as “Height”, “Age”, “Amount”, “price” etc.
· Classification This has to do with a classification problem such that the expected output is a categorical set such as
Examples of algorithms under regression are Simple Linear Regression, Multiple Linear Regression, and Polynomial Regression etc. While algorithms that are used in Classification includes Random Forest, Support Vector Machines (SVM), Decision Tree and so on.
2. Unsupervised Learning: This is type of learning where a computer algorithm is fed with inputs (training sets) without any outcome (results). In this kind of learning model, the algorithm is allowed to figure out the dataset and look for important structures and trends in the data. Just like giving a student set of balls and watch as the student sorts it into “big”, “medium” “small”.
Unsupervised learning has two main categories, which are:
· Clustering: This is based on categorizing datasets based on some intrinsic characteristics and grouping of the data set. For example, grouping football players based on their playing style. Algorithms that fall under this space include K-means, Hierarchical clustering etc.
· Association rule: This is based on learning model that focuses on the discovery of trends in a large data set. For example, determining the trend of people that drive Mercedes and own a duplex in a city. Algorithms relevant here include Apriori algorithm, Eclat and so on.
I trust you enjoyed reading this piece. This is just the foundation of applied machine learning. Watch out for my next article on “Data Preprocessing”.
For practical session with me kindly endeavor to download Anaconda and install Spyder. Spyder will be our main programming interface in this MACHINE learning tutorials.
Have fun and don’t stop learning!
Written by; Raji Adam Bifola (MCP, MCSA). Data Scientist/BI Analyst at Techspecialist Consulting Limited