Introduction to Machine Learning(ML)

Today, machine learning has become one of the most powerful technologies in the world. We cannot even find a place where we cannot apply machine learning. It is in every field and has spread across the whole world.

The most common example is that in Gmail we have features to classify emails as spam or not. Moreover, it can be used for Image Recognition, Voice Recognition, Prediction, Medical diagnoses and much more.

With time, we have seen how machines are becoming more and more intelligent with the help of ML (Machine Learning). ML has been applied to many fields today such as in Business, Health Care, Finance, Agriculture and several other sectors. So now let’s see the history of machine learning.

History of Machine Learning

The name of machine learning was coined in 1959 by Arthur Samuel while at IBM. He is an American pioneer in the field of computer gaming and artificial intelligence.

Here are some definitions for Machine Learning

• Arthur Samuel defines machine learning is a field of study that gives computers the ability to learn without being explicitly programmed
• In 1997, Tom Mitchell gave a “well-posed” mathematical and relational definition: A computer program is said to learn from experience E concerning some task T and some performance measure M, if its performance on T, as measured by M, improves with experience E.

Some examples of machine learning:

• Web search engines (Google search, etc)
• Image recognition is one of the most common uses of machine learning (Google image search, face detection in an image, character recognition, etc.)
• Spam filters
• Natural language processing (NLP)
• Self-customizing programs (like Amazon, Netflix product recommendations) and more.

Types of machine learning problems

There are two main types of machine learning problems; supervised and unsupervised learning.

Supervised

In supervised learning, you train the machine using well-labeled data where we want to get predictions based on some set of examples. It means our data is already tagged with the correct answers. It is like learning which takes place in the presence of a teacher or a supervisor.

For example, if you want a machine to help you to predict how long it will take to drive from your workplace to home. The machine will have official speed limits and recommended speeds, similar to speeds defined for road types, historical data pertaining to average speed data over certain time durations (this may be averaged data and sometimes at particular times of day), actual travel times from previous users, and real-time traffic information. Thus, based on these data, the machine will come up with the best prediction they can make.

From the above example, you can see the machine is already having labeled data to get the best prediction.

The supervised learning problems can be further grouped into regression and problems.

• Regression: This is when the output variable is a real value; e.g. dollars or weight etc.
• Classification: This is a problem defined when the output variable is a category, such as back or blue or disease and no-disease.

Classification and Regression Algorithms

Here are some algorithms used in supervised learning:

• Support vector machine,
• Neural network,
• Linear and logistics regression
• Random forest
• Classification trees.

When we are implementing programs, we need to use one or many algorithms to overcome a problem.

So now let’s see what unsupervised learning is.

Unsupervised

Unsupervised learning is not meant to supervise the model, instead, it is required to allow the model to work on its own to discover information. In this case, the machine will deal with unlabelled data.

Unsupervised machine learning finds all kind of unknown pattern in the data and it helps you to find features which can be useful for categorization. This is taken place in real-time where all data be analyzed and labeled in the presence of learners also it is easier to collect unlabelled data from a computer than labeled data which needs manual intervention.

Unsupervised learning problems can be further grouped into clustering and association problems.

• Clustering: This is where you want to discover the inherent groupings in the data; e.g. grouping customers by purchasing behavior.

Clustering Algorithm

• Association: An association rule learning problem is where you want to discover rules that describe large portions of your data, such as people that buy X also tend to buy Y.

Here are some algorithms for unsupervised learning:

• Cluster algorithms
• K-means
• Hierarchical clustering
• Apriori algorithm for association rule learning problems.

Supervised vs. Unsupervised Learning

After going through the above content, I hope you may have understood the basics of machine learning. Please give your feedback (positive and negative) about this article or just hit the clap button and share this post with friends and colleagues.

Thanks for reading!

Reference:

Supervised vs Unsupervised Learning: Key Differences

Introduction to Machine Learning for beginners