The Data Daily

What is Machine Learning?

Last updated: 11-08-2019

Read original article here

What is Machine Learning?

This article gives you a brief introduction that will help you understand what is Machine Learning. I will start by giving its overview and some definitions from different sources. Then, I will discuss what are the types of machine learning. How Machine Learning works with simple examples, and will also explain how Machine Learning is being used in various industries. So lets dive into it…

Machine learning is a fundamental subdivision field of artificial intelligence. It is the science of getting computers to learn and act like humans do, and improve their learning over time in autonomous fashion, by feeding them data and information in the form of observations and real-world interactions. When exposed to new data, these computer programs are enabled to learn, change, develop, and grow by themselves. In short; more the data, higher will be the accuracy to learn and predict the results.

There are multiple definitions exists of machine learning depending upon the understanding. So lets see some of the top definitions about machine learning:

According to McKinsey & Co.:

and the simple one by Stanford:

Machine learning implementation types can be classified into three different categories Supervised, Unsupervised and Reinforcement learning depending upon the nature of the data it receives. Now lets discuss each of them.

Supervised learning is the process like you are learning under someone’s supervision. In supervised learning, the process of an algorithm learning from the training dataset can be thought of as a teacher supervising the learning process. The correct answers are known, the algorithm iteratively makes predictions on the training data and its been corrected by the teacher. The learning phase continue to progress until algorithm achieves an acceptable level of performance. In supervised learning, data is given with associate labels.

Supervised learning problems can be further grouped into regression and classification problems.

Classification Problem: We are given an input, for example a fruit basket filled with lots of fruits, and the task is to predict the correct output or label, based on the previous knowledge we have, for example which fruit is in the picture (apple, banana, orange, etc.). In the simplest cases, the answers are in the form of yes/no (we call these binary classification problems). For your ease, see below screenshot to understand the complete picture:

Here, Fruit size, color, shape and name are the labels of the dataset.

Regression Problem: A regression problem is when the output variable is a real value, floating point value or in a continuous quantity, such as ‘age’ or ‘weight’. A more realistic machine learning example is one involving lots of variables, like an algorithm that predicts the price of an apartment in different areas in Karachi based on square footage, location, safety and proximity to public transport.

In Unsupervised learning, the information used to train is neither classified nor labeled. The goal for unsupervised learning is to model the underlying structure or distribution in the data in order to learn more about the data to form “clusters”, or reducing the data to a small number of important “dimensions”. There is no correct answers and there is no teacher. Algorithms are left to their own devises to discover and present the interesting structure in the data. Data visualization can also be considered unsupervised learning.

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

Clustering Problem: A clustering problem is where you want to discover the inherent groupings in the data, such as grouping customers by purchasing behavior. Let’s understand the concept with clustering genders based on hair length example. To determine gender, different similarity measure could be used to categorize male and female genders. This could be done by finding the similarity between two hair lengths and keep them in the same group if the similarity is less(Difference of hair length is less). The same process could continue until all the hair length properly grouped into two categories.

Association Problem: Association rules are if/then statements that help uncover relationships between seemingly unrelated data in a relational database or other information repository. An example of an association rule would be “If a customer buys a dozen eggs, he is 80% likely to also purchase milk.” An association rule has two parts, an antecedent (if) and a consequent (then). An antecedent is an item found in the data. A consequent is an item that is found in combination with the antecedent.

It is a learning method that interacts with its environment by producing actions and discovers errors or rewards. It is also known as reward base learning or we can say it works on the principles of feedback. Lets say you have provided the image of an apple to the machine and then system identifies it as a ball which is wrong, so you provide the negative feedback to the machine saying that its an apple image. So machine will learn from the feedback and finally if it comes across of any image of an apple, it will be able to classify it correctly. That’s what reinforcement learning is all about.

Machine learning in healthcare is one such area which is seeing gradual acceptance in the healthcare industry. Google recently developed a machine-learning algorithm to identify cancerous tumors in mammograms, and researchers in Stanford University are using deep learning to identify skin cancer. Machine Learning (ML) is already lending a hand in diverse situations in healthcare. ML in healthcare helps to analyze thousands of different data points and suggest outcomes, provide timely risk scores, precise resource allocation, and has many other applications.

The video surveillance system nowadays are powered by AI that makes it possible to detect crime before they happen. They track unusual behavior of people like standing motionless for a long time, stumbling, or napping on benches etc. The system can thus give an alert to human attendants, which can ultimately help to avoid mishaps. And when such activities are reported and counted to be true, they help to improve the surveillance services. This happens with machine learning doing its job at the backend.

Experts predict online credit card fraud to soar to a whopping $32 billion in 2020. That’s more than the profit made by Coca Cola and JP Morgan Chase combined. That’s something to worry about. Fraud Detection is one of the most necessary Applications of Machine Learning. The number of transactions has increased due to a plethora of payment channels — credit/debit cards, smartphones, numerous wallets, UPI and much more. At the same time, the amount of criminals have become adept at finding loopholes. Whenever a customer carries out a transaction — the Machine Learning model thoroughly x-rays their profile searching for suspicious patterns. In Machine Learning, problems like fraud detection are usually framed as classification problems

Social media services utilizes machine learning for their own and user benefits. You have seen one of the most common applications of Machine Learning is Automatic Friend Tagging Suggestions in Facebook or any other social media platform. You have also noticed People You May Know Suggestion in Facebook where it shows some of those person profiles that you can become friends with. It is done by continuously notices the friends that you connect with, the profiles that you visit very often, your interests, workplace, or a group that you share with someone etc. Both these suggestion features are using machine learning behind.

Businesses organizations that are in the retail industry or e-commerce companies have been using advanced machine learning applications including Recommendation systems, Chat-bot applications, Predictive Analytics system, etc. to innovate and enhance their business processes. A number of big Retail and E-commerce industries like Walmart, Amazon, Alibaba, Flipkart have successfully incorporated AI and Machine Learning technologies across their entire sales cycles from logistics to sales to post-sales services, thus improve results as well as business processes.

To summarize this article, following are some important takeaways from this article:

I hope you get a rough idea about what machine learning is and I’m sure that you will explain it to someone else easily. Please comment your feedback as it will help me to expand my skills further. Thank you!

See you soon…

Read the rest of this article here