Machine Learning (ML): Integral Part of AI

Updated: Oct 16, 2021

Table of Contents

1. Introduction

2. What is Machine Learning (ML)?

3. Artificial Intelligence (AI) vs. Machine Learning

4. Importance of Machine Learning

5. Working of Machine Learning

6. 3 Types of Machine Learning Based on Learning Problems

7. Characteristics of Machine Learning

8. Key Catch ups


As we know, we are living in a world of humans and machines. Humans have been evolving and learning from past experiences for millions of years. On the other hand, the era of machines and robots have just begun now. We can consider it in a way that currently, we are living in the primitive age of machines while the future of machines is enormous and is beyond a scope of imagination. Now in today’s world, these machines or the robots have to be programmed before they start following our instructions; but what if the machine started learning on their own from their experiences, worked like us, felt like us, did things more accurately than us, they might even start a war of their own. Now, these things sound fascinating and little scaring. But let’s just not forget that it’s just the beginning of the new era. The best is yet to come.

Well, Machine learning (ML) is a concept which allows the machines to learn from examples and experiences and that too without being explicitly programmed. So, instead of us writing the code, what we do is feed the data to the generic algorithm and the algorithm or the machine will still logic based on the given data.

What is Machine Learning (ML)?

As per the experts, Machine Learning is a subfield of Artificial Intelligence that allows software applications to learn and become more accurate at predicting outcomes without being explicitly programmed. ML is one of the most exciting technologies that one would ever come across. As it is evident from the name, it gives the ability to computers to sense more similar to humans: The ability to learn. ML is actively being used today, perhaps in many more places than one would expect. For e.g. when we browse sites like Netflix or YouTube, we are recommended certain videos and shows to watch based on our interests. That’s Machine Learning! Companies like Amazon and Google are also creating smart speakers that utilize machine learning, e.g., Siri, to become our personal assistants.

In simple terms, Machine learning is the use of massive amounts of data to answer complex questions.

Artificial Intelligence (AI) vs. Machine Learning

Now, before going further, it is very important to clear very common misperception. People often think that Artificial Intelligence (AI) and Machine Learning are the same since they have common applications. For e.g. Siri is an application of AI Machine learning, so how are these technologies related? Artificial Intelligence is the science of getting machine to mimic behavior of humans. Machine learning is a subset of AI that focuses on getting machines to make decisions by feeding them data. So, to sum it up, AI and Machine learning are interconnected fields. Machine Learning is an Artificial Intelligence which provides a set of algorithms and neutral networks to solve data-driven problems. However, AI is not restricted to only Machine Learning. It covers a vast domain of fields including deep learning, natural language processing, object detection, computer vision, robotics expert systems and so on which we are going to look forward in further blogs.

Importance of Machine Learning

Machine Learning is evolving continuously. With evolution come a rise in demand and its importance. Machine Learning is important because its high value predictions that can guide better decisions and smart actions in real-time without human intervention. It helps analyze large chunks of data, easing the tasks of enterprises for over viewing trends in customer behaviors and business operational patterns, as well as supporting the development of new products. Many of today’s leading companies like Facebook and Google make ML a central part of their operations. Machine Learning has changed the way data extraction and interpretation works by involving automatic sets of generic methods that have replaced traditional statistical techniques giving a significant competitive differentiator for many companies.

Working of Machine Learning

Undoubtedly, ML is one of the most exciting subsets of Artificial Intelligence. It competes the task of learning of learning from data with specific inputs to the machine. For this, it is important to understand what makes Machine Learning work and how can it be used in upcoming time.

  • Data Collection and Preparation: Everything from choosing where to get the data, up to the point where it is clean and ready for feature and collection.

  • Feature selection and Feature engineering: This includes all changes to the data from once it has been cleaned up to when it is ingested into the machine learning model.

  • Choosing the ML algorithm and training our first model: Getting a “better than baseline” result upon which we can improve hopefully. In simple terms, choosing and developing a model that does better than baseline.

  • Evaluating our model: This includes the selection of the measure of our success as well as deciding on an evaluation process; seemingly a smaller step than others, but important to our end result.

  • Model tweaking, regularization, and hyper parameter tuning: This is where we iteratively go from a “good enough” model to our best effort.

3 Types of Machine Learning Based on Learning Problems

1. Supervised Learning

Supervised Learning is the one, where we can consider the learning guided by a teacher. We have dataset which acts as a teacher and it role is to train the model or the machine. Once the model gets trained, it starts making a prediction or decision when new data is given to it. For e.g., a photograph with a traffic sign, and the task is to predict the correct label or output, for example which traffic sign is in the picture (speed limit, stop sign, etc.). In the simplest cases, the answers are in the form of yes/no (i.e., binary classification problem).

2. Unsupervised Learning

This model learns through observation and finds structure in the data. Once the model is given a dataset, it automatically finds patterns and relationships in the dataset by creating clusters in it. What is cannot do is adding labels to the cluster, or reducing the data to small number of important ‘dimensions’. Data visualization can be considered as an unsupervised learning.

3. Reinforcement Learning

It is the ability of an agent to interact with the environment and find out what is the best outcome. It follows the concept of hit and trial method. The agent is rewarded or penalized with a point for a correct or a wrong answer, and on the basis of the positive reward points gained, the model trains itself. And again, once trained, it gets ready to predict the new data presented to it. It is commonly used in situations where an AI agent like a self-driven car must operate in an environment and where it gives feedback about good or bad choices that is available with some delay. It is also used in games where the outcome may be decided only at the end of the game.

Characteristics of Machine Learning

  • Automation of everything: It is the ability to automate various decision-making tasks. It is responsible for cutting the workload and time. This automation is the reason that ML is very reliable and helps us to think more creatively. For e.g., some common use we see in our daily life is social media sentiment analysis and chat bots. The moment a negative tweet is made related to a product or service of a company, a chatbot instantly replies as first-level customer support.

  • Wide range of applications: It means that we can apply ML on any of the major fields. ML has its role everywhere from banking, defense, education, business and medical to science and technology. This helps to create more opportunities.

  • Scope of improvement: ML is constantly evolving with time. There is a lot of scope in ML to become the top technology in future due to its adequate research areas. This helps us to improve both hardware and software. Giants like Amazon, Flipkart, etc. collect a huge amount of new data every day. The accuracy of finding associated products or recommendation, engine improves with this huge amount of training data available.

  • Multi-dimensional and multi-variety data handling: ML plays the biggest role when it comes to data handling at this time. It can handle any type of data. It can process and analyze any type of data that normal systems can’t. Data is the most important part of any Machine learning model.

Key Catch ups

  • Machine Learning is a subfield of Artificial Intelligence. Instead of relying on explicit programming, it is a system through which computers use a massive set of data and apply algorithms to ‘train’ and ‘teach’ themselves and make predictions.

  • ML system enables us to quickly apply knowledge and training from large data sets to excel at facial and speech recognition and many more tasks.

  • There are 3 type of Machine Learning based on learning options: Supervised Learning, Unsupervised Learning and Reinforcement Learning.

  • ML touches industries from government to education to healthcare. It can be used in business, medical, social media, customer services, self-driven cars and many more. It is now regarded as core tool for decision making.

  • Businesses like IBM, Amazon, Microsoft, Google and others offers tools for machine learning. There are free platforms as well.

About Author

This blog is written by Anushka Jain. She is pursuing BCA and have a strong interest in arts, painting and writing the blogs.

30 views0 comments

Recent Posts

See All