November 7, 2023

Kathlyn Phan

11th Grade

Fountain Valley High School

Have you ever used ChatGPT before? How about a customer service chatbot? You may have interacted with artificial intelligence before at some point. In doing so, you might have unintentionally helped further develop one of the fastest-growing fields in recent years. Machine learning has rapidly become an in-demand skill because of how versatile, powerful, and continuously self-improving it is. Although some might question how trustworthy a computer is in analyzing mass amounts of data, machine learning has proven its worth over the past few years by examining and recognizing algorithms that help improve other fields such as marketing, business, education, and much more.

In 1949, Donald Hebb created a brain-cell interaction model in his book, “The Organization of Behavior”, which would soon inspire the creation of machine learning. The model influenced the idea of machine learning because there were similarities between how neurons and a computer program could interact and communicate with themselves to trigger specific actions to occur. However, it was not until 1959 that the term “machine learning” was coined by Arthur Samuel when he created a computer program that calculated the chances of each side winning during a checkers game. The program would remember the positions of the game pieces from past games that it analyzed and store them to use them as a reference to score the probability of the winning side.

Machine learning is a branch of artificial intelligence (AI) that recognizes patterns, handles large amounts of data, and makes decisions based on the algorithm. The main appeal of machine learning is how it is continuously improving itself to produce more accurate results as it is fed more data and feedback. The “learning” in its name comes from how easily adaptable it is to complete a variety of tasks. Within machine learning, there are three types: supervised learning, unsupervised learning, and reinforcement learning. 

Supervised learning is the most popular form of machine learning because it is fed historical data to use as input and desired results to use as output. From there, the program learns to develop an algorithm to consistently predict and provide the desired outcome. It commonly uses neural networks, decision trees, linear regression, etc. Unsupervised learning is also provided with data to use as its input, but there is no specific desired output. This type of machine learning is best for making predictions and identifying patterns within a dataset. Reinforcement learning is different from the other two because it is used in a dynamic setting where it has to interact with its environment to come up with the desired output. Out of the three types of machine learning, reinforcement learning is the most similar to human learning because it simultaneously takes in data without any guide and adapts its algorithm to progressively improve itself.

Machine learning is a powerful tool that is used in numerous industries to automate tasks. For instance, it is used in healthcare, natural language processing, data analysis, and finance. In the healthcare field, machine learning helps with disease diagnosis and treatment recommendations. The program has a vast amount of historical medical data recorded, so it can recognize the symptoms of a disease and use it to make an educated prediction of the patient’s disease. From there, the program can recommend to the patient what treatment or medicine to take in order to treat their disease. Chatbots are a great example of the application of machine learning in natural language processing (NLP) because their algorithm processes language-based inputs to understand what the desired outcome is with contextual understanding. Chatbots are widely used by companies as customer service because they aid with answering frequently asked questions that do not necessarily need a human to answer. Machine learning is mainly used in data analytics and finance to make predictions based on statistics, trends, and complex datasets. It uses trends in financial markets to assess risk, detect fraud, and score credit. By automating tasks for companies, businesses are able to be more productive and efficient with little to no human intervention.

Although machine learning is reliable and convenient, some security pitfalls are potentially dangerous. An example of this would be model stealing where attackers steal the learning model and make a replica of it. This is dangerous because the learning model is created from sensitive information comprised of customers’ personal and financial information. Data is the most essential part of how a machine learning program can perform, so data privacy is crucial to prevent corruption and ensure people’s sensitive information is protected from hackers.

As computer programs become more and more powerful, they are simultaneously becoming more independent and accurate in their performance. In the future, we can expect machine learning to continue improving its performance and replace a large portion of the workforce. Machine learning plays a pivotal role as a tool in countless fields outside of computer science from human resources and communications to finance and automotive.