Machine Learning and Recent Updates on the Field
Today’s subject is Machine Learning. Although it seems like a very technical term, it isn’t really. On this article, I will try to explain Machine Learning and its recent usages along the field.
What is Machine Learning
Machine learning is a field of study that is interested in development and study of the statistical algorithms that can learn from data and discovers new patterns and trends.
Machine Learning (ML for short) finds its usage in natural language processing, computer vision, speech recognition, email filtering, agriculture, and medicine (Wikipedia, 2024). It is mostly used for pattern recognition on big data and these are the usage areas that is branched from this specialty.
Machine Learning History
Although the term “Machine Learning” was coined in 1959 by Alfred Samuel, an IBM employee and invented a program to calculate winning chances on checkers, history of machine learning goes back to 1949 (Wikipedia, 2024). When Donald Hebb published his book “The Organization of Behavior”, he explained the theoretical neural structure formed by neurons that works together. This was the groundwork for machine learning.
After Donald and Alfred, (Kononenko, 2001) three different branches appeared. Classical work in symbolic learning is described by (Hunt E., 1966), in statistical methods by (Nilsson, 1965) and in neural networks by (Rosenblatt, 1962). Through the years, they have emerged as one branch and created methods such as; statistical or pattern recognition methods, such as the k-nearest neighbours, discriminant analysis, and Bayesian classifiers, inductive learning of symbolic rules, such as top-down induction of decision trees, decision rules and induction of logic programs, and artificial neural networks, such as the multilayered feedforward neural network with backpropagation learning, the Kohonen’s self-organizing network and the Hopfield’s associative memory.
Machine Learning Usage Areas
Machine learning, due to its flexibility and handling big data founds lots of usage areas to work. Usage areas can be considered as, developing practical software for computer vision, speech recognition, natural language processing, robot control, and other ap- plications (Jordan & Mitchell, 2015). The machine learning and recent updates on the fields effected the computer science and across a range of industries concerned with data-intensive issues, such as consumer services, the diagnosis of faults in complex systems, and the control of logistics chains.
Do you use Siri on your iPhone a lot? Well, you can thanks to Machine Learning for that. Because of handling big data that comes from human- computer interaction, Machine Learning can work its magic. Most highlighted specialty of Machine Learning is that it keeps learning during the process. When you are talking with Siri or talking with ChatGPT, you are training the system during process. It means constant feedback and learning loop.
Recent Articles and Development on the Machine Learning
Surprise! You don’t have to teach something to the Machine by constant feedback loop in the current situation. According to a research that is published in Nature, software get better at detecting Covid-19 by just watching YouTube videos of people coughing, breathing and throat clearing. (Lenharo, 2024).
Another surprise! They have put a camera to a child and trained the model for pairing video frames with co-occurred with uttered words, after that, AI model started to understand a variety of words around it. Also, it started to build a relationship between words. Although some scientist’s states that people are born with built-in expectations and logical constraints that make creation of relationship between words possible. However, this study confirms that it is not the case.
Surprise (not with exclamation mark because you must get used to it). As we have said, Machine Learning can be used for identifying patterns on the big data. A study published in Nature (Avelle & Marco, 2024) analyzed 500 million threads, posts and conversations across eight platforms over 34 years. And publishers found by the help of Machine Learning, toxicity is not result of social media but humans. Also, one more surprise, toxicity does not diminishes that media’s attractiveness. This means that it is not important that media is toxic or not, if it is attractive to us, we are staying attractive all the time.
Conclusion
In conclusion, machine learning used a lot in different areas that I couldn’t count them on this small article. However, the earlier the better. You need to get familiar with AI and Machine Learning in order not to be stay behind and look to AI as similar as a technological God. It is no God. It is merely learning from a big dataset and makes some assumptions through it. We will going to use it until more advanced or more useful technology replace it. But getting familiar with it and using it in another fields will going to increase technological advances and usages.
References
Wikipedia. (2024, 08 02). Wikipedia. Retrieved from Machine Learning: https://en.wikipedia.org/wiki/Machine_learning
Kononenko, I. (2001). Machine Learning for Medical Diagnosis: History, State of the Art and Perspective. Artificial Intelligence in medicin.
Hunt E., M. J. (1966). Experiments in Induction. Academic Press.
Nilsson, N. (1965). Learning Machines. McGraw-Hill.
Rosenblatt, F. (1962). Principles of Neurodynamics. Washington, DC: Spartan Books.
Jordan, M. I., & Mitchell, T. M. (2015). Machine learning: Trends, perspectives, and prospects. Science.
Lenharo, M. (2024). Google AI could soon use a person’s cough to diagnose disease. Nature.
Avelle, M., & Marco, N. D. (2024). Persistent interaction patterns across social media platforms and over time. Nature.
