Is Machine Learning Always the Right Tool?
Let’s stop narrowing our possibilities by focusing only on one part of data-science.
You have probably heard of Machine Learning [ML] (e.g. neural networks, deep learning). Promising approach, based on neuroscience and statistics, in solving real world problems. It could be said that the last couple of years was the rise of Machine Learning, and it still is. Researchers from various fields are applying it to explore the extension of its value. Also, the development of accessible tools, in open-source programming languages, such as Tensforlow and PyTorch made it easy to use ML. Instead of hard coding and understanding the math behind it, we can use these handy packages.
It is worth to mention that not only researchers are experimenting with ML, but also a majority of companies which are active in data-science and artificial intelligence are emphasizing the usage of ML towards their clients. Although, sometimes it seems like ML became their sales pitch, ML turned into some big hype, big trend which you do not want to miss out. Regardless you are familiar with it or not, as long as you provide ML services you are a company which provides services in data science. If your clients make use of it, they can call themselves data-driven organizations. Which has led to the question, did we as data-science companies become too obsessive with ML? Is there…
