Applying Mathematics — Data Science & Machine Learning
Whether you call it "math", "maths" or the traditional way of calling it "mathematics", if you have an average or above minimal knowledge of counting numbers, then you already possess a skill to solve and analyze problems.
Being a mathematics major, your opportunities are immensely numerous in terms of job and career advancements. From careers in accounting, banking, actuarial science, meteorology, teaching, financial analysis, statistics, investment analysis, research (math), data science, and machine learning engineering; No one would argue that, right?
My drive point is about pursuing a career in Data science and Machine learning engineering because these two are one of the world’s most paying jobs today. I’ll take my time to break it down to the simplest of understanding.
If you are studying mathematics or you already have a major in mathematics, Let’s ride! as I take you through the basics of both fields. If your field isn’t related to mathematics but you can easily add numbers like a pre-school kid, you’re also good to go! Not all data scientist and machine learning engineers are mathematicians, just know you’d have to work on increasing your mathematical skills overtime.
You see, both fields are subdomains under artificial intelligence. They are the key components that make AI functions. Knowing mathematics is the bedrock of all sciences, almost all the techniques of data science and machine learning have a deep mathematical foundation.
Data science is the study of data. It involves developing methods of recording, storing, and analyzing data to effectively extract useful information. The goal of data science is to gain insights and knowledge from any type of data — both structured and unstructured.
This is a fully finished/handpicked data that fits nicely into a database. It is highly organized and easily analyzed. Most IT guys are used to working with this kind of data. When you think of structured data, it’s just like a kind of data that fits nicely into an Excel Spreadsheet. For example, dates, phone numbers, addresses, customer names, product names, POS transaction information, etc.
The beauty of it is its inherent structure and orderliness makes it simple to understand and analyze.
It’s just the opposite. It doesn’t fit nicely into a spreadsheet or database. It can be textual or non-textual. It can be human or machine-generated. Examples are media files (audio, images, and video), text files, emails, data from social networks, mobile data (text messages and GPS locations), and communications (chats and calls). These examples are largely human-generated, but machine-generated data can also be unstructured, like satellite images, scientific data, surveillance images and video, weather sensor data, etc.
Hey you, I know you getting bored already. Haha. Open your mind as I take you through the evolution of data science.
Evolution of data science
Data Science has been around for a very long time, it started in the early 20th century. But it wasn't called data science in those times, it was usually called computational analysis.
The reason it was named data science was because of BIG DATA.
The term “Big data” came to live due to the fact that there was a skyrocket in the usage of internet services in 2010. These data were too enormous to the extent that Internet server companies recorded over 2 zettabytes ( 1 trillion gigabytes) of data. Woah! Truly Big data.
This field needed people to sort this enormous data. So, big tech companies like Google, Facebook, IBM, Apple, Microsoft, Intel, Cisco, and Oracle began to leverage and exploit these enormous data to improve their businesses, customer service and obviously to make profits.
Ever wondered Google always has answers to most of your day-to-day questions? It’s Big data. Big data is also used to get insights for Internet marketing and customer needs. You’ll understand better when we get to machine learning.
Being a Data Scientist
The process of learning to become a data scientist is not easy. You must know how programming languages work and how to run codes. I’ve heard people say data science doesn’t need programming and you just need to know how to use Excel spreadsheet, that’s a big fat lie. When you start going deep into data science you’d understand the need of learning how to program.
Here are the most used programming languages in data science you can learn:
Data science also needs applied math. You might know the cores of mathematics to some extent, but if you don’t know how to apply it, it’s just useless. To be a very good data scientist, you must be good at linear algebra, discrete math, differential equations, numerical analysis, abstract algebra, number theory, real analysis, complex analysis, and topology. That seems bogus, right? It isn’t Compulsory you know all these at the start. Over time, you’d have to learn all these and get used to it.
Lest I forget, you need to be quite good at statistics too. As you might know, statistics deals with collection, organization, analysis, interpretation and presentation of data. Statistics is essential. Nevertheless, the most important skill in data science is critical thinking.
Working as a data scientist, no one really cares where the source of your data comes from, we just want a data that is authentic and can be understood by all, only if you're being given a source of data to work on. You see, data is infinitesimal, it has no boundaries.
Data science is wider than this basic explanation I’ve just given. One of the best ways to thrive as a data scientist is to be able to communicate insights using data, mostly for non-technical people. This way, you can sell your refined data to anyone and make big bucks. Like I said earlier in this article, it is one of the most paid jobs in the world today. On average big tech companies pay data scientist between $90,000 - $150,0000. It’s worth it.
P.S: This is more or less like a basic explanation of what data science is all about. Go deeper by searching the internet for more insights. If you’d like to advance more, you could go for a masters in data analytics.
Machine Learning (ML)
Machine learning is a subdomain of artificial intelligence (AI) that provides systems with the ability to automatically learn and improve from experience without being explicitly programmed. ML focuses on the development of computer programs that can access data and use it to learn for themselves.
Our ability to learn and get better at doing things through experience is a part of being human. When you were born, you didn't know anything and you couldn't do much for yourself. But soon enough you were learning and becoming capable day by day. The same process works with computers.
Machine learning brings both structured data and statistics combined with high-end algorithms to enable computers to learn to do a given task without being programmed.
For example, if I want to enable a computer system to differentiate between a dog and a cat, I’ll have to feed it with thousands of images of both dogs and cats stating the big differences between dogs and cats, dogs have bigger heads, cats have furs, etc. As time goes on, the computer system will be able to differentiate both.
But of course, there can be mistakes. The more data the computer receives, the more it finely tunes its algorithm becomes and the more accurate it can be in its predictions.
Machine learning is widely applied in the world today. Take, for example, Sometimes Facebook notifies you that you were in someone’s picture or someone posted your picture without you being tagged in the post.
Facebook was able to recognize you because you’ve been posting your pictures and it already has your pictures in its archive. You fed it with your data (pictures), it automatically learns it’s you without being manually programmed to do so in the background.
Now you can understand better that fine-tuned data structure + ML algorithm = AI (Artificial intelligence). It’s as simple as that.
Another example is this, if you are conversant with using Google, Google tends to use a predictive search feature that uses a predictive search algorithm based on popular searches to predict a user’s search query as it is typed, providing a dropdown list of suggestions that changes as the user adds more characters to the search input.
At times, Google also provides suggestions based on your previous search queries. This is feature is possible because Google leveraged and exploited the data (texts) you and other people have been typing, to improve their service and make more profits. This doesn’t only work on Google, it also works with other series of services and services on the internet.
Other examples of ML at work are:
- Speech recognition
- Self-driving vehicles
- Recommendations based on your previous history
- Targeted emails (e.g spam)
- Ranking posts on social media (e.g Twitter trends)
- Virtual personal assistants (e.g Alexa, Siri)
- Virtual map predictions
- Email and malware filtering
- Online customer support
- Product recommendations
- Online fraud detection
With my basic explanations, you can now understand that data science is like a pipeline that connects down to machine learning.
Top Skills Of being a Machine Learning Engineer
- Computer science fundamentals and programming
- Probability and statistics
- Data modelling and evaluation
- Applying machine learning algorithms and libraries
- Software engineering and system design
All these skills look complex but they aren’t really complex. When you start diving deep into it, they get simpler.
ML engineering relatively combines software engineering with data exploration. Both Data science and machine learning work hand-in-hand, in the sense that machine learning skills can be applied and data science and vice-versa.
In Nigeria today, ML engineers and data scientist earn between N400k - N1.2M. Not bad for the efforts put into it after all.
I guess by now, you're now interested in both fields. Some people tend to work as both data scientist and ML engineers, that's an added advantage.
Working as any of both you can work in any kind of explorable sectors like technology, medicine, transportation, banking & finance, agriculture, industrialization, real estate, research & development, and so many other sectors as long as they have an enormous data you can penetrate and explore.
To learn more about both fields, start by taking online courses on python, database management, data science, basics of machine learning, basics of programming, and most especially mathematics. Get certifications from recognized online learning platforms like Udemy, Coursera, Udacity, etc.
Without maths, a career in data science and machine learning is just guesswork.
Go on and start learning now. The world of data science and machine learning is waiting for you!
Extra Resources (Click the Links)
How to Get Started in Data Science
Essential Math for Data Science
How I Became a Data Scientist Despite Having Been a Math Major
7 Open Source Data Science Projects you Should Add to your Resume
edX: Linear algebra: foundations to frontiers
Coursera: Mathematics for machine learning: linear algebra