Can You Study Data Science Without Maths? - Check Here

Can You Study Data Science Without Maths?

To answer this question directly, you can not study data science without maths. Data science is a subject built on the pillars of mathematics. However, some streams of data science need maths more than others. For example, you need to understand basic statistics to understand the concepts of average, mean, median, percentages and more. You need to understand probability and basic algebra for the basics. When it comes to specialisations and choosing a career afterwards, you can go with roles as per your proficiency in maths for example, for roles such as data analyst, business intelligence developer, data engineer, just a good grasp over basic concepts of maths are sufficient.

Also read,

• 7 Career Paths for Psychology Graduate Beyond Counselling

Data Science for Non-Math Backgrounds

Technically you can take admission in data science as an arts or commerce student, but you still need to have maths skills. But there is a reason why students who have studied maths are preferred for data science. You will have to work with averages, medians, percentages, probability, variables - for which you need mathematics.

However, there are some roles you can fit into without maths, involving but not limited to data cleaning, data visualisation, using libraries, business analysis and more.

What Math Chapters Actually Matter in Data Science?

Here are some of the important chapters in maths which are important for data science. Take a look below to

• Descriptive & Inferential Statistics (Essential): This chapter consists of mean, median, standard deviation, hypothesis testing (p-values), confidence intervals, A/B testing and more. This helps drawing false conclusions from dirty data.

• Linear Algebra (Fundamental for ML): This chapter consists of Vectors, matrices, dot products, dimensionality reduction (PCA) and more. ML engineers need it as data is represented as matrices; computers process tabular and image data through matrix operations.

• Calculus (Targeted, Not Comprehensive): This chapter consists of Derivatives, partial derivatives, gradient descent and more. This helps understand how machine learning models minimise errors, though it is mostly calculated using a computer automatically.

Where does maths actually matter in data science?

You definitely need to have working knowledge of maths to study and understand the fundamentals of data science. But after you graduate, you can choose which path you walk, as in some roles you do not need that deep knowledge of maths. For example,

• Data Analyst: You need a low to moderate level of maths knowledge in this role. Here, your core responsibilities are SQl, Excel, managing business dashboard and more. You can get away with low proficiency in maths.

• Data Scientist: You need a moderate level of maths knowledge to be a data scientist. Here, your job responsibilities are training established machine learning models, evaluating metrics, among other things. You typically are not the ones doing maths here, the machine does it for you, you just need to validate the results.

• Machine Learning Engineer: Here things get serious. You definitely need high knowledge in maths to be in this role. In this role, you will be responsible for optimising models, check systems and more. A strong foundation in linear algebra and calculus is fundamental to this role.

• AI/ ML Researcher: This is where your maths knowledge needs to be very high. AI-ML researchers develop new algorithms/ architectures, write papers, and provide theoretical proof. Having advanced graduate level knowledge in maths is essential to be in this role.

Possible Learning Path for Data Science

Many students often go through maths text books before writing a single line of code. But data science works a bit differently. Here is a possible approach you can take:

• Phase 1: learn programming languages which are essential to data science, like Python, Pandas and visualisation tools. Practice by building simple projects using pre-built libraries.

• Phase 2: This phase will help you understand your projects in depth. How to spot problems if something goes wrong, where to look when debugging and more.

• Phase 3: In this phase, you will learn to improve your targeted maths proficiency. Study only the specific mathematical concept needed to solve that exact limitation.