How can I start working towards an AI/ML career?
As a absolute beginner to Data Science, I have recently completed the Python for Data Science at Cognitive Class. When I progressed to Data Analysis, I found myself struggling to understand the statistical concepts and syntax as it began to skip explanations.
As my lecturer did not teach us about the concepts behind statistical methods, I have little knowledge on the concepts and only know how to follow the formulas. I wish to pursue a Data Sciene internship but I doubt I can complete so many courses in a month or two.
My questions are:
- How do I work towards Data Science?
- Assuming that I'll be using ML frameworks such as scikit-learn, what are the mathematics required to achieve this? Should I start from pre-calculus?
- If I'm not using frameworks, what kind of mathematics would I need?
- What mathematics foundation would I need to start learning Data Science?
Learn Something New Everyday,
Connect With The Best Developers!
Note that artificial intelligence, machine learning, data science and data analytics have some overlap, but they're not the same. A good first step might be to be more clear about your goal.
How do I work towards Data Science?
There's courses and books, but don't forget to practice, e.g. on Kaggle.com.
What mathematics foundation would I need to start learning Data Science?
Statistics is very important and linear algebra is also useful.
I doubt I can complete so many courses in a month or two.
No, you cannot.
Simply start to work in Data Science field and practice,practice,practice and again - practice on different DS problems. As Mark said - Kaggle.com is a very good site for DS practical training.
Data Science is multidisciplinary field and involves mathematics, statistics and computer science. I would say that there is no a "key" field in math, which would make you a super-strong in data science. Everything equally counts as it broadens your knowledge and data analysis "intuition". The more you know - the more tools you can choose from and more insights you can gain from the data. So if you wish a recommendation - you can start from calculus, differential equations, series, limits, group theory, sets and many more. Of course you MUST know linear algebra, curve fitting, least-squares and other error estimation methods, ROC curve - ML classification quality analysis, etc, etc, etc. In the end - just START :-)