Which one for data analytics job: Python, R or Julia?
Araz Nasirian, PhD
AI & Business Analytics Assistant Professor (Lecturer) @ RMIT University | Ph.D. Analytics
I wrote this article based on my personal experience in using Python, R and Julia. Too many people and students do not know which one to choose to get a decent job in data analytics market. Here are some hints:
Python is a general-purpose language that can be used by multiple departments in a company, making it easy to integrate across different teams. I like the NumPy and Pandas libraries. Scikit-Learn is also good. There are many resources available for learning Python, making it an easy language to pick up. The most important point about Python is everyone knows it. Even people who never code in their entire life.
R is my favorite language for data analysis. I love the dplyr library, which I found it much more powerful than Pandas. I love piping in R. The ggplot2 library has no rival for data visualization. The RStudio environment is clearly designed for data analytics, and I really enjoy using it. However, I haven't found the text mining packages in R to be as good as the ones in Python, such as NLTK. The drowback is it is a bit slow. But, not too many people use big datasets. So it is fast enough in a lot of cases.
领英推荐
Julia syntax is very similar to Python and very far from R. In my opinion, it is the best language for optimization and mathematical programming. Julia is also several times faster than Python. I started learning Julia in 2016, when there were very limited resources for troubleshooting and discussing Julia topics. However, there are now many more resources available, including the excellent discourse.julialang.org forum. The main drowback of Julia is it is still unknown in the Australian job market. So when you say Julia, they think you talk about your gf.
Concluding note:
I believe that Julia will become a serious rival to Python for mathematical modeling in the next few years. It is fast and expressive, making it a great choice for this task. It comes from MIT. Everything from MIT is good!
#Python #R #Julia #DataAnalytics #BusinessAnalytics #
?
Analyst at the Australian Government | Data Science | AI
1 年Python for ML and DL. R for data wrangling and visualisation
PhD student at Stevens Institute of Technology
1 年Thank you for sharing this insightful article. I can not be agree more with you that there is no rival for R in data visualization and I love to work in RStudio for all of my data analysis projects. I saw Julia also as a powerful tool while it needs more time to be known by the job market. It means it is the time to carve a niche for ourselves.
Computer Science and Robotics
1 年Surely, Python!
Business Analyst | Data Analyst
1 年I haven't tried Julia yet, but I've heard good things about it, your post made me think about it more seriously ??