I have an interest problem to solve (in Python, in case anybody is wondering to give a specific answer).

I want to predict the salary range, given the job title. The inference should be able to handle any job title input. With that, I mean that the input could vary widely like:

“IT Engineer $$$$ with Great Benefits,”

“Sous Chef at a Restaurant,” “Sr. Engineer in IT,”

“None,”

“Product person,”

“Marketing Specialist with Remote Work,”

“Data Scientist in a Tech Startup,”

“Junior Software Developer,”

“Senior Sales Representative Company A Houston Texas,”

“Chef de Partida,”

Assume that you only have 3 columns available to solve this problem. The job title, salary from and salary to.

The challenge includes normalizing these job titles, which might involve steps like cleaning, preprocessing, applying LDA, and other necessary techniques to make accurate predictions.

The whole concept is to provide something straightforward which can then be scaled. It’s not about creating something advanced for no reason.

I’ve opened up this thread to hear your take on it, different aspects and approaches to this problem. Any answer is welcome but I would more focus on the conceptual side of things!

Looking forward reading the comments section!

I’ve tried quite some stuff but I wouldn’t like to bias the audience just yet. I’m more than happy to share though!