Classyfing job titles

[RocketFan]

• https://www.kaggle.com/code/samuelguebo/classifying-job-offers-through-nlp#Test-the-webservice - example
• https://medium.com/ziprecruiter-tech/classifying-job-titles-with-noisy-labels-using-reinforce-ce1a4bde05e2 - more advance approach, but without actual code
• https://towardsai.net/p/nlp/natural-language-processing-nlp-with-python-tutorial-for-beginners-1f54e610a1a0 - nice tutorial for nlp and libraries for it in python

[RocketFan]

Jobs were grouped into unlabeled categories with unsupervised learning approach. After short investigation some categories as technology and management workers can be easily found from top 10 words, but with some of them are still grouped with common labels such as people, benefits, gender etc. The ones which don't have any meaning in context of the job type should be removed from nlp analysis.