Hello, this is my first post, so I hope I’m doing it right, and thank you for reading.
I am currently enrolled in the Hugging Face NLP course but find myself constantly relating it to my project. At the moment, I feel somewhat overwhelmed with the options available, so I’d like to seek advice. My current project involves a dataset of job posting descriptions. My goal is to analyze these descriptions to extract required skills (like Python, SQL, etc.) and the years of experience needed for each skill. Initially, I considered using a text classification model, training it to identify skills and experience years, and then extracting this information from the text, potentially matching skills and experience based on their proximity in the text. However, I’ve also contemplated a question-answer model. Given that the texts can be lengthy and contain multiple skills or varying years of experience, this approach might be overly complex.
I would greatly appreciate your opinion on the best method to tackle this project, especially considering that models for analyzing CVs and similar tasks already exist. Maybe there’s a better approach I should consider?