A groundbreaking development in the field of Natural Language Processing (NLP) has emerged from the Stanford Institute for Human Centered Artificial Intelligence (Stanford HAI) at Stanford University. Researchers, led by Sarah Bana, have developed a machine learning algorithm that can predict a job’s salary based on the language used in the job posting. The algorithm, called BERT (Bidirectional Encoder Representations from Transformers), was trained on 800,000 job postings and their corresponding salaries from the job platform Greenwich.HR. The researchers then tested BERT on an additional 200,000 job postings and found that it accurately predicted the salary for 87% of the postings.
The problem of job postings not including salary information has long been a frustration for job seekers. Without this information, candidates are forced to apply “blindly” and may waste time applying for jobs that do not meet their salary requirements. Bana and her team hope that their algorithm will make the job application process more transparent and improve workforce training and development.
BERT is a powerful tool for NLP, developed by Google, that can analyze natural language and determine the meaning behind certain words. By training BERT on job postings and salaries, the researchers were able to identify patterns and correlations between the language used in job postings and the corresponding salaries. This breakthrough could have significant implications for the job market and could help to level the playing field for job seekers.
Overall, the development of BERT and its ability to predict job salaries based on language used in job postings is a significant advancement in the field of NLP. With the potential to make the job application process more transparent and improve workforce training and development, BERT could have a significant impact on the job market in the years to come.
