AI-POWERED JOB MARKET INSIGHTS TO PREDICT FUTURE DEMAND FOR SKILLS ACROSS VARIOUS JOB MARKETS

Abstract

The global job market is experiencing unprecedented transformation, driven by rapid technological advancement, artificial intelligence, and automation. Reports indicate that over 85% of jobs expected by 2030 have yet to be created, while nearly 40% of the current workforce will require reskilling within the next five years. Although automation may displace approximately 75 million jobs, it is also projected to create 133 million new roles, underscoring a major shift in skill demand. Traditional job market analysis methods are largely manual, reactive, and dependent on static labor reports, limiting their ability to process real-time, large-scale employment data. Moreover, conventional machine learning approaches such as the Gradient Boosting Classifier (GBC) often exhibit lower predictive accuracy and efficiency when forecasting evolving skill requirements. To address these challenges, this study proposes an AI-powered hybrid deep learning framework to predict job market trends by classifying roles across companies and sectors as growing, declining, or stable. The system processes large-scale job postings and labor trend data through comprehensive preprocessing, including keyword extraction, text normalization, and feature encoding. A Convolutional Neural Network (CNN) is employed to capture deep semantic patterns in job descriptions, while a Random Forest Classifier (RFC) provides robust and interpretable classification. Experimental results demonstrate that the proposed hybrid model significantly outperforms GBC in terms of accuracy and efficiency. The framework offers actionable insights into emerging skill demands, supporting informed workforce planning, reskilling strategies, and policy development.