Noise is Signal: Density-Based Outliers as Leading Indicators of Occupational Emergence in Labor Market Text

The paper presents the Emergence-Density Inversion (EDI) hypothesis, arguing that noise in job postings can signal emerging occupations rather than incoherence. Analyzing 84,988 job postings, the study shows that high-EOS outlier groups transition to stable clusters significantly faster than low-EOS groups, with an improved clustering prediction F1 score of 0.74 using the Extended Emerging Occupation Score (EOS). This approach identifies new roles like Prompt Engineer and AI Safety Researcher up to three quarters before they stabilize, providing valuable insights for practitioners in labor market analytics and workforce planning.