CareerCorpus: A comprehensive dataset of annotated resumes

The CareerCorpus dataset contains 302 annotated resumes collected from Kaggle (LiveCareer.com) and LinkedIn, covering six occupational categories: Teacher, Finance, Apparel, Accountant, Banking, and Research Assistant. The dataset supports multi-class classification for resume categorization. Each resume received dual annotations from domain experts with verified professional or academic backgrounds in their respective fields. Financial categories (Finance, Accountant, Banking) were annotated by professionals with 5+ years of accounting experience and ICMAB certifications, while specialized categories were annotated by industry practitioners and university lecturers. Data preprocessing involved HTML-to-text conversion using GPT-5, standardized formatting, removal of personally identifiable information (PII), duplicate elimination, and text normalization. Both annotations are preserved in the dataset to enable flexible consensus methods and annotation uncertainty analysis. Inter-annotator agreement varies by category: Apparel (r = 0.89), Finance (r = 0.68), Research Assistant (r = 0.67), Teacher (r = 0.56), Banking (r = 0.38), and Accountant (r = 0.35). Overall mean squared error is 0.023 and mean absolute error is 0.106 across categories. The dataset is released in Excel format (.xlsx) with separate files for each category, available through Mendeley Data under a CC-BY-4.0 license. The dataset can be used for resume classification, skill extraction, recruitment analytics, and related natural language processing research.