54K Resume Corpus: Structured Professional Profiles

54,000 professional resumes aggregated from public sources such as Kaggle and GitHub. The resumes were originally in PDF format, converted to text via OCR, and structured using LLMs. The dataset was uploaded by Suriyaganesh on November 14, 2024.

Use Cases

• Training resume parsing models based on structured text extracted from PDFs.
• Developing job recommendation algorithms based on professional profile data.
• Analyzing skill trends across industries based on aggregated resume content.
• Benchmarking OCR and LLM-based text structuring techniques for document processing.

Strengths

• Contains 54,000 resume entries, providing a substantial corpus.
• Aggregated from multiple public sources, likely increasing diversity.
• Resumes were processed with OCR and LLMs to create a structured format.

Limitations

• Column-level documentation is absent; field semantics must be inferred after download.
• Row count is known, but specific column names and data types are unknown.
• Data may reflect source bias inherent to the original Kaggle and GitHub repositories.

Provenance

Source

Aggregated from Kaggle and GitHub repositories.

Collection Method

PDFs converted to text via OCR, then structured using LLMs.

Freshness

Last updated 2024-11-14 13:10:13