The Resume Corpus: A Large Dataset for Research in Information Extraction Systems

We publish a Chinese Resume Corpus for researches of information extraction. The corpus contains 178 thousand resume documents and over 33 million words. The resume documents with unstructured form contain many types of common information like name, gender, birthday, education, work experience, and so forth. We evaluate this corpus by using four types of mainstream neural network models for demonstrating the potential of our corpus to promote the development of information extraction research.