Research of web recruitment information extraction based on NLU

摘要

With the rapid development of Internet technology, network has become main source of recruitment information. How to extract the useful recruitment information automatically from these recruitment websites is an important subject nowadays. This thesis studied processing the half-structure text by Natural Language Understanding and improved the existing language model to realize Chinese named entity recognition. Based on this, the author designed a web recruitment information extraction system. The system has a good extensibility and adaptability and precision and recall can reach above 70%.