Natural language processing (NLP) has been designed to convert narrative text into structured data. Although some general NLP architectures have been developed, a task-specific NLP framework to facilitate the effective use of data is still a challenge in lexical resource limited regions, such as China. The purpose of this study is to design and develop a task-specific NLP framework to extract targeted information from particular documents by adopting dedicated algorithms on current limited lexical resources. In this framework, a shared and evolving ontology mechanism was designed. The result has shown that such a free text driven platform will accelerate the NLP technology acceptance in China.