We propose a recognition method based on statistics through analysis the grammatical and semantic characteristics of the Chinese organization name. This recognition method includes three elements: frequency, part of speech, word length. We use the data in mature collection as training data; separately calculate a candidate organization name's word frequency, part of speech and word length of the contribution. Finally get this candidate organization name's contribution, compare with a given threshold and achieve the recognition of a Chinese organization name.