Temporal information processing plays an important role in many application areas such as information retrieval, question answering, machine translation, and text summarization. This paper proposes a transformation-based error-driven learning approach to extracting temporal expressions from Chinese unstructured texts. The temporal expression annotator used in the approach is developed based on a Chinese time ontology, which includes concepts of temporal expressions and their taxonomical relations. Experiments in three domains show that our algorithm obtained promising results.