Study on context-based domain ontology concept extraction and relation extraction
基于上下文的领域本体概念和关系的提取

Recently, ontology learning focuses on concept extraction and conceptual relation extraction. For concept extraction, domain relevance combined with domain consistent had yielded better results, and the algorithm of association rules was mainly adopted for relation extraction. Since the traditional methods only considered the word frequency, there were many substantial inaccuracies in learning results. To overcome these shortcomings, this paper proposed a new learning method based on context. In this way, represented semantic similarity between words and could overcome these shortcomings. The experimental results show that this method can effectively improve the performance of ontology learning system.