针对现有跨语言检索模型普遍存在的翻译准确性差、效率低、成本高等不足,在深入分析主题图技术在揭示语词概念之间的语义关系和多语言支持等方面的优越性能的基础上,提出一个基于主题图的英汉跨语言检索模型,该模型采用索引翻译的策略来实现跨语言检索,其突出特点是能够在提高翻译准确性的同时,有效降低翻译成本,并且实现起来也比较简单。 Aiming at the poor translation accuracy, low efficiency and high cost which are ubiquitous in the existing cross-language retrieval models, this paper analyzes the superiority of the topic graph technology in revealing the semantic relations and multi-lingual support among the word concepts, An English-Chinese cross-language retrieval model based on thematic maps is proposed. This model adopts the strategy of index translation to achieve cross-language retrieval. The outstanding feature of this model is that it can effectively reduce the translation cost while improving the translation accuracy and is simple to implement.