To quickly and efficiently retrieve the target component from a large software component library, a component retrieval method based on identification of faceted classification and cluster tree was proposed. The component with facet classification identification was described by using the set of component identification, which overcomes the impact of subjective factors when only using facets classification to describe and retrieve components. By introducing cluster tree, the component cluster tree was established by analysis clustering of components based on semantic similarity, thus narrowing the retrieval area, reducing the number of comparisons with component libary, and improving the search efficiency. Finally, the proposed method was experimented and compared with other common retrieval methods. The results show that the precision of the proposed method is 88.3% and the recall ratio is 93.1%; moreover, the proposed method also has a good retrieval effect when searching in a large-scale component library.