During the process of component reusing

users often have no all-around understanding of the component description mechanism.So its difficult for users to propose exact retrieval requirements

which leads to lower precision.We refer to association mining theory in data mining field

and present a self-adaptive component retrieval model with user-feedback and the association mining-based self-adaptive learning algorithm.It can obtain the internal relationships between users explicit retrieval conditions and their implicit requirements through mining in the retrieval history

thus making the retrieval conditions more complete and precise

which increase the precision of component retrieval.The experimental results show that this solution is feasible and effective.