The current researches of knowledge discovery are introduced under the environment of incomplete data, and then an approach is presented for knowledge discovery in incomplete databases through the two parts. Firstly, the method of how to guess the missing data is in detail discussed and the definition as well as the mining method of distance based association rule is given. Then based on this, an algorithm is described in detail for discovering knowledge in incomplete databases, the complexity of the algorithm is analyzed and the experimental results are also given. The paper finally concludes with a comparison of the approach proposed in the paper with other related ones.