Abstract: Traditional Apriori algorithm can only deal with Boolean data and can't process the medical records of breast cancer patients with continuous data. In this regard， a prediction method of breast cancer diffusion based on the improved Apriori algorithm is proposed，which based on a new support calculation method including the fuzzy set theory. Experimental results show that the improved algorithm can process records of breast cancer patients with continuous data. Compared with traditional algorithms，it can mine more and higher quality rules，and obtain the hidden rules between the pathogenic factors and diffusion of breast cancer patients. It verifies that the improved Apriori algorithm is of guiding significance for the treatment of breast cancer patients.

Key words: data mining, Apriori algorithm, breast cancer diffusion prediction, fuzzy sets