In order to improve personal data security in the elec- tronic commerce, and to avoid information exposing of per- sonal privacy, personal data was used as the encrypted principle, applying homomorphism and random perturba- tion to strengthen personal data security. This paper an- alyze the usage of personal data in electronic commerce firstly, Then discuss the P3P to realize security of personal data. This mechanism cannot guarantee that Web sites do act according to their policies once they have obtained user?s personal data. In light of this, a new algorithm was used to acquire improved security. The proposed algorithm uses homomorphism as principle to preserve privacy. This method can make the important data to be appeared with secret content in the web service, and prevent personal data from being misused. The simulation results proved that the modified method can protect personal privacy effectively, and can carry on data mining to provide the characteristic service for the customer, and modified method has shorter response time.