The invention relates to a malicious user identification method. The malicious user identification method includes the following steps that: user identifiers of global users are obtained, and credit corresponding to the user identifiers is calculated, and global user credit distribution is generated according to the credit corresponding to the user identifiers of the global users; user identifiers of conditional users are screened from the user identifiers of the global users according to query conditions; credit corresponding to the user identifiers of the conditional users is obtained, and conditional user credit distribution is generated according to the credit corresponding to the user identifiers of the conditional users; the number of global high-quality users and the number of conditional high-quality users of which the credit is larger than a credit threshold value are respectively obtained from the global user credit distribution and the conditional user credit distribution; and a preset threshold value of the number of high-quality users is obtained, and the difference value of the number of the global high-quality users and the number of the conditional high-quality users is calculated, and whether the query conditions are corresponding to malicious users can be judged according to the difference value and the threshold value of the number of the high-quality users. The above malicious user identification method can improve the accuracy of identification.