In practice, it is difficult to accurately identify the wear state of hob during the complex hobbing process. A hob wear state identification method, which based on singular spectrum analysis (SSA) and weighted Naive Bayes, is proposed to cope with this problem. Firstly, the Z-direction vibration signal of the main shaft in the whole life working cycle of the hob is acquired, and the vibration signal is decomposed into a noise dominant component and a signal dominant component by SSA. Then, kurtosis and pulse factors in the noise dominant component are selected as feature vector of the hob wear state. Owing to the poor Naive Bayes classification effect caused by inconsistent importance of different feature classification, Naive Bayes which weighted with information gain ratio is employed to optimize. Finally, the feature vector is input into the weighted Naive Bayes for training to achieve the automatic identification of the hob wear state. The experimental results show that the method can quickly and accurately identify the wear state of hob with the identification rate up to 98.3%.