A Tibetan and Uygur Sensitive Word Automatically Add System Based on Co-occurrence

In this paper, we first built a sensitive word vocabulary and classified the sensitive words. Then in order to add Tibetan and Uygur sensitive word automatically, we adopted a method based on co-occurrence. In our system, we used a simple algorithm to calculate the relevance of sensitive words. According to the relevance of words, the new sensitive words are added to the vocabulary automatically.