Two new indicators are introduced in this paper - "citation rate during sleep period" and "citation rate in recent five years" - on the basis of B-value method to restrain the number of citations of sleeping beauty during sleeping period and wakening period. This method is used to identify "sleeping beauty" literature in the two different fields to verify its advantages and analyze the sleeping and awakening mechanism of sleeping beauty. By comparing and analyzing the sleeping and awakening mechanism of sleeping beauty in these two disciplines, this study provides guidance for different disciplines on how to find the sleeping beauty publications as early as possible. Improved B-value Algorithm was used to identify sleeping beauties published between 1986 and 2007 in the Web of Science (WoS) database in the two fields of WR and ISLS. The results show that the improved B-value recognition method can identify 10 sleeping beauties in ISLS and WR fields, which cannot be recognized only by B-value method, and the recognition accuracy is improved by 43.4%.In terms of the citation curve, the cited frequency of Sleeping Beauty Literature identified by the B-value method only fluctuates greatly during the sleeping period, and the average annual cited frequency of sleeping Beauty Literature is more than 2. Moreover, the citation frequency of three articles decreased significantly, which did not accord with the characteristics of Sleeping Beauty Literature. From the perspective of influencing factors of papers,the weak academic influence of journals and authors are the important reasons for the paper to become a "sleeping beauty". Moreover, research suggest that compared with the ISLS field, the "sleeping beauty" in the WR field is more influenced by the authority of the journals when awakened.