Abstract: Mobile phones and other mobile devices, make micro-blog and other social networks become an important channel for information dissemination and sharing. But at the same time, a large number of reactionary, false and pornographic information filled the entire network, and the rumors of the increasingly prominent adverse effects. Furthermore, the emergence of some rumors has seriously affected people’s access to network information and normal use. How to detect all kinds of rumors in the network, and to find out the source of the rumor and the way of communication has become an urgent problem for public security network department to be solved. Micro-blog rumors, for example, according to the existing longest common subsequence(LCS) algorithm, the corresponding improvement is made in the construction of sequence table. The preliminary experiments show that the improved algorithm can be more efficient to compare and originate micro-blog rumors, so as to help public security network department to find the source of micro-blog rumors.

Key words: micro-blog rumors, origin, LCS algorithm, dynamic programming