Abstract: In order to ensure fair and fair ticketing and protect the interests of people ticket buying, this paper designed an abnormal user intelligence recognition algorithm of railway Internet ticking based on index weight by using the big data technology and combining with existing user ticket buying behavior data. Using the user behavior data in 2017,the accuracy of abnormal user prediction reached 80%. The test results verify the feasibility of the algorithm, which can effectively improve the accuracy of abnormal user identification, and provide technical support for ensuring the safe and stable operation of the 12306 railway Internet Ticketing and Reservation System and maintaining fair and fair ticketing environment.