The invention discloses a user behavior analysis method and system based on big data. According to the method, user behavior data are collected through a client side in real time, user behaviors and contextual information of a page URL are combined, the real scene that a user browses Web pages is reproduced to the greatest extent, comprehensive user behavior tracks are extracted, and effective data assurance is provided for analyzing the user behaviors; security assurance is provided for the user behavior data through a safety analysis module, and a user behavior data body model is used for modeling the user behaviors, so that behavior information semantic levels are shared and reused, and the interoperability and the reliability of the model are improved; the user behavior and context data are collected in real time for analysis, so that a result is more reliable; body and behavior information is stored through a column storage database, and therefore the foundation of massive data management is laid; the powerful processing capacity and the large-scale data storage capacity of the cloud computing technology, the body and a reasoning and knowledge discovery method of the body are combined, the massive user behavior data are analyzed in real time to obtain user interest in time, and then effective and accurate user push is achieved.