The invention discloses a man-
machine verification detection method based on JS (
JavaScript) interaction behaviors, which is characterized by comprising the following steps that step 1, a
server sends a calling request to a
scoring system, and the
server stores a real Token based on a
session ID (Identity) of a service; 2, in the
client side, the JS-SDK carries out decryption
verification on the mixToken to obtain a realToken, and the realToken is subjected to
confusion arrangement to generate a JsmixToken; 3, a monitoring collection module is set to obtain parameter information of the interaction event, and the parameter information is reported to the
server; and 4, a scoring component is set to perform deserialization analysis on the parameter information, and the parameter information is input into a behavior analysis module to obtain an interaction
event analysis result. According to the method, automatic measurement and scoring can be achieved based on interaction behavior
data analysis, parameter information of interaction events generated by users is monitored and collected through a monitoring and collecting module, a response request of login
simulation and a counterfeit request of
mimicry interaction are analyzed, an evaluation standard is provided for a
service system, and an external environment technology is combined to continuously optimize an analysis model, and the accuracy of man-
machine detection is continuously improved.