The invention discloses a
mobile phone APP advertisement interception method based on a neural network and a Mitmproxy agent, and belongs to the technical field of
secure communication. The problem that in the prior art, a traditional advertisement intercepting method is low in intercepting precision and is difficult to adapt to a mobile terminal is solved. According to the method, the traffic is captured through a Mitmproxy agent,
a domain name filtering module based on
domain name rule classification is constructed to filter domain names requested in the traffic, and pop-up advertisements are intercepted; analyzing the traffic into a
JSON format, constructing an advertisement recognition module, obtaining a preprocessed
data set, inputting the preprocessed
data set into the constructed deep neural
network model for training, and obtaining a trained neural
network model and an analysis result; and
processing the embedded advertisement identified by the trained neural
network model through a set advertisement interception module to complete
mobile phone APP advertisement interception. The method improves the accuracy and efficiency of advertisement recognition, and can be applied to mobile terminal advertisement interception.