Method and system for detecting abnormal data in mobile advertisement click traffic

The invention relates to a method and system for detecting abnormal data in mobile advertisement click traffic, and the method comprises the steps: obtaining a mobile advertisement click traffic table data set, carrying out the preprocessing of numerical features and classification features in the data set, and randomly dividing the data set into a training set and a test set; inputting the training set into a pre-constructed abnormal mobile traffic table data model, representing a learning borrowing task by using traffic table data, and training the abnormal mobile traffic table data model by providing context knowledge priori through self-supervised pre-training to obtain a trained abnormal mobile traffic table data model; and inputting the test set into the trained abnormal mobile traffic table data model to obtain false click traffic data in the mobile advertisement click traffic. According to the method, false invalid click traffic data can be effectively detected, loss caused by invalid traffic is avoid