A Behavior Prediction Method Based on Associative Periodic Attention Mechanism

Abstract

The invention discloses a behavior prediction method based on an association cycle attention mechanism. Firstly, the user's learning behavior data is collected according to the log records, the discrete feature is realized through the one-hot vector, and then the learner's behavior feature is constructed through dimensionality reduction; and then Use information entropy to detect learner behavior cycle to serve the later prediction stage; behavior prediction considers the dual influence of sequence behavior and historical behavior to predict behavior, and the detected cycle-associated attention mechanism finds the attention target for it, and then introduces it into the prediction The prediction is done in the base learner LSTM. Compared with the current behavior prediction method, the present invention comprehensively considers the impact of historical behavior and sequence events on current behavior, and innovatively combines the attention mechanism of associated cycles to greatly improve the accuracy of prediction.

Description

technical field [0001] The invention belongs to the field of computer technology, and relates to a behavior prediction method, in particular to a behavior prediction method based on an association cycle attention mechanism; it can be applied to learner behavior prediction oriented to large-scale online education, specifically based on user learning behavior prediction Effective learning behaviors that may occur for the user within a certain period of time. Background technique [0002] Scientists at Northeastern University found that 93 percent of human behavior is predictable. Albert Laszlo Barabas, a well-known professor at Northeastern University, and his colleagues studied the activity patterns of anonymous mobile phone users. They found that although people generally think that the behavior of this embodiment is random and unpredictable, it is Surprisingly, human activities, in fact, follow regular patterns. Their study was published in the journal Science. [0003] ...

AI Technical Summary

Problems solved by technology

Because online education lacks the supervision mechanism of traditional education, the phenomenon of user loss occurs

The reason why users drop out of school may be that the learning resources are not suitable, the learning ability does not match, the learning method is incorrect, or the lack of communication between users leads to insufficient learning motivation and driving force, etc.

Images