Recommendation system click rate prediction method based on deep neural network

Abstract

The invention discloses a recommendation system click rate prediction method based on a deep neural network, and the method comprises the steps: collecting a user click behavior as a sample, extracting numerical features of the sample with a numerical value relation, and inputting the numerical features into a GBDT tree model for training, and obtaining a GBDT leaf node matrix E1; inputting a behavior sequence formed by clicking the articles by all the users in the sample into an Attention network to obtain an interest intensity matrix E2 of all the users in the sample for the articles; summing and averaging the article feature vectors of the click interaction of the user to obtain a click interaction matrix E3 corresponding to the user, splicing E1, E2 and E3, and inputting the E1, E2 andE3 into a deep neural network model with three hidden layers and one output layer to output a prediction result. According to the method, user clicking behaviors are decomposed into attribute characteristics, a GBDT tree model, an Attention network and a deep neural network model are subjected to nonlinear fitting, a recommendation system clicking rate prediction model is constructed, a prediction result is obtained through model training, and the method has the advantages of deep mining of recent interests of users, high generalization degree and high expansibility.

Description

Technical field [0001] The present invention relates to the field of information recommendation systems, in particular to a method for predicting click-through rate of recommendation systems based on deep neural networks. Background technique [0002] With the development of information technology and the Internet industry, information overload has become a challenge for people to process information. For users, how to quickly and accurately locate the content they need in exponentially growing resources is a very important and challenging thing. For merchants, how to present appropriate items to users in a timely manner so as to promote transaction volume and economic growth is also a difficult task. The birth of the recommendation system greatly eases this difficulty. [0003] The recommendation system is an information filtering system that mainly uses user portraits, item information, and user behavior data such as searches, clicks, and favorites to recommend items that may b...

AI Technical Summary

Problems solved by technology

[0006] (2) It is difficult to obtain high-order cross features

[0007] (3) For high-dimensional sparse id data, the generalization ability is not enough,

Although the DeepFM model has become the mainstream model for CTR estimation, it is found in practical applications that the scalability of DeepFM is not enough

There are still some important continuous features in the recommendation task, and sometimes a small amount of feature engineering is still required, such as counting the number of clicks of users in recent days, the difference of user click time, etc. If the continuous features are directly input into the neural network model In , not only complex normalization is required, but also it cannot have a good effect on prediction

Images