NetEase Cloud Dataset: Active User Identification and Deep Neural Network Based CTR
Yue Quan, Kaiwei Wang, Sichen Guan · 2021
The online short-video platform has become the new emerging market of the online social media era. NetEase Cloud Music is one of the popular music apps that is transformed into the latest form of music app that contains short-videos. This study has self-defined the user active level and using linear regression to analyze the mathematical relationship of active level with user features. Neural Network is a great model when dealing with a significant amount of data. The online-short video click-through rate (CTR) is calculated using the approach of the neural network. Mean-Squared error is also using by changing the number of variables to find the minimum error. The goal of this paper is to stress the relationship between active user level defined by ourselves and the FNN prediction. It is also essential to calculate the CTR of users with further research on the Reco system.