TIME SERIES ANALYSIS METHODS IN TRACKING USER ACTIVITY: FROM EVENT COLLECTION TO BEHAVIOR MODELING

Abstract This article explores the methods of time series analysis used to track user activity, focusing on event collection, preprocessing, and the modeling of user behavior. The paper examines traditional statistical methods like AutoRegressive Integrated Moving Average (ARIMA) and exponential smoothing, as well as modern machine learning approaches such as Recurrent Neural Networks (RNN), Long Short-Term Memory (LSTM), and Gated Recurrent Unit (GRU). The article also discusses the challenges of working with user data and demonstrates how it can be utilized to predict and understand audience behavior patterns.