User Behavior Research Of English Mobile Learning App

With the acceleration of global integration and the deepening of lifelong education concept,people's demand for English mobile learning is increasing.Because of the inconvenience of space-time and other factors,traditional learning methods limit people's learning.Therefore,breaking through the limitation of time and space,using mobile devices to learn English anytime and anywhere is becoming more and more popular,and English mobile learning APP has emerged as the times require.At the same time,with the increasing number and time spent on mobile phones,hundreds of millions of behavioral data have been generated,and these massive user data have great commercial value.Based on the user behavior data of English mobile learning APP,this paper mainly studies three aspects of user attributes,behavior and loss,describes the behavior characteristics of English mobile learning APP user groups,and provides some reference value for precision marketing.Firstly,based on user portraits,this paper describes the user attributes and user characteristics of APP target population in English mobile learning,mainly describing the user population characteristics and distribution.Based on the clustering of more than a dozen user behavior data,the results show that APP users can be divided into four categories.The first category: older people,lower educational background,negative learning tendency,and negative game tendency.This kind of population has a lower frequency of using various applications in mobile phones.The second category: male majority,low age,low educational background,high learning inclination,high game inclination,high frequency of use of various applications.The third category: higher education,fewer mobile phone applications and higher frequency of use.The fourth category: women are the majority,older,higher education,the highest frequency of use of various types of mobile phone applications.Secondly,based on Apriori association rule algorithm,this paper gives the basic attributes and behavior characteristics of English mobile learning APP to stabilize customers.This kind of stable customers tend to be female,aged between 18 and 22,with undergraduate education and second-tier cities.Based on these conclusions,enterprises can directly target users with demand and achieve better traffic conversion rate.Finally,according to the actual situation,a new definition of APP user churn in English mobile learning is defined,and a Logistic model is established to analyze the influencing factors of user churn.This paper establishes a user churn prediction model using SVM,and the prediction accuracy is above 85%.The model can help enterprises to find users who may lose,and enable enterprises to implement effective marketing strategies to retain users in time to reduce the user loss rate.