Clustering-Based Hotspot Analysis And Alarm Compression For Mobile Internet

In recent years,with the popularity of mobile network infrastructure and smart phones,the mobile Internet has ushered in rapid development.The mobile Internet has penetrated into every aspect of people's lives.From food and clothing to entertainment,learning and work,these are all related to mobile applications.At the same time,netizens are increasingly demanding mobile application service experience.In addition to providing basic functional services for netizens,mobile applications also need to meet the requirements of convenience,efficiency,and aesthetics when using Internet services.The improvement of the use experience of netizens requires both upper-layer mobile application service providers to continuously optimize and upgrade their products according to the needs of netizens,and mobile network infrastructure to provide a guarantee for the stable transmission of data by upper-layer applications.Mobile application user data analysis and mobile communication network alarm management are two important contents that help to enhance the experience of netizens.Mobile application user data analysis analyzes user data to understand the actual needs,preferences,and habits of netizens when using mobile applications to guide mobile application optimization.The alarm management of the mobile communication network provides early warning of possible failures in the mobile communication network,thereby helping to eliminate potential faults in the mobile communication network in time and ensuring stable operation of the mobile communication network.This paper analyzes these two contents,and conducts in-depth research on two specific problems,that is the user data hotspot analysis and alarm aggregation.The main work and contributions of this paper are as follows:(1)A clustering-based data hotspot analysis method is proposed.Hotspot detection in data aims at finding out those areas with high density of data,and presenting these areas in a interpretable way.In this work,hotspot detecting algorithm is designed to deal with multi-dimensional data containing numerical features as well as categorical features.The core of the algorithm is the clustering algorithm CLTree+,a significant improvement over the baseline CLTree.CLTree+is able to deal with numerical features and categorical features,and the clustering result of numerical features with periodical characteristics is also improved.Besides,the computational efficiency of CLTree+is also improved.CLTree+is applied to transaction data of large Internet businesses and find out a few areas with high density of data,and these areas are presented as the easy to interpret combinations of attributes and its values(2)An alarm aggregation algorithm is designed for the alarm data of mobile communication networks.In an actual production environment,too many alarms are an important cause of poor practicality of the alarm management system.By analyzing the real alarm data of the mobile communication network,this paper finds out the reasons for the excessive number of alarms,and has formulated a series of alarm information consolidation strategies.These strategies are used to aggregate real alarm data generated by mobile communication networks with a aggregation rate of 3.6%.The results of the aggregation have been evaluated by experts in the relevant fields,who confirmed the correctness of the algorithm.