Analysis of Lost Users of Operators Based on Big Data
1. Project Background
With the rapid development of business, the competition in the mobile business market has intensified. How to retain online users and absorb new customers to the greatest extent is one of the most concerned issues for telecom companies. Competitors' promotions, the introduction of soft-landing measures for companies' tariffs, and the constant changes in policies and regulations have affected customers' consumption psychology and behavior, resulting in constantly changing customer churn characteristics. For telecom operators, loss will bring a series of problems such as declining market share, increasing marketing costs, and declining profits for telecom companies. While developing the monthly increase of users, how to retain and win more users is a very important task.
With the continuous development and application of big data mining technology, mobile operators hope to use data mining technology to identify which users may churn and when will churn occur. By establishing a churn prediction model, analyzing users' historical data and current data, extracting key data to assist decision-making, discovering hidden relationships and patterns, and predicting possible future behaviors, mobile operators can help mobile operators achieve these requirements.
2. Original data
(1) Main data: basic information and usage behavior information of operator users.
(2) Data set description: modeling user data set user_info_m
(3) Data format: The original file is a compressed package, which is in CSV format after decompression.
(4) Field description: as shown in Table 1 below:
Table 1 Data Description
Name Field Description
MONTH_ID Month
USER_ID User ID
INNET_MONTH Internet Time
IS_AGREE Whether the contract is valid or not User
AGREE_EXP_DATE Contract plan expiration time
CREDIT_LEVEL Credit level
VIP_LVL VIP level
ACCT_FEE This month's fee ( Yuan)
CALL_DURA Call duration (seconds)
NO_ROAM_LOCAL_CALL_DURA Local call duration (seconds)
NO_ROAM_GN_LONG_CALL_DURA Domestic long-distance call duration (seconds)
GN_ROAM_CALL_DURA Domestic roaming call duration (seconds)
CDR_NUM Number of calls (times)
NO_ROAM_CDR_NUM Number of non-roaming calls (times)
NO_ROAM_LOCAL_CDR_NUM Number of local calls (times)
NO_ROAM_GN_LONG_CDR_NUM Number of domestic long-distance calls (times)
GN_ROAM_CDR_NUM Number of domestic roaming calls (times)
P2P_SMS_CNT_UP Number of text messages sent (pieces)
TOTAL_FLUX Internet traffic (MB)
LOCAL_FLUX Local non-roaming Internet traffic (MB)
GN_ROAM_FLUX Domestic roaming Internet traffic (MB)
CALL_DAYS Call days CALLING_DAYS
Calling days CALLED_DAYS Called
days CALL_RING
Voice calling circle CALLING_RING
Calling calling circle
CALLED_RING Called calling circle
TERM_TYPE Terminal hardware type (0=indistinguishable, 4=4g, 3=3g, 2=2g)
IS_LOST Whether the user lost the mark in March (1=yes, 0=no), the value of January and February is empty
three, Mining goals
(1) Use the data of users' short messages, traffic, calls, consumption and basic customer information, and use data mining technology to group users.
(2) Analyze the usage patterns of different groups of users, and identify the important characteristics of customer churn in each group.
(3) Establish user churn models for different groups, and provide operators with differentiated opinions and suggestions based on the results.
http://www.daixie0.com/contents/21/1281.html
The core members of the team mainly include Silicon Valley engineers, BAT front-line engineers, top 5 master and doctoral students in China, and are proficient in German and English! Our main business scope is to do programming assignments, course design and so on.
Our field of direction: window programming, numerical algorithm, AI, artificial intelligence, financial statistics, econometric analysis, big data, network programming, WEB programming, communication programming, game programming, multimedia linux, plug-in programming program, API, image processing, embedded/MCU database programming, console process and thread, network security, assembly language hardware Programming software design engineering standards and regulations. The ghostwriting and ghostwriting programming languages or tools include but are not limited to the following:
C/C++/C# ghostwriting
Java ghostwriting
IT ghostwriting
Python ghostwriting
Tutored programming assignments
Matlab ghostwriting
Haskell ghostwriting
Processing ghostwriting
Building a Linux environment
Rust ghostwriting
Data Structure Assginment
MIPS ghostwriting
Machine Learning homework ghostwriting
Oracle/SQL/PostgreSQL/Pig database ghostwriting/doing/coaching
web development, website development, website work
ASP.NET website development
Finance Insurance Statistics Statistics, Regression, Iteration
Prolog ghostwriting
Computer Computational method
Because professional, so trustworthy. If necessary, please add QQ: 99515681 or email: [email protected]
WeChat: codinghelp