Latent mobile phone changing user discovery method based on multiple-model fusion

A discovery method and multi-model technology, applied in the field of potential replacement users based on multi-model fusion, which can solve problems such as meaningless user interference, difficulty in estimating marketing costs, and reducing brand reputation.

Inactive Publication Date: 2017-06-13
CHONGQING UNIV OF POSTS & TELECOMM
View PDF2 Cites 45 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

However, random pushes cause meaningless interference to most users
For merchants, spamming pushes may reduce brand reputation, and it is difficult to estimate marketing costs

Method used

the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
View more

Image

Smart Image Click on the blue labels to locate them in the text.
Viewing Examples
Smart Image
  • Latent mobile phone changing user discovery method based on multiple-model fusion
  • Latent mobile phone changing user discovery method based on multiple-model fusion
  • Latent mobile phone changing user discovery method based on multiple-model fusion

Examples

Experimental program
Comparison scheme
Effect test

Embodiment Construction

[0020] In order to further illustrate the solution of the present invention, the technical solution is described in detail with the user data of Guizhou Mobile from 2011 to 2016 as an example:

[0021] 1. Data Collection

[0022] The operator obtains the user's consumption information in the last year and various parameters of the user's historical model, including the processor, memory, pixels, and whether to support navigation, etc., and collects the user's basic information and user historical change track information.

[0023] 2. Data preprocessing

[0024] Including the processing of user basic information and user historical change track information, according to the description of the two data tables and physical understanding, the following processing is performed:

[0025] Due to the lack of some features in the user’s basic information, the numerical features and enumerated features are treated differently, such as flow, mou, apru and other numerical features. If there are mis...

the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
Login to view more

PUM

No PUM Login to view more

Abstract

The invention provides a latent mobile phone changing user discovery method based on multiple-model fusion. The method mainly comprises the steps of acquiring and preprocessing user consumption data and mobile phone changing track data, characteristic constructing, model establishing and model fusing, predicating, etc. According to the method of the invention, diversity and difference of a machine learning algorithm are used for combining a plurality of models, obtaining better effect and realizing higher generalization capability of the fused model. Training is performed through historical data of the user and predication is performed based on a data set, thereby obtaining high-value users who will change the mobile phone, and accurately pushing mobile phone information to the user.

Description

Technical field [0001] The invention relates to a data cleaning method, an abnormal data processing method, a user replacement marking method, a feature engineering construction method, and a technology that adopts multi-model fusion to perform replacement user prediction. Background technique [0002] Operators are generating massive business data almost every minute and every second. How to use this data and generate value has become an urgent need for operators. However, random pushes cause meaningless interference to most users. For merchants, spamming pushes may reduce brand reputation and it is difficult to estimate marketing costs. Based on data mining technology and methods, the accurate user consumption behavior and historical change track provided by the operator are used to portray user portraits, understand user needs, and make full use of data mining and machine learning technology to provide strong support for the business development of operators. [0003] Use data...

Claims

the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
Login to view more

Application Information

Patent Timeline
no application Login to view more
Patent Type & Authority Applications(China)
IPC IPC(8): G06Q10/04G06Q30/02G06K9/62
CPCG06Q10/04G06Q30/0201G06F18/22G06F18/24147G06F18/241G06F18/2411
Inventor 王进夏翠萍杨阳王鸿李智星邓欣陈乔松胡峰雷大江
Owner CHONGQING UNIV OF POSTS & TELECOMM
Who we serve
  • R&D Engineer
  • R&D Manager
  • IP Professional
Why Eureka
  • Industry Leading Data Capabilities
  • Powerful AI technology
  • Patent DNA Extraction
Social media
Try Eureka
PatSnap group products

Browse by: Latest US Patents, China's latest patents, Technical Efficacy Thesaurus, Application Domain, Technology Topic.

© 2024 PatSnap. All rights reserved.Legal|Privacy policy|Modern Slavery Act Transparency Statement|Sitemap