Mobile Internet-based big data mining method and system

Abstract

The invention relates to a mobile Internet-based big data mining method. The mobile Internet-based big data mining method includes the following steps that: updated customer social behavior data related to customer basic data are obtained; the customer social behavior data are standardized; the standardized data are utilized to establish or update a behavior category data model; a prediction algorithm is adopted to establish or update the prediction model of a customer behavior stage; and data indicating whether customer behaviors enter a follow-up stage are obtained and are adopted as validation data feedback, and the data model and the prediction model are outputted. According to the method of the invention, based on service features of each stage of service demands, a data processing process from data extraction, data modeling, prediction modeling to prediction analysis output feedback can be formed; the result of previous processing is received and is adopted as the input of the next iteration, the self-learning functions of the prediction model and the data model can be formed; and based on incrementally updated data content and data categories, flexible and reliable data processing processes for service prediction in different sectors can be realized. The invention also provides a mobile Internet-based big data mining system.

Description

technical field [0001] The invention relates to a data processing method and a data processing system, in particular to a discrete heterogeneous data processing method and a data processing system. Background technique [0002] In existing data analysis projects, the implicit correlation between data is not significant, which is often due to the time-discrete nature of data generation, inaccurate data generation time, and random drift within a certain period or period. Data is also heterogeneous, and data from different sources has different dimensions. Data normalization can reduce this difference but also cause information loss. When the collected data is based on the behavior of the data source, the process of data extraction, transfer, and loading cannot be organically combined with the communication, calculation, and storage architecture in computer technology, resulting in the inability to conduct effective data analysis on the massive objective behavior data in the re...

AI Technical Summary

Problems solved by technology

[0004] In view of this, the embodiment of the present invention provides a mobile Internet-based big data mining method to solve the technical problem that existing data processing cannot effectively process heterogeneous types of time-discretized behavior data to form objective behavior analysis and prediction

[0005] The embodiment of the present invention also provides a mobile Internet-based big data mining system, which solves the technical problem that existing data processing cannot effectively process heterogeneous time discretized behavior data to form objective behavior analysis and prediction

Images