System for estimating age of a user based on mass data

摘要

An initial weight is configured for each kind of basic age data, an age weight of the user in different basic age data is obtained according to the initial weight and age similarity of the user in different kinds of basic age data, and the age with the largest age weight is determined as the age of the user.

主权项

1. A system for estimating age of a user, comprising:
memory; one or more processors; and at least one unit stored in the memory for execution by the one or more processors, the at least one unit comprises: a weight configuring unit, to obtain different kinds of basic age data of the user from network products that comprise an instant messaging tool and a Social Networking Service (SNS), wherein the different kinds of basic age data are provided by the user by filling in ages through the different kinds of network products, obtain reference age data of the user through network questionnaire, compare the basic age data with the reference age data to obtain an accuracy ratio of the basic age data; and configure an initial weight for the basic age data according to the accuracy ratio; a weight processing unit, to obtain an age weight of the user in each kind of basic age data according to a sum of the initial weight and an age weight score of the kind of basic age data; wherein the age weight score of the kind of basic age data is configured according to the initial weight of the kind of the basic age data and an age similarity of the user in the different kinds of basic age data; and an age estimating unit, to search the different kinds of basic age data for the basic age data with a largest age weight, and estimate the age of the user according to the basic age data with the largest age weight; wherein the accuracy ratio of the basic age data comprises searching the basic age data for a selected group of users whose ages conform to the reference age data, dividing the number of the users searched out by the total number of users in the basic age data, and determining the divided result as the accuracy ratio of the basic age data.