Natural person information setting method and electronic device

Patent metadata

Field	Value
Publication number	US-10824649-B2
Application number	US-201816018761-A
Country	US
Kind code	B2
Filing date	Jun 26, 2018
Priority date	Aug 16, 2012
Publication date	Nov 3, 2020
Grant date	Nov 3, 2020

Abstract

A natural person information setting method, an account information providing method, and electronic devices include: an associated set of each account is determined according to a degree of association and an exclusion index of another account with the each account; convergence processing is performed on all associated sets, such that after the convergence processing, no associated account in any one associated set appears in another associated set; and identical natural person information is set for any associated account in a same associated set having gone through the convergence processing, and a same natural person is set for any associated account in the same associated set according to the identical natural person information.

First claim

The invention claimed is:

1. A friend recommendation method, comprising:
receiving, by a sever, a request for friend recommendation from a terminal, and obtaining a list of strange accounts sharing common friends with an account of the terminal;
determining, by the server, friend accounts to be recommended for the terminal from the list of strange accounts according to the number of friend accounts shared by the account of the terminal and the strange accounts;
setting, by the server, natural person information of each friend account shared by the account of the terminal and the friend accounts to be recommended;
modifying, by the server, according to the natural person information, the number of friend accounts shared by the account of the terminal and the friend accounts to be recommended; and
providing, by the server, according to the modified number of friend accounts, the terminal with account information of the friend accounts to be recommended;
wherein the setting, by the server, natural person information of each friend account shared by the account of the terminal and the friend accounts to be recommended comprises:
for each friend account, determining an associated set of each account according to a degree of association and an exclusion index of another account with the each account;
performing convergence processing on all associated sets, such that after the convergence processing, no associated account in any one associated set appears in another associated set; and
setting identical natural person information for any associated account in a same associated set having gone through the convergence processing and setting a same natural person for any associated account in the same associated set according to the identical natural person information,
wherein the determining an associated set of each account according to a degree of association and an exclusion index of another account with the each account comprises:
filtering, according to a degree of association and an exclusion index of another account in a tentative associated set of an account for which natural person information is to be set with an account in the tentative associated set, a candidate associated account comprised in the tentative associated set to obtain an associated set of the account for which the natural person information is to be set, wherein the associated set of the account for which the natural person information is to be set comprises an associated account; and
successively taking the each account as the account for which the natural person information is to be set, and acquiring the associated set of the each account.

2. The method according to claim 1 , further comprising: determining, by the server, according to information on static association and information on dynamic association of the another account with the each account, the degree of association of the another account with the each account, wherein
the information on static association of the another account with the each account comprises at least one of: a similarity between personal data of the each account and personal data of the another account, a similarity between contact information of the each account and contact information of the another account, or a similarity between interest information of the each account and interest information of the another account; and
the information on dynamic association of the another account with the each account comprises at least one of: a similarity between login behavior information of the each account and login behavior information of the another account, or a similarity between online behavior information of the each account and online behavior information of the another account.

3. The method according to claim 1 , further comprising: determining, by the server, according to information on static exclusion and information on dynamic exclusion of the another account with the each account, the exclusion index of the another account with the each account, wherein
the information on static exclusion comprises a similarity between physical location information of the each account and physical location information of the another account; and
the information on dynamic exclusion comprises at least one of: a similarity between login behavior information of the each account and login behavior information of the another account, a similarity between online behavior information of the each account and online behavior information of the another account, or information on an interaction of the another account with the each account.

4. The method according to claim 1 , wherein the tentative associated set is determined according to the degree of association of the another account with the each account.

5. The method according to claim 4 , wherein the determining a tentative associated set of an account for which natural person information is to be set according to the degree of association of the another account with the each account comprises:
putting the account for which the natural person information is to be set into the tentative associated set;
determining a degree of association of another account with the account for which the natural person information is to be set, and, when it is determined that the degree of association of the another account with the account for which the natural person information is to be set is greater than a first set value, putting the another account into the tentative associated set; and
determining a degree of association of another account with an account put in the tentative associated set, when it is determined that the degree of association of the another account with the account put in the tentative associated set is greater than a first set value and that the another account has not been put in the tentative associated set, putting the another account into the tentative associated set, and repeating the determining and putting step until no more account is put in the tentative associated set, wherein the tentative associated set comprises the candidate associated account.

6. The method according to claim 4 , wherein the filtering, according to a degree of association and an exclusion index of another account in the tentative associated set with an account in the tentative associated set, the candidate associated account in the tentative associated set to obtain an associated set of the account for which the natural person information is to be set comprises:
determining whether the account for which the natural person information is to be set and the candidate associated account in the tentative associated set are owned by different owners according to a different-owner condition of
E>=p*D,

where D is the degree of association of the candidate associated account with the account for which the natural person information is to be set in the tentative associated set, E is the exclusion index of the candidate associated account with the account for which the natural person information is to be set in the tentative associated set, and p is a constant, with 0<p<1;
removing a candidate associated account meeting the different-owner condition from the tentative associated set; and
determining whether any two candidate associated accounts in the tentative associated set that have gone through the last determining and removing step are owned by different owners according to the different-owner condition of
E>=p*D,

where D is a degree of association between the two candidate associated accounts in the tentative associated set that are undergoing the current determining step, E is an exclusion index between the two candidate associated accounts in the tentative associated set that are undergoing the current determining step, and p is a con

Assignees

• Tencent Tech Shenzhen Co Ltd

Inventors

• Mai Junming
• Yi Lingling
• Chen Chuan
• Lei Xiaoping
• Ji tashan
• Liu Yuewen
• He Peng
• Li Yuhuang
• Chen Weihua

Classifications

• G06F16/285 Primary

  Clustering or classification · CPC title

• G06Q10/48

  using social graphs · CPC title

• G06Q10/42

  Determination of affinities or common interests between users · CPC title

• G06F16/215 Primary

  Improving data quality; Data cleansing, e.g. de-duplication, removing invalid entries or correcting typographical errors · CPC title

• G06F17/00

  Digital computing or data processing equipment or methods, specially adapted for specific functions (information retrieval, database structures or file system structures therefor G06F16/00) · CPC title