Matching products with service scenarios

Patent metadata

Field	Value
Publication number	US-2021272153-A1
Application number	US-202117322894-A
Country	US
Kind code	A1
Filing date	May 17, 2021
Priority date	May 4, 2015
Publication date	Sep 2, 2021
Grant date	—

Abstract

For each service scenario out of a plurality of service scenarios, matching features of a to-be-matched product corresponding to the service scenario are acquired based on user features of users accessing the service scenario. A respective user feature mapping value of the service scenario is calculated based on the matching features of the to-be-matched product corresponding to the service scenario. Out of the plurality of service scenarios, a target service scenario of the to-be-matched product is selected based on the respective user feature mapping value of the service scenario.

First claim

1 .- 20 . (canceled)

21 . A computer-implemented method, comprising:
acquiring, by data processing apparatus, for each service scenario of a plurality of service scenarios, respective representative features of a plurality of users who have accessed the service scenario; obtaining, by the data processing apparatus, features of a to-be-matched product to be presented to users accessing one of the service scenarios; calculating, by the data processing apparatus, a respective user feature mapping value of each service scenario of the plurality of the service scenarios based on matching features of the to-be-matched product corresponding to the service scenario to respective representative features of the plurality of the users who have accessed the service scenario, comprising:
quantifying a first user feature corresponding to a first matching feature of the to-be-matched product corresponding to each service scenario according to a quantification rule; and
calculating the respective user feature mapping value of each service scenario based on the first user feature corresponding to the first matching feature according to a mapping rule; and
selecting, by the data processing apparatus, among the plurality of the service scenarios, a target service scenario for the to-be-matched product based on the respective user feature mapping value of the target service scenario.

22 . The computer-implemented method of claim 21 , wherein the respective representative features of the plurality of the users who have accessed each service scenario comprises features derived from registration information or an access record of the plurality of the users who have accessed each service scenario.

23 . The computer-implemented method of claim 21 , wherein each service scenario is a different respective access interface for accessing a different respective service of a web site.

24 . The computer-implemented method of claim 21 , wherein the quantification rule is the same for the plurality of the service scenarios, and wherein the mapping rule is the same for the plurality of the service scenarios.

25 . The computer-implemented method of claim 21 , wherein selecting, among the plurality of the service scenarios, the target service scenario for the to-be-matched product based on the respective user feature mapping value of the target service scenario comprises:
selecting a service scenario having a lowest user feature mapping value among a plurality of respective user feature mapping values of the plurality of the service scenarios as the target service scenario for the to-be-matched product; or selecting a service scenario having a highest user feature mapping value among a plurality of respective user feature mapping values of the plurality of the service scenarios as the target service scenario for the to-be-matched product.

26 . The computer-implemented method of claim 21 , wherein obtaining features of the to-be-matched product to be presented to the users accessing one of the service scenarios comprises acquiring the matching features of the to-be-matched product corresponding to one of the service scenarios based on the respective representative features of the plurality of the users who have accessed the service scenarios according to a machine learning algorithm.

27 . The computer-implemented method of claim 26 , wherein the machine learning algorithm comprises at least one or more of a logistic regression algorithm, a Gradient Boosting Decision Tree (GBDT) algorithm, a decision tree algorithm, or a deep learning algorithm.

28 . The computer-implemented method of claim 21 , wherein calculating the respective user feature mapping value of each service scenario based on the matching features of the to-be-matched product corresponding to each service scenario comprises:
obtaining a mapping function between the respective user feature mapping value of each service scenario and the matching features of the to-be-matched product corresponding to each service scenario using a machine learning method; and calculating the respective user feature mapping value of each service scenario based on the mapping function.

29 . A non-transitory, computer-readable medium storing one or more instructions executable by a computer system to perform operations comprising:
acquiring, by data processing apparatus, for each service scenario of a plurality of service scenarios, respective representative features of a plurality of users who have accessed the service scenario; obtaining, by the data processing apparatus, features of a to-be-matched product to be presented to users accessing one of the service scenarios; calculating, by the data processing apparatus, a respective user feature mapping value of each service scenario of the plurality of the service scenarios based on matching features of the to-be-matched product corresponding to the service scenario to respective representative features of the plurality of the users who have accessed the service scenario, comprising:
quantifying a first user feature corresponding to a first matching feature of the to-be-matched product corresponding to each service scenario according to a quantification rule; and
calculating the respective user feature mapping value of each service scenario based on the first user feature corresponding to the first matching feature according to a mapping rule; and
selecting, by the data processing apparatus, among the plurality of the service scenarios, a target service scenario for the to-be-matched product based on the respective user feature mapping value of the target service scenario.

30 . The computer-readable medium of claim 29 , wherein the respective representative features of the plurality of the users who have accessed each service scenario comprises features derived from registration information or an access record of the plurality of the users who have accessed each service scenario.

31 . The computer-readable medium of claim 29 , wherein each service scenario is a different respective access interface for accessing a different respective service of a website.

32 . The computer-readable medium of claim 29 , wherein the quantification rule is the same for the plurality of the service scenarios, and wherein the mapping rule is the same for the plurality of the service scenarios.

33 . The computer-readable medium of claim 29 , wherein selecting, among the plurality of the service scenarios, the target service scenario for the to-be-matched product based on the respective user feature mapping value of the target service scenario comprises:
selecting a service scenario having a lowest user feature mapping value among a plurality of respective user feature mapping values of the plurality of the service scenarios as the target service scenario for the to-be-matched product; or selecting a service scenario having a highest user feature mapping value among a plurality of respective user feature mapping values of the plurality of the service scenarios as the target service scenario for the to-be-matched product.

34 . The computer-readable medium of claim 29 , wherein obtaining features of the to-be-matched product to be presented to the users accessing one of the service scenarios comprises acquiring the matching features of the to-be-matched product corresponding to one of the service scenarios based on the respective representative features of the plurality of the users who have accessed the service scenarios according to a machine learning algorithm.

35 . The computer-readable medium of claim 34 , wherein the machine learning al

Assignees

• Advanced New Technologies Co Ltd

Inventors

• Shen Dong
• Xiao Hanping
• Liu Tangheng
• Ye Jiajie

Classifications

• G06F16/00

  Information retrieval; Database structures therefor; File system structures therefor · CPC title

• G06Q30/00

  Commerce · CPC title

• G06Q50/10

  Services · CPC title

• G06Q30/0601

  Electronic shopping [e-shopping] · CPC title

• G06F16/957

  Browsing optimisation, e.g. caching or content distillation · CPC title