Systems and methods for online content recommendation

Patent metadata

Field	Value
Publication number	US-2016371589-A1
Application number	US-201514748333-A
Country	US
Kind code	A1
Filing date	Jun 24, 2015
Priority date	Jun 17, 2015
Publication date	Dec 22, 2016
Grant date	—

Abstract

The present disclosure relates to computer systems implementing methods for online content recommendation. The computer systems may be configured to receive a training sample from a first client device corresponding to a predefined feedback interacting with online content displayed on the first client device; update a preexisting training database in real-time based on the received training sample to generate an updated training sample, wherein prior to being updated based on the training sample received from the first client, the training database includes a set of historical training samples; conduct a regression training to a computer learning model in real-time, using the updated training sample, to produce a set of trained parameters for an online content recommendation model; call the set of trained parameters in real-time to determine recommend online content for a second user with the online content recommendation model; and send the recommended online content to a second client device of the second user.

First claim

1 . A computer system, comprising:
a storage medium comprising a set of instructions for online content recommendation; and a processor in communication with the storage medium, wherein when executing the set of instructions, the processor is directed to:
receive a training sample from a first client device in communication with the computer system, the training sample corresponding to a predefined feedback action taken by a first user interacting with online content displayed on the first client device;
update a preexisting training database in real-time based on the received training sample to generate an updated training sample, wherein prior to being updated based on the training sample received from the first client, the training database includes a set of historical training samples;
conduct a regression training to a computer learning model in real-time, using the updated training sample, to produce a set of trained parameters for an online content recommendation model;
call the set of trained parameters in real-time to determine recommend online content for a second user with the online content recommendation model; and
send the recommended online content to a second client device of the second user.

2 . The computer system of claim 1 , wherein to receive the training sample, the processor is further directed to:
receive in real-time a user request for the online content from the first client device; identify and request features from the user request; cache the request features for a predetermined amount of time; receive in real-time a predefined feedback over the online content from the first client device; and when the predefined feedback is received within the predetermined amount of time from the receipt of the user request, combine in real-time the request features and the predefined feedback as the training sample.

3 . The computer system of claim 2 , wherein the user request is received from the first client device when the first client device opens a user interface to load the online content.

4 . The computer system of claim 2 , wherein the request features comprise at least one of demographic information of the user, device parameters of the client device, and a timestamp showing a time that the user request occurs.

5 . The computer system of claim 2 , wherein the predefined feedback comprises at least one of a click of the online content, a “like” or “dislike” action associated with the online content, and sharing the online content and/or forwarding to a website.

6 . The computer system of claim 1 , wherein the set of trained parameters of the content recommendation model is output every predetermined number of iterations of the regression training.

7 . The computer system of claim 1 , wherein to determine the recommended online content, the processor is further directed to:
call the set of trained parameters of the content recommendation model; based on the request features and for each item of an plurality of online items, determine a probability value that the user will perform a predefined feedback action over the online item; select a predetermined number of online items from the plurality of online items that have highest probability values; and identify the selected online items as the recommended online content to display on the second client device.

8 . A method for online content recommendation, comprising:
receiving, by a computer system, a training sample from a first client device in communication with the computer system, the training sample corresponding to a predefined feedback action taken by a first user interacting with online content displayed on the first client device; updating, by the computer system, a preexisting training database in real-time based on the received training sample to generate an updated training sample, wherein prior to being updated base don the training sample received from the first client, the training database includes a set of historical training samples; conducting, by the computer system, a regression training to a computer learning model in real-time, using the updated training sample, to produce a set of trained parameters for an online content recommendation model; calling the set of trained parameters in real-time to determine recommended online content for a second user with the online content recommendation model; and sending, by the computer system, the recommended online content to a second client device of the second user.

9 . The method of claim 8 , wherein the receiving of the training sample comprising:
receiving in real-time a user request for the online content from the first client device; identifying requesting features from the user request; caching the request features for a predetermined amount of time; receiving in real-time a predefined feedback over the online content from the first client device; and when the predefined feedback is received within the predetermined amount of time from the receipt of the user request, combining in real-time the request features and the received feedback as the training sample.

10 . The method of claim 9 , wherein the user request is received from the first client device when the first client device opens a user interface to load the online content.

11 . The method of claim 9 , wherein the request features comprise at least one of demographic information of the user, device parameters of the client device, and a timestamp showing a time that the user request occurs.

12 . The method of claim 9 , wherein the predefined feedback comprises at least one of a click of the online content, a “like” or “dislike” action associated with the online content, and sharing the online content and/or forwarding to a website.

13 . The method of claim 8 , wherein the set of trained parameters of the content recommendation model is output every predetermined number of iterations of the regression training.

14 . The method of claim 8 , wherein the determining of the recommended online content comprises:
calling the set of trained parameters of the content recommendation model; based on the request features and for each item of an plurality of online items, determining a probability value that the user will perform a predetermined action over the online item; selecting a predetermined number of online items from the plurality of online items that have highest probability values; and identifying the selected online items as the recommended online content to display on the second client device.

15 . A computer-readable non-transitory storage medium, comprising a set of instructions for online content recommendation, wherein the set of instructions, when executed by a processor of a computer system, directs the processor to perform operations:
receiving a training sample from a first client device in communication with the computer system, the training sample corresponding to a predefined feedback action taken by a first user interacting with online content displayed on the first client device; updating a preexisting training database in real-time based on the received training sample to generate an updated training sample, wherein prior to being updated based on the training sample received from the first client, the training database includes a set of historical training samples; conducting a regression training to a computer learning model in real-time, using the updated training sample to produce a set of trained parameters for an online content recommendation model; call the set of trained parameters in real-time to determine recommended online content for a s

Assignees

• Yahoo Inc

Inventors

• Golbandi Nadav
• Wang Chao

Classifications

• G06N5/04 Primary

  Inference or reasoning models · CPC title

• G06N7/005

  Physics · mapped topic

• G06N99/005

  Physics · mapped topic

• G06F16/9535

  Search customisation based on user profiles and personalisation · CPC title

• G06Q30/0241

  Advertisements · CPC title