Unlock instant, AI-driven research and patent intelligence for your innovation.

Online learning-based network resource recommendation method and device

A network resource and recommendation method technology, applied in special data processing applications, instruments, electrical digital data processing, etc., can solve problems such as lack of recommendation, fixed recommended information, and many Internet resources, so as to improve the accuracy of recommendation , Improve the effect of reaction speed

Active Publication Date: 2015-02-04
BAIDU ONLINE NETWORK TECH (BEIJIBG) CO LTD
View PDF5 Cites 16 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

[0002] In the fields of video, search, film and television resource websites, Internet search engines, and recommendation engines (including news recommendations, product recommendations, advertising recommendations, etc.), the problem usually encountered is that there are many Internet resources, but the resource competition problem is only limited. Internet resources can be displayed to users, so how to choose appropriate Internet resources to recommend to users is an important issue
[0003] In the prior art, when a network device recommends Internet resources to users, it often calculates the weight of the resources according to user behavior and other information, and then selects the recommended information according to the weight ranking. However, in this existing Internet resource recommendation process, the inventor The following problems were found: the recommended information tends to be centralized and fixed, and some new or unpopular Internet resources cannot be recommended due to the lack of reference result information

Method used

the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
View more

Image

Smart Image Click on the blue labels to locate them in the text.
Viewing Examples
Smart Image
  • Online learning-based network resource recommendation method and device
  • Online learning-based network resource recommendation method and device
  • Online learning-based network resource recommendation method and device

Examples

Experimental program
Comparison scheme
Effect test

Embodiment 1

[0021] figure 1 It is a flow chart of the online learning-based network resource recommendation method provided by Embodiment 1 of the present invention. This embodiment is applicable to the situation of online recommendation of network resources, including but not limited to content providing websites that provide browsing and / or downloading of network resources Recommend network resources, such as film and television recommendations, application APP recommendations, etc. The method can be executed by a device for recommending network resources based on online learning configured in a server, such as figure 1 As shown, the network resource recommendation method based on online learning described in this embodiment includes:

[0022] S101. Train a data model for recommending network resources according to user feedback logs on displayed network resources.

[0023] The network resources referred to in this embodiment include multiple types of online resources, such as media f...

Embodiment 2

[0039] figure 2 It is a flow chart of the method for recommending network resources based on online learning described in Embodiment 2 of the present invention, as shown in figure 2 As shown, the network resource recommendation method based on online learning described in this embodiment includes:

[0040] S201. Obtain a user feedback log on the displayed network resource.

[0041] Wherein the feedback log includes user behavior information.

[0042] There are several ways to obtain a log of user feedback on a displayed web resource. For example, through real-time acquisition, that is, when the user has preset operations on the displayed network resources, including clicking, downloading, browsing, online installation, collection, and evaluation (such as comments, ratings, likes, etc.) of the displayed network resources by the user Record in real time when one or more preset behaviors occur. Another example is to obtain periodically, that is, to periodically extract the ...

Embodiment 3

[0086] image 3 It is a structural block diagram of the device for recommending network resources based on online learning described in Embodiment 3 of the present invention, as shown in image 3 As shown, the device for recommending network resources based on online learning described in this embodiment includes:

[0087] A model training unit 301, configured to train a data model for recommending network resources according to user feedback logs on displayed network resources;

[0088] A resource recommendation unit 302, configured to score and recommend network resources to be recommended based on the data model according to the E&E algorithm;

[0089] The resource display unit 303 is configured to display recommended network resources.

[0090] Further, the scoring and recommendation according to the E&E algorithm in the resource recommendation unit 302 specifically includes: according to the ε-greedy algorithm, the upper confidence limit UCB algorithm, the sampling devi...

the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
Login to View More

PUM

No PUM Login to View More

Abstract

The embodiment of the invention discloses an online learning-based network resource recommendation method and device. The method comprises the following steps: training a data model for recommending network resources according to feedback logs of users aiming at displayed network resources; scoring and recommending the network resources to be recommended based on the data model according to an E&E (Exploration and Exploitation) algorithm; displaying the recommended network resources. According to the embodiment of the invention, the data model is trained according to feedback of users to realize quick repetitive iteration of the data model, so that the response speed of a recommendation strategy and the accuracy for recommendation can be improved.

Description

technical field [0001] The invention relates to the technical field of the Internet, in particular to a method and device for recommending network resources based on online learning. Background technique [0002] In the fields of video, search, film and television resource websites, Internet search engines, and recommendation engines (including news recommendations, product recommendations, advertising recommendations, etc.), the problem usually encountered is that there are many Internet resources, but the resource competition problem is only limited. Internet resources can be displayed to users, so how to select appropriate Internet resources to recommend to users is a more important issue. [0003] In the prior art, when a network device recommends Internet resources to users, it often calculates the weight of the resources according to user behavior and other information, and then selects the recommended information according to the weight ranking. However, in this exist...

Claims

the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
Login to View More

Application Information

Patent Timeline
no application Login to View More
IPC IPC(8): G06F17/30
CPCG06F16/9535
Inventor 王凡田浩徐倩陈立信贤卫吴泽衡周方圆郑德荣
Owner BAIDU ONLINE NETWORK TECH (BEIJIBG) CO LTD