IP Library › Granted Patent US 12,481,706
Granted Patent B2
US 12,481,706 · App. 17/303,060 · Granted Nov 25, 2025

Method and apparatus for sending information

Inventors: Lijuan Cui (Beijing, CN); Yudong Li (Beijing, CN)
Assignee: Baidu Online Network Technology (Beijing) Co., Ltd.
G06F16/9038G06F16/904G06F16/906G06F16/908G06F16/951G06F16/9538G06F18/22G06N20/00
View Patent ↗
Loading inventors, assignments & file history…
Monitor This Case
Get email alerts when status or documents change.
Order Certified Copies
Most orders are placed with the USPTO same day — all within 24 business hours.
Order via The Patent Place →
Pre-filled with this patent's details
Quick Facts
Patent No.
US 12,481,706
App. No.
17/303,060
Granted
Nov 25, 2025
Kind
B2
Abstract

Embodiments of the present application provides a method and apparatus for sending information. A specific implementation of the method includes: identifying search information input through a terminal by a user, to obtain the user's demand; acquiring, based on the user's demand, at least one information card from at least one category of preset information set, where the information set includes information and information cards used to describe the information; performing information search based on the search information, to obtain a search result; screening, based on feature information of the search result, the at least one information card to obtain a target information card; and sending to-be-displayed information to the terminal based on the target information card and the search result.

Claims (75)

1 . A method for sending information, the method being performed by a processor, the method comprising:

identifying search information input through a terminal to obtain a user demand;

acquiring, based on the user demand, a plurality of information cards from at least one category of preset information set, wherein a preset information set comprises information and information cards used to describe the information, wherein an information card is a result of integrating and visualizing various kinds of information, and the information card is an entry to the various kinds of information and contains a picture and a play link;

searching information from Internet data collected by web crawlers based on the search information, to obtain a search result from the Internet data, the search result is a result of a text category, and determining feature information of the search result, wherein the feature information of the search result comprises locations of a keyword in a title and an abstract of the search result and statistics on a number of occurrences of the keyword in the title and the abstract of the search result;

screening, based on feature information of the search result obtained from the Internet data, the plurality of information cards to obtain a target information card; and

sending the target information card and the search result to the terminal as to-be-displayed information, and displaying on the terminal the target information card and the search result, so that various kinds of information integrated by the target information card is accessible through the target information card displayed on the terminal,

wherein the screening, based on the feature information of the search result obtained from the Internet data, the plurality of information cards to obtain the target information card comprises:

performing a first screening on an information card of the plurality of information cards, comprising:

inputting the feature information of the search result, statistics for categories of the plurality of information cards, and a category of the information card into a to-be-used screening model, to obtain screening information for the information card, wherein the to-be-used screening model is pre-trained with a preset sample set by using a machine learning method, the statistics for categories of the plurality of information cards includes numbers and proportions of the categories of the plurality of information cards, wherein the screening information for the information card is display or non-display; and

determining that the information card is the target information card, in response to determining that the screening information for the information card is display.

2 . The method of claim 1 , wherein the performing the first screening comprises:

determining, based on the category of the information card, the to-be-used screening model from at least one screening model trained in advance.

3 . The method of claim 1 , wherein the determining whether the information card is the target information card based on the obtained screening information for the information card comprises:

determining that the information card is not the target information card, in response to determining that the screening information for the information card is non-display.

4 . The method of claim 1 , wherein the screening, based on the feature information of the search result, the plurality of information cards to obtain the target information card comprises:

performing a second screening on an information card of the plurality of information cards, comprising:

matching the feature information and the category of the information card with a preset feature combination; and

determining whether the information card is the target information card based on result of the matching.

5 . The method of claim 1 , wherein the sending the target information card and the search result to the terminal as the to-be-displayed information comprises:

calculating a similarity between information contained in the target information card and the search information, and calculating a similarity between the search result and the search information;

determining attribute information of the target information card and attribute information of the search result, wherein different scores are set for different attribute information;

performing weighted sum on correlations of each target information card and the search result and scores corresponding to the attribute information, to obtain a result of the scoring;

sorting the target information card and the search result according to results of the scoring; and

sending the sorted target information card and search result to the terminal as the to-be-displayed information.

6 . The method of claim 1 , wherein the searching information from Internet data based on the search information to obtain the search result comprise:

segmenting the search information to obtain at least one segmented word; and

searching information from the Internet data based on the obtained at least one segmented word, to obtain the search result.

7 . A server, comprising:

one or more processors;

a storage device, on which one or more programs are stored,

the one or more programs, when executed by the one or more processors, cause the one or more processors to implement operations comprising:

identifying search information input through a terminal to obtain a user demand;

acquiring, based on the user demand, a plurality of information cards from at least one category of preset information set, wherein a preset information set comprises information and information cards used to describe the information, wherein an information card is a result of integrating and visualizing various kinds of information, and the information card is an entry to the various kinds of information and contains a picture and a play link;

searching information from Internet data collected by web crawlers based on the search information, to obtain a search result from the Internet data, the search result is a result of a text category, and determining feature information of the search result, wherein the feature information of the search result comprises locations of a keyword in a title and an abstract of the search result and statistics on a number of occurrences of the keyword in the title and the abstract of the search result;

screening, based on feature information of the search result obtained from the Internet data, the plurality of information cards to obtain a target information card; and

sending the target information card and the search result to the terminal as to-be-displayed information, and displaying on the terminal the target information card and the search result, so that various kinds of information integrated by the target information card is accessible through the target information card displayed on the terminal,

wherein the screening, based on the feature information of the search result obtained from the Internet data, the plurality of information cards to obtain the target information card comprises:

performing a first screening on an information card of the plurality of information cards, comprising:

inputting the feature information of the search result, statistics for categories of the plurality of information cards, and a category of the information card into a to-be-used screening model, to obtain screening information for the information card, wherein the to-be-used screening model is pre-trained with a preset sample set by using a machine learning method, the statistics for categories of the plurality of information cards includes numbers and proportions of the categories of the plurality of information cards, wherein the screening information for the information card is display or non-display; and

determining that the information card is the target information card, in response to determining that the screening information for the information card is display.

8 . The server of claim 7 , wherein the performing the first screening comprises:

determining, based on the category of the information card, the to-be-used screening model from at least one screening model trained in advance.

9 . The server of claim 7 , wherein the determining whether the information card is the target information card based on the obtained screening information for the information card comprises:

determining that the information card is not the target information card, in response to determining that the screening information for the information card is non-display.

10 . The server of claim 7 , wherein the screening, based on the feature information of the search result, the plurality of information cards to obtain the target information card comprises:

performing a second screening on an information card of the plurality of information cards, comprising:

matching the feature information and the category of the information card with a preset feature combination; and

determining whether the information card is the target information card based on result of the matching.

11 . The server of claim 7 , wherein the sending the target information card and the search result to the terminal as the to-be-displayed information comprises:

calculating a similarity between information contained in the target information card and the search information, and calculating a similarity between the search result and the search information;

determining attribute information of the target information card and attribute information of the search result, wherein different scores are set for different attribute information;

performing weighted sum on correlations of each target information card and the search result and scores corresponding to the attribute information, to obtain a result of the scoring;

sorting the target information card and the search result according to results of the scoring; and

sending the sorted target information card and search result to the terminal as the to-be-displayed information.

12 . The server of claim 7 , wherein the searching information from Internet data based on the search information to obtain the search result comprise:

segmenting the search information to obtain at least one segmented word; and

searching information from the Internet data based on the obtained at least one segmented word, to obtain the search result.

13 . A non-transitory computer readable medium storing one or more computer programs that, when executed by a processor, cause the processor to implement operations, the operations comprising:

identifying search information input through a terminal to obtain a user demand;

acquiring, based on the user demand, a plurality of information cards from at least one category of preset information set, wherein a preset information set comprises information and information cards used to describe the information, wherein an information card is a result of integrating and visualizing various kinds of information, and the information card is an entry to the various kinds of information and contains a picture and a play link;

searching information from Internet data collected by web crawlers based on the search information, to obtain a search result from the Internet data, the search result is a result of a text category, and determining feature information of the search result, wherein the feature information of the search result comprises locations of a keyword in a title and an abstract of the search result and statistics on a number of occurrences of the keyword in the title and the abstract of the search result;

screening, based on feature information of the search result obtained from the Internet data, the plurality information cards to obtain a target information card; and

sending the target information card and the search result to the terminal as to-be-displayed information, and displaying on the terminal the target information card and the search result, so that various kinds of information integrated by the target information card is accessible through the target information card displayed on the terminal,

wherein the screening, based on the feature information of the search result obtained from the Internet data, the plurality of information cards to obtain the target information card comprises:

performing a first screening on an information card of the plurality of information cards, comprising:

inputting the feature information of the search result, statistics for categories of the plurality of information cards, and a category of the information card into a to-be-used screening model, to obtain screening information for the information card, wherein the to-be-used screening model is pre-trained with a preset sample set by using a machine learning method, the statistics for categories of the plurality of information cards includes numbers and proportions of the categories of the plurality of information cards, wherein the screening information for the information card is display or non-display; and

determining that the information card is the target information card, in response to determining that the screening information for the information card is display.

14 . The medium of claim 13 , wherein the performing the first screening comprises:

determining, based on the category of the information card, the to-be-used screening model from at least one screening model trained in advance.

15 . The medium of claim 13 , wherein the determining whether the information card is the target information card based on the obtained screening information for the information card comprises:

determining that the information card is not the target information card, in response to determining that the screening information for the information card is non-display.

16 . The medium of claim 13 , wherein the screening, based on the feature information of the search result, the plurality of information cards to obtain the target information card comprises:

performing a second screening on an information card of the plurality of information cards, comprising:

matching the feature information and the category of the information card with a preset feature combination; and

determining whether the information card is the target information card based on result of the matching.

Assignments (1)
ASSIGNMENT OF ASSIGNOR'S INTEREST Recorded Jul 22, 2021
From: CUI, LIJUAN; LI, YUDONG
To: BAIDU ONLINE NETWORK TECHNOLOGY (BEIJING) CO., LTD.
Reel/Frame 056946/0600 →
Priority Claims (1)
CN 202010247322.1 · Mar 31, 2020 · national
Continuity (2)
Continuation PCTCN2020116528 · Sep 21, 2020
Related Publication 20210271720A1 · Sep 2, 2021
References Cited (55)
US 7603348B2 · He · 2009 [cited by examiner]
US 9213765B2 · Wong et al. · 2015 [cited by applicant]
US 9996626B1 · Bailey · 2018 [cited by examiner]
US 10410224B1 · Levanon · 2019 [cited by examiner]
US 10620801B1 · Fein · 2020 [cited by examiner]
US 10706450B1 · Tavernier · 2020 [cited by examiner]
US 10733197B2 · Lin et al. · 2020 [cited by applicant]
US 20100223257A1 · Milic-Frayling · 2010 [cited by examiner]
US 20120076414A1 · Xu · 2012 [cited by examiner]
US 20130060755A1 · He · 2013 [cited by examiner]
US 20130283140A1 · Wu · 2013 [cited by examiner]
US 20140172652A1 · Pobbathi · 2014 [cited by examiner]
US 20150108213A1 · Sasaki · 2015 [cited by examiner]
US 20150278266A1 · Tang · 2015 [cited by examiner]
US 20150324361A1 · Glass et al. · 2015 [cited by applicant]
US 20170371883A1 · Bailiang · 2017 [cited by examiner]
US 20180342003A1 · Siddiqui · 2018 [cited by examiner]
US 20190325062A1 · Rogulenko · 2019 [cited by examiner]
US 20200143102A1 · Ziraknejad · 2020 [cited by examiner]
US 20200159861A1 · Baeck · 2020 [cited by examiner]
US 20200210647A1 · Panuganty · 2020 [cited by examiner]
US 20200226169A1 · Zhang · 2020 [cited by examiner]
US 20210049165A1 · Najork · 2021 [cited by examiner]
US 20210118024A1 · Sollami · 2021 [cited by examiner]
US 20210133269A1 · Shah · 2021 [cited by examiner]
US 20210224321A1 · Chiarandini · 2021 [cited by examiner]
US 20220222418A1 · Ziraknejad · 2022 [cited by examiner]
CN 101052181 · 2007 [cited by applicant]
CN 101145153 · 2008 [cited by applicant]
CN 101331475 · 2008 [cited by applicant]
CN 101477557 · 2009 [cited by applicant]
CN 101694656 · 2010 [cited by applicant]
CN 104239481 · 2014 [cited by applicant]
CN 104503980 · 2015 [cited by applicant]
CN 105138535 · 2015 [cited by applicant]
CN 110765275 · 2015 [cited by applicant]
CN 105354225 · 2016 [cited by applicant]
CN 105426436 · 2016 [cited by applicant]
CN 105786969 · 2016 [cited by applicant]
CN 105893343 · 2016 [cited by applicant]
CN 106649780 · 2017 [cited by applicant]
CN 107577807 · 2018 [cited by applicant]
CN 107885872 · 2018 [cited by applicant]
CN 107992602 · 2018 [cited by applicant]
CN 109344336 · 2019 [cited by applicant]
CN 109522465 · 2019 [cited by applicant]
CN 109949124 · 2019 [cited by applicant]
CN 110309431 · 2019 [cited by applicant]
CN 111475722 · 2020 [cited by applicant]
EP 2740053A1 · 2014 [cited by applicant]
EP 2740053B1 · 2019 [cited by examiner]
JP 2014522056 · 2014 [cited by applicant]
JP 2015153158 · 2015 [cited by applicant]
May 16, 2022 Extended European Search Report from EP 20886169.0 (8 pgs). [cited by applicant]
International Search Report of Sep. 21, 2020 for International Patent Application No. PCT/CN2020/116528. 11 pages. [cited by applicant]