Granted Patent B2

US 12,511,484 · App. 18/031,461 · Granted Dec 30, 2025

Inventors: Jianwei Fang (Shanghai, CN); Qi Xiang (Shanghai, CN); Ping Xu (Shanghai, CN)

Assignee: PATEO CONNECT+ TECHNOLOGY (SHANGHAI) CORPORATION

G06F40/295G06F16/9537G06F16/9558G06F40/242

Abstract

Claims (101)

1 . A point-of-interest (POI) information management method, comprising:

obtaining a webpage link, the webpage link being from a first application;

obtaining, at a second application, a webpage text associated with the webpage link;

obtaining at least one item of POI information matching the webpage text from a POI information repository;

presenting the at least one item of the POI information;

in response to a pre-determined operation being performed on a first item of the POI information of the at least one item of the POI information, presenting a route from a current location to an address indicated in the first item of the POI information and a first operable icon; and

in response to detecting another pre-determined operation is performed on the first operable icon, setting the route as a target trip,

wherein the obtaining the at least one item of the POI information matching the webpage text from the POI information repository includes:

determining a content category of the webpage text;

in response to the content category being determined to be a second content category, determining at least one word tagged with a pre-determined tag from the webpage text based on an entity recognition model; and

obtaining the POI information associated with the at least one word from the POI information repository to be the at least one item of the POI information.

2 . The method according to claim 1 , wherein obtaining the at least one item of the POI information matching the webpage text from the POI information repository includes:

in response to the content category being determined to be a first content category, obtaining a set of address information from the webpage text based on an address matching rule;

obtaining items of the POI information matching the set of address information from the POI information repository, the items of the POI information including a plurality of POI identifiers;

determining at least one POI identifier that matches the webpage text from the plurality of POI identifiers based on a natural language processing model; and

obtaining the POI information associated with the at least one POI identifier from the items of the POI information to be the at least one item of the POI information.

3 . The method according to claim 1 , wherein determining the content category of the webpage text includes:

determining a first frequency of a first term associated with the first content category and a second frequency of a second term associated with the second content category in the webpage text; and

in response to determination of the first frequency being greater than the second frequency, determining the content category to be the first content category; or

in response to determination of the first frequency being smaller than the second frequency, determining the content category to be the second content category; or

in response to determination of the first frequency being equal to the second frequency, obtaining a set of words matching a word database from the webpage text and inputting the set of words into a text classification model to determine whether the content category is the first content category or the second content category.

4 . The method according to claim 3 , wherein determining the first frequency and the second frequency includes:

obtaining title information from the webpage text; and

determining the first frequency of the first term associated with the first content category and the second frequency of the second term associated with the second content category in the title information.

5 . The method according to claim 1 , wherein determining the at least one POI identifier that matches the webpage text from the plurality of POI identifiers based on the natural language processing model includes:

generating a dictionary based on the plurality of POI identifiers;

performing a word segmentation process on the webpage text according to the dictionary based on the natural language processing model to obtain a word segmentation result; and

determining the at least one POI identifier that matches the word segmentation result in the dictionary.

6 . The method according to claim 5 , wherein determining the at least one POI identifier that matches the webpage text from the plurality of POI identifiers based on the natural language processing model further includes:

determining at least one frequency of the at least one POI identifier in the word segmentation result; and

according to the at least one frequency, sorting the at least one POI identifier.

7 . The method according to claim 6 , wherein determining the at least one POI identifier that matches the webpage text from the plurality of POI identifiers based on the natural language processing model further includes:

deleting the POI identifier whose frequency in the word segmentation result is lower than a pre-determined frequency threshold from the at least one POI identifier that has been sorted.

8 . The method according to claim 1 , wherein obtaining the POI information associated with the at least one word from the POI information repository to be the at least one item of the POI information includes:

determining region identifiers tagged as location tags from the webpage text based on the entity recognition model;

determining a first region identifier with the highest frequency occurring in the webpage text from the region identifiers; and

obtaining from the POI information repository the at least one item of the POI information that is located within a region associated with the first region identifier and that matches the at least one word.

9 . The method according to claim 1 , further comprising:

obtaining an image associated with the webpage link;

obtaining text information from the image based on a text recognition model;

obtaining a first POI location matching the text information from the POI information repository;

determining at least one first credibility score associated with the at least one item of the POI information based on a distance between the at least one POI location in the at least one item of the POI information and the first POI location;

obtaining at least one item of user evaluation data associated with the at least one item of the POI information from a pre-determined server;

based on the at least one item of user evaluation data, determining at least one second credibility score associated with the at least one item of the POI information;

based on the at least one first credibility score and the at least one second credibility score, determining at least one final credibility score associated with the at least one item of the POI information; and

presenting the at least one final credibility score.

10 . The method according to claim 9 , wherein presenting the at least one item of the POI information includes:

based on the at least one final credibility score, sorting the at least one item of the POI information to obtain a sorted result; and

presenting the sorted result.

11 . The method according to claim 1 , further comprising:

presenting at least one selectable icon associated with the at least one item of the POI information; and

in response to a selection operation being performed on a first selectable icon of the at least one selectable icon, adding the first item of the POI information associated with the first selectable icon to a favorite POI list.

12 . The method according to claim 1 , wherein the entity recognition model comprises a bidirectional long-short-term memory network (BiLSTM) and a conditional random field (CRF).

13 . The method according to claim 1 , wherein a first content category comprises a food-related category, and the second content category comprises an attractions-related category.

14 . The method according to claim 1 , wherein the webpage link is obtained via a clipboard or a sharing function.

15 . An electronic device, comprising:

at least one processor; and

a memory coupled to the at least one processor;

wherein the memory stores instructions being executed by the at least one processor, and when being executed by the at least one processor, the instructions cause the at least one processor to perform:

obtaining a webpage link, the webpage link being from a first application;

obtaining, at a second application, a webpage text associated with the webpage link;

obtaining at least one item of POI information matching the webpage text from a POI information repository;

presenting the at least one item of the POI information;

in response to a pre-determined operation being performed on a first item of the POI information of the at least one item of the POI information, presenting a route from a current location to an address indicated in the first item of the POI information and a first operable icon; and

in response to detecting another pre-determined operation is performed on the first operable icon, setting the route as a target trip,

wherein when obtaining the at least one item of the POI information matching the webpage text from the POI information repository, the at least one processor is further configured to:

determine a content category of the webpage text;

in response to the content category being determined to be a second content category, determine at least one word tagged with a pre-determined tag from the webpage text based on an entity recognition model; and

obtain the POI information associated with the at least one word from the POI information repository to be the at least one item of the POI information.

16 . The electronic device according to claim 15 , wherein when obtaining the at least one item of the POI information matching the webpage text from the POI information repository, the at least one processor is further configured to:

in response to the content category being determined to be a first content category, obtain a set of address information from the webpage text based on an address matching rule;

obtain items of the POI information matching the set of address information from the POI information repository, the items of the POI information including a plurality of POI identifiers;

determine at least one POI identifier that matches the webpage text from the plurality of POI identifiers based on a natural language processing model; and

obtain the POI information associated with the at least one POI identifier from the items of the POI information to be the at least one item of the POI information.

17 . The electronic device according to claim 16 , wherein when determining the content category of the webpage text, the at least one processor is further configured to:

determine a first frequency of a first term associated with the first content category and a second frequency of a second term associated with the second content category in the webpage text; and

in response to determination of the first frequency being greater than the second frequency, determine the content category to be the first content category; or

in response to determination of the first frequency being smaller than the second frequency, determine the content category to be the second content category; or

in response to determination of the first frequency being equal to the second frequency, obtain a set of words matching a word database from the webpage text and inputting the set of words into a text classification model to determine whether the content category is the first content category or the second content category.

18 . A non-volatile computer-readable storage medium storing computer instructions, wherein:

when being executed by at least one processor, the instructions cause the at least one processor to perform:

obtaining a webpage link, the webpage link being from a first application;

obtaining, at a second application, a webpage text associated with the webpage link;

obtaining at least one item of POI information matching the webpage text from a POI information repository;

presenting the at least one item of the POI information;

in response to a pre-determined operation being performed on a first item of the POI information of the at least one item of the POI information, presenting a route from a current location to an address indicated in the first item of the POI information and a first operable icon; and

in response to detecting another pre-determined operation is performed on the first operable icon, setting the route as a target trip,

wherein when obtaining the at least one item of the POI information matching the webpage text from the POI information repository, the at least one processor is further configured to:

determine a content category of the webpage text;

in response to the content category being determined to be a second content category, determine at least one word tagged with a pre-determined tag from the webpage text based on an entity recognition model; and

obtain the POI information associated with the at least one word from the POI information repository to be the at least one item of the POI information.

19 . The non-volatile computer-readable storage medium according to claim 18 , wherein when obtaining the at least one item of the POI information matching the webpage text from the POI information repository, the at least one processor is further configured to:

in response to the content category being determined to be a first content category, obtain a set of address information from the webpage text based on an address matching rule;

obtain items of the POI information matching the set of address information from the POI information repository, the items of the POI information including a plurality of POI identifiers;

determine at least one POI identifier that matches the webpage text from the plurality of POI identifiers based on a natural language processing model; and

obtain the POI information associated with the at least one POI identifier from the items of the POI information to be the at least one item of the POI information.

20 . The non-volatile computer-readable storage medium according to claim 19 , wherein when determining the content category of the webpage text, the at least one processor is further configured to:

determine a first frequency of a first term associated with the first content category and a second frequency of a second term associated with the second content category in the webpage text; and

in response to determination of the first frequency being greater than the second frequency, determine the content category to be the first content category; or

in response to determination of the first frequency being smaller than the second frequency, determine the content category to be the second content category; or

in response to determination of the first frequency being equal to the second frequency, obtain a set of words matching a word database from the webpage text and inputting the set of words into a text classification model to determine whether the content category is the first content category or the second content category.

Assignments (1)

Priority Claims (1)

Continuity (1)

References Cited (18)