Method and electronic device for providing commodity object search information

1 . A method for providing commodity object search information, the method comprising:

determining, according to a commodity object search request submitted by a user, a plurality of commodity objects meeting a search intent;

analyzing the plurality of commodity objects to obtain a plurality of commodity object combinations;

generating text contents corresponding to the plurality of commodity object combinations under a plurality of categories, a respective text content showing recommendation reason information corresponding to a respective commodity object combination, the recommendation reason information including a recommendation reason to display a respective commodity under a respective category; and

providing, to a client, the plurality of commodity object combinations and the text contents under the plurality of categories to be displayed in a commodity search result page of the client.

2 . The method according to claim 1 , wherein:

the commodity object search request comprises initial search condition information input by the user; and

the initial search condition information comprises a long statement expressed in a natural language.

3 . The method according to claim 2 , wherein the determining, according to the commodity object search request submitted by the user, the plurality of commodity objects meeting the search intent comprises:

performing model understanding on the long statement expressed in the natural language by using an artificial intelligence (AI) large-scale parametric model to acquire search intent information of the user; and

determining the plurality of commodity objects meeting the search intent based on the search intent information.

4 . The method according to claim 3 , further comprising:

before receiving the commodity object search request, generating a prompt text content for guiding a search initiated by inputting the long statement expressed in the natural language.

5 . The method according to claim 4 , further comprising providing the prompt text content to the client to display the prompt text content in an interface used for receiving a search input condition.

6 . The method according to claim 3 , further comprising:

using an AI large-scale parametric model to perform model understanding on user behavior information generated by the user while browsing the commodity search result page or related commodity object information;

generating a question text content for questions to be raised to the user;

providing the question text content to the client to display the question text content on the commodity search result page;

after receiving a response content that the user submits based on the question text content, performing model understanding on a question-and-response content and the initial search condition information through the AI large-scale parametric model; and

updating the search intent information of the user, and providing a new commodity object search result according to updated search intent information.

7 . The method according to claim 6 , wherein:

a process of generating a dialogue text content and receiving the response content from a user has multiple rounds, so that the AI large-scale parametric model gradually acquires and updates the search intent information of the user by performing model understanding on a dialogue content accumulated in the multiple rounds of question-and-response processes and the initial search condition information.

8 . A method for providing commodity object search information, the method comprising:

receiving a commodity object search request initiated by a user inputting initial search condition information on a target page;

submitting the commodity object search request to a server to acquire commodity object search results;

generating, according to the commodity object search results returned by the server, a search result page; and

displaying the commodity object search results, the commodity object search results including a plurality of commodity object combinations and generated text contents, a respective text content showing recommendation reason information corresponding to a respective commodity object combination.

9 . The method according to claim 8 , further comprising:

receiving a question text content provided by the server; and

displaying the question text content on the search result page.

10 . The method according to claim 9 , wherein the question text content comprises: text content used for questions to be raised to the user and generated after an artificial intelligence (AI) large-scale parametric model performs model understanding on user behavior information generated by the user when browsing the search result page or related commodity object information.

11 . The method according to claim 10 , further comprising:

after receiving a response content from the user with respect to the question text content, submitting the response content to the server, so that the server performs model understanding on a question-and-response content and the initial search condition information using an AI large-scale parametric model and provides a new commodity object search result after updating search intent information of the user; and

according to a new commodity object search result returned by the server, refreshing and displaying the search result page.

12 . The method according to claim 11 , wherein:

a process of generating dialogue text content and receiving response content from a user has multiple rounds, so that model understanding is performed on a dialogue content accumulated in the multiple rounds of question-and-response processes and the initial search condition information, and the search intent information of the user is gradually acquired and updated.

13 . The method according to claim 12 , further comprising:

providing a dialogue content display area in the search result page, wherein the dialogue content display area, by default, is configured to display a latest round of dialogue content; and

after receiving a request for backtracking a historical dialogue content, displaying the historical dialogue content in the dialogue content display area.

14 . An apparatus comprising:

one or more processors; and

one or more memories storing thereon computer-readable instructions that, when executed by the one or more processors, cause the one or more processors to perform acts comprising:

determining, according to a commodity object search request submitted by a user, a plurality of commodity objects meeting a search intent;

analyzing the plurality of commodity objects to obtain a plurality of commodity object combinations;

generating text contents corresponding to the plurality of commodity object combinations, a respective text content showing recommendation reason information corresponding to a respective commodity object combination; and

providing, to a client, the plurality of commodity object combinations and the text contents to be displayed in a commodity search result page of the client.

15 . The apparatus according to claim 14 , wherein:

the commodity object search request comprises initial search condition information input by the user; and

the initial search condition information comprises a long statement expressed in a natural language.

16 . The apparatus according to claim 15 , wherein the determining, according to the commodity object search request submitted by the user, the plurality of commodity objects meeting the search intent comprises:

performing model understanding on the long statement expressed in the natural language by using an artificial intelligence (AI) large-scale parametric model to acquire search intent information of the user; and

determining the plurality of commodity objects meeting the search intent based on the search intent information.

17 . The apparatus according to claim 16 , wherein the acts further comprise:

using an AI large-scale parametric model to perform model understanding on user behavior information generated by the user while browsing the search result page or related commodity object information;

generating a question text content for questions to be raised to the user;

providing the question text content to the client to display the question text content on the commodity search result page;

after receiving a response content that the user submits based on the question text content, performing model understanding on a question-and-response content and initial search condition information through the AI large-scale parametric model; and

updating the search intent information of the user, and providing a new commodity object search result according to updated search intent information.

18 . The apparatus according to claim 17 , wherein:

a process of generating a dialogue text content and receiving the response content from a user has multiple rounds, so that the AI large-scale parametric model gradually acquires and updates the search intent information of the user by performing model understanding on a dialogue content accumulated in the multiple rounds of question-and-response processes and the initial search condition information.

19 . The apparatus according to claim 15 , wherein the acts further comprise:

before receiving the commodity object search request, generating a prompt text content for guiding a search initiated by inputting the long statement expressed in the natural language.

20 . The apparatus according to claim 19 , wherein the acts further comprise:

providing the prompt text content to the client to display the prompt text content in an interface used for receiving a search input condition.