Search result generation method and apparatus, and device and storage medium

Abstract

Disclosed in the present invention are a search result generation method and apparatus, and a device and a storage medium, which are used for improving the accuracy of computing summaries. In the present application, the method comprises: obtaining at least one webpage corresponding to a search term; generating first text corresponding to the search term, wherein the first text is text associated with the search term; for each webpage, extracting a summary of the webpage from a main body of the webpage on the basis of the search term and the first text; and generating search results for the search term on the basis of the summary of the webpage. In the present application, first text is obtained on the basis of a search term, such that the search term is extended, and thus the search term is more complete; and a summary is determined jointly on the basis of the search term and the first text, thereby solving the problem of the effect of summarization being poor when the search term is incomplete or has a question-and-answer intent. In the present application, when search results are displayed, keywords corresponding to the search term in the summary and keywords corresponding to the first text can be displayed in a highlighted manner, thereby facilitating a user in intuitively understanding the search results.

Description

Search result generating method, device, equipment and storage medium

[0001] This application claims priority to the Chinese patent application filed with the State Intellectual Property Office on February 6, 2024, with application number 202410173786.0 and application name “A method, device, equipment and storage medium for generating search results”, the entire contents of which are incorporated by reference into this application. Technical Field

[0002] The present invention relates to the field of Internet technology, and in particular to a search result generating method, apparatus, device and storage medium. Background Art

[0003] In today's internet age, search engines are one of the most important and commonly used tools for people to access targeted information in a vast ocean of information. After a user enters a search term into a search engine, it displays multiple results. Each result includes a title, URL, and summary. The summary allows users to quickly understand the relevant snippets of the result's webpage content related to the search term, helping them determine whether the result is relevant to the search term, useful to them, and whether they should click on it.

[0004] Every web page contains a large amount of text, and the summary is derived from this text. Current summary calculation methods primarily rely on the correlation between the search term's keywords and the keywords in the web page's text. Ultimately, the most relevant sentences from the web page's content are selected as the summary. This method yields suboptimal summaries when users enter complex search terms, resulting in search results that don't meet user intent.

[0005] Summary of the Invention

[0006] The purpose of the present invention is to provide a search result generating method, apparatus, device and storage medium to solve the technical problems in the prior art.

[0007] In a first aspect, an embodiment of the present application provides a method for generating search results, the method comprising:

[0008] Obtain at least one web page corresponding to a search term; generate a first text corresponding to the search term, where the first text is text associated with the search term; for each web page, extract a summary of the web page from the body of the web page based on the search term and the first text; and generate search results for the search term based on the summary of the web page.

[0009] In this application, by obtaining the first text based on the search term, the search term is extended, making the search term more complete, and the summary is jointly determined based on the search term and the first text, solving the problem of poor summary effect when the search term is incomplete or the search term has a question-and-answer intention.

[0010] In some possible embodiments, extracting a summary of a web page from the main text of the web page based on a search term and a first text includes: performing word segmentation processing on the search term and the first text respectively to obtain a text keyword set; the text keyword set includes: at least one keyword and a weight corresponding to each keyword; performing sentence segmentation processing on the main text of the web page to obtain at least one main text sentence; and determining a summary of the web page based on the relevance of each main text sentence to a keyword in the text keyword set.

[0011] In the present application, when calculating the relevance between each body sentence of a web page and the search term, the first text is added to make the search term more complete, thereby ensuring that the calculated summary is more consistent with the user's intention.

[0012] In some possible embodiments, the summary of a web page is determined based on the relevance between each body sentence and keywords in a text keyword set, including: performing word segmentation processing on each body sentence to obtain a body keyword set corresponding to each body sentence; the body keyword set includes: at least one keyword and a weight corresponding to each keyword; based on the text keyword set and the body keyword set corresponding to each body sentence, obtaining a relevance score between each body sentence and a search term; and obtaining a summary corresponding to the web page based on the relevance score corresponding to each body sentence.

[0013] In the present application, when calculating the relevance between each body sentence of a web page and the search term, the first text is added to make the search term more complete, thereby ensuring that the calculated summary is more consistent with the user's intention.

[0014] In some possible embodiments, the search term and the first text are segmented separately to obtain a text keyword set, including: segmenting the search term to obtain at least one first keyword and a weight corresponding to each first keyword; segmenting the first text to obtain at least one second keyword and a weight corresponding to each second keyword; and obtaining a text keyword set based on the first keyword, the weight corresponding to each first keyword, the second keyword, and the weight corresponding to the second keyword.

[0015] In the present application, a text keyword set is constructed based on the search term and the first text, thereby ensuring the integrity of the keywords in the obtained text keyword set and further ensuring the accuracy of the summary determined subsequently.

[0016] In some possible embodiments, before determining the summary of the webpage based on the correlation between each body sentence and the keywords in the text keyword set, the method further includes: performing deduplication processing on the text keyword set to obtain a deduplicated text keyword set.

[0017] In this application, computing resources are saved by performing deduplication processing on a text keyword set.

[0018] In some possible embodiments, before determining the summary of the webpage according to the relevance between each body sentence and the keywords in the text keyword set, the method further includes: normalizing the weights of the keywords in the text keyword set.

[0019] In some possible embodiments, the weights of keywords in a text keyword set are normalized, including: normalizing the third keyword in the text keyword set based on a preset weight ratio corresponding to the first keyword to obtain a normalized weight corresponding to each third keyword in the text keyword set; the third keyword is a keyword corresponding to a search term in the text keyword set; normalizing the fourth keyword included in the text keyword set based on a preset weight ratio corresponding to the second keyword to obtain a normalized weight corresponding to each fourth keyword in the text keyword set; the fourth keyword is a keyword in the text keyword set other than the third keyword; using the normalized weight corresponding to each third keyword to update the weight corresponding to each third keyword in the text keyword set; using the normalized weight corresponding to each fourth keyword to update the weight corresponding to each fourth keyword in the text keyword set.

[0020] In this application, the accuracy is further improved by normalizing the weights of keywords.

[0021] In some possible embodiments, normalizing the third keyword in the text keyword set based on a preset weight ratio corresponding to the first keyword includes: normalizing the third keyword in the text keyword set based on the preset weight ratio corresponding to the first keyword and a first normalization formula; wherein the first normalization formula includes:

[0022] Among them, qw′ i is the normalized weight of the third keyword in the text keyword set, α is the preset weight ratio corresponding to the first keyword, qw i is the weight corresponding to the i-th third keyword in the text keyword set, and m is the number of third keywords in the text keyword set.

[0023] In some possible embodiments, normalizing the fourth keyword included in the text keyword set based on the preset weight ratio corresponding to the second keyword includes: normalizing the fourth keyword in the text keyword set based on the preset weight ratio corresponding to the second keyword and a second normalization formula; wherein the second normalization formula includes:

[0024] Among them, tw′ i is the normalized weight of the fourth keyword in the text keyword set, 1-α is the preset weight ratio corresponding to the second keyword, tw i is the weight corresponding to the i-th fourth keyword in the text keyword set, and n is the number of fourth keywords in the text keyword set.

[0025] In some possible embodiments, based on the text keyword set and the text keyword set corresponding to each text sentence, a relevance score between each text sentence and the search term is obtained, including: executing, for the text keyword set corresponding to each text sentence: determining the number of identical keywords in the text keyword set and the text keyword set; and obtaining a relevance score between the text sentence corresponding to the text keyword set and the search term based on the number of identical keywords.

[0026] In some possible embodiments, obtaining a relevance score between a text sentence corresponding to a text keyword set and a search term based on the number of identical keywords includes obtaining a relevance score between a text sentence corresponding to a text keyword set and a search term based on the number of identical keywords and a relevance formula; the relevance formula includes: Among them: revelance s is the relevance score between the text sentence and the search term corresponding to the text keyword set, l is the number of identical keywords in the text keyword set and the text keyword set, m i is the weight of the i-th keyword in the text keyword set, s i is the weight of the ith keyword in the text keyword set.

[0027] In some possible embodiments, before generating search results for a search term based on the abstract of a web page, the method further includes: determining a first keyword and a second keyword in the abstract of the web page based on a text keyword set; the first keyword is a keyword obtained by segmenting the search term, and the second keyword is a keyword obtained by segmenting the first text; and sending the first keyword and the second keyword in the abstract of the web page to a terminal device.

[0028] In some possible embodiments, the method further includes: determining a target keyword in response to a selection operation on a search result; updating the search term based on the target keyword to obtain an updated search term; and returning to execute the step of obtaining at least one web page corresponding to the search term based on the updated search term.

[0029] In this application, the user can interact with the keywords in the abstract. After determining that the user has selected the search result, the target keyword is determined and a secondary search is performed again, further ensuring the accuracy of the search.

[0030] In some possible embodiments, the method further includes: obtaining a modified first text in response to a modification operation on the first text; and returning to the step of extracting a summary of the web page from the body of the web page based on the search term and the first text based on the modified text.

[0031] In the present application, the user can interact with the first text, and a secondary search can be performed based on the user's modifications to the first text, thereby ensuring the accuracy of the search.

[0032] In some possible embodiments, generating a first text corresponding to a search term includes: inputting the search term into a generative artificial intelligence (AIGC) system, and obtaining the first text output by the AIGC system.

[0033] In this application, the AIGC system is used to determine the first text corresponding to the search term, and the first text is used to supplement the search term, thereby ensuring the accuracy of the obtained abstract.

[0034] In a second aspect, an embodiment of the present application further provides a display method, the method comprising:

[0035] A first interface is displayed, wherein the first interface includes a first control; a search term input by a user in the first control is obtained; search results are obtained based on the search term, wherein the search results include: a summary of at least one web page; the summary of each web page is extracted from the main body of the web page based on the search term and a first text, wherein the first text is generated based on the search term, and the first text is text associated with the search term; and the search results are displayed in a second interface.

[0036] In this application, by obtaining the first text based on the search term, the search term is extended, making the search term more complete, and the summary is jointly determined based on the search term and the first text, so that the displayed summary is more in line with the user's intention, solving the problem of poor summary effect when the search term is incomplete or the search term has a question-and-answer intention.

[0037] In some possible embodiments, the method further includes: displaying a second control in the second interface; and detecting a selection operation on the second control, displaying the first text in a third interface.

[0038] In this application, by setting the second control in the second interface, the user can intuitively see the first text, thereby improving the user experience.

[0039] In some possible embodiments, after displaying the first text in the third interface, the method further includes: obtaining a modified first text in response to a modification operation on the first text; obtaining updated search results based on the search term and the modified first text; the updated search results include: a summary of at least one web page; the summary of each web page is extracted from the body of the web page based on the search term and the modified first text; and displaying the updated search results in the second interface.

[0040] In the present application, the user can interact with the first text, and a secondary search can be performed based on the user's modifications to the first text, thereby ensuring the accuracy of the search.

[0041] In some possible embodiments, displaying search results in the second interface includes: giving a first emphasis to the first keyword in the abstract corresponding to each web page in the second interface; and giving a second emphasis to the second keyword in the abstract corresponding to each web page; the first keyword is a keyword obtained by performing word segmentation processing on the search term, and the second keyword is a keyword obtained by performing word segmentation processing on the first text.

[0042] In this application, by displaying the first keyword and the second keyword with different emphases respectively, it is convenient for users to see the search results intuitively, thereby improving the user experience.

[0043] In some possible embodiments, the first emphasis display includes: red display, highlighted display, bottom wavy line display, and / or bold display; the second emphasis display includes: red display, highlighted display, bottom wavy line display, and / or bold display; the first emphasis display is different from the second emphasis display.

[0044] In this application, the first key display is different from the second key display, which makes it easier for users to distinguish the first keyword from the second keyword.

[0045] In some possible embodiments, the method also includes: determining a target keyword in response to a selection operation on the search results; updating the search term based on the target keyword to obtain an updated search term; obtaining an updated search result based on the updated search term; the updated search results include: a summary of at least one web page; the summary of each web page is extracted from the body of the web page based on the updated search term and a second text, the second text is generated based on the updated search term, and the second text is a text associated with the updated search term; and displaying the updated search results in a second interface.

[0046] In this application, the user can interact with the keywords in the abstract. After determining that the user has selected the search result, the target keyword is determined and a secondary search is performed again, further ensuring the accuracy of the search.

[0047] In a third aspect, an embodiment of the present application further provides a search result generating device, comprising:

[0048] A web page acquisition module, configured to obtain at least one web page corresponding to a search term;

[0049] a text determination module, configured to generate a first text corresponding to the search term, wherein the first text is a text associated with the search term;

[0050] a summary acquisition module, configured to extract a summary of each web page from the main body of the web page according to the search term and the first text;

[0051] The search result generating module is used to generate the search results of the search term according to the abstract of the web page.

[0052] In some possible embodiments, the summary acquisition module is specifically used to: perform word segmentation processing on the search term and the first text respectively to obtain a text keyword set; the text keyword set includes: at least one keyword and the weight corresponding to each keyword; perform sentence segmentation processing on the main text of the web page to obtain at least one main text sentence; determine the summary of the web page based on the relevance of each main text sentence with the keywords in the text keyword set.

[0053] In some possible embodiments, the summary acquisition module is specifically used to: perform word segmentation processing on each text sentence to obtain a text keyword set corresponding to each text sentence; the text keyword set includes: at least one keyword and a weight corresponding to each keyword; based on the text keyword set and the text keyword set corresponding to each text sentence, obtain a relevance score between each text sentence and the search term; based on the relevance score corresponding to each text sentence, obtain a summary corresponding to the web page.

[0054] In some possible embodiments, the summary acquisition module is specifically used to: perform word segmentation on the search terms to obtain at least one first keyword and the weight corresponding to each first keyword; perform word segmentation on the first text to obtain at least one second keyword and the weight corresponding to each second keyword; and obtain a text keyword set based on the first keyword, the weight corresponding to each first keyword, the second keyword, and the weight corresponding to the second keyword.

[0055] In some possible embodiments, the summary acquisition module is further configured to: perform deduplication processing on the text keyword set to obtain a deduplicated text keyword set.

[0056] In some possible embodiments, the summary acquisition module is further configured to: normalize the weights of keywords in the text keyword set.

[0057] In some possible embodiments, the summary acquisition module is also used to: normalize the weights of keywords in the text keyword set, including: normalizing the third keyword in the text keyword set based on the preset weight ratio corresponding to the first keyword to obtain the normalized weight corresponding to each third keyword in the text keyword set; the third keyword is the keyword corresponding to the search term in the text keyword set; normalizing the fourth keyword included in the text keyword set based on the preset weight ratio corresponding to the second keyword to obtain the normalized weight corresponding to each fourth keyword in the text keyword set; the fourth keyword is the keyword in the text keyword set other than the third keyword; using the normalized weight corresponding to each third keyword to update the weight corresponding to each third keyword in the text keyword set; using the normalized weight corresponding to each fourth keyword to update the weight corresponding to each fourth keyword in the text keyword set.

[0058] In some possible embodiments, the summary acquisition module is specifically configured to: perform normalization processing on the third keyword in the text keyword set based on a preset weight ratio corresponding to the first keyword and a first normalization formula; wherein the first normalization formula includes:

[0059] Among them, qw′ i is the normalized weight of the third keyword in the text keyword set, α is the preset weight ratio corresponding to the first keyword, qw i is the weight corresponding to the i-th third keyword in the text keyword set, and n is the number of third keywords in the text keyword set.

[0060] In some possible embodiments, the summary acquisition module is specifically configured to: perform normalization processing on the fourth keyword in the text keyword set based on a preset weight ratio corresponding to the second keyword and a second normalization formula; wherein the second normalization formula includes:

[0061] Among them, tw′ i is the normalized weight of the fourth keyword in the text keyword set, 1-α is the preset weight ratio corresponding to the second keyword, tw i is the weight corresponding to the i-th fourth keyword in the text keyword set, and n is the number of fourth keywords in the text keyword set.

[0062] In some possible embodiments, the summary acquisition module is specifically used to: perform, for each body keyword set corresponding to a body sentence: determine the number of identical keywords in the body keyword set and the text keyword set; and obtain a relevance score between the body sentence corresponding to the body keyword set and the search term based on the number of identical keywords.

[0063] In some possible embodiments, the summary acquisition module is specifically configured to obtain a relevance score between the text sentences corresponding to the text keyword set and the search term based on the number of identical keywords and a relevance formula; the relevance formula includes: Among them: revelance s is the relevance score between the text sentence and the search term corresponding to the text keyword set, l is the number of identical keywords in the text keyword set and the text keyword set, m i is the i-th keyword in the text keyword set, s i is the i-th keyword in the text keyword set.

[0064] In some possible embodiments, the summary acquisition module is also used to: determine the first keyword and the second keyword in the summary of the web page based on the text keyword set; the first keyword is the keyword obtained by segmenting the search term, and the second keyword is the keyword obtained by segmenting the first text; and send the first keyword and the second keyword in the summary of the web page to the terminal device.

[0065] In some possible embodiments, the web page acquisition module is also used to: determine the target keyword in response to a selection operation on the search results; update the search term based on the target keyword to obtain an updated search term; and return to execute the step of obtaining at least one web page corresponding to the search term based on the updated search term.

[0066] In some possible embodiments, the summary acquisition module is further used to: obtain a modified first text in response to a modification operation on the first text; and based on the modified text, return to the step of extracting a summary of the web page from the body of the web page based on the search term and the first text.

[0067] In some possible embodiments, the summary acquisition module is specifically used to: input a search term into a generative artificial intelligence AIGC system to obtain a first text output by the AIGC system.

[0068] In a fourth aspect, an embodiment of the present application further provides a display device, comprising:

[0069] A first interface display module, configured to display a first interface, wherein the first interface includes a first control;

[0070] A search term acquisition module, used to obtain the search term entered by the user in the first control;

[0071] A search result acquisition module, configured to obtain search results based on a search term, the search results including: a summary of at least one web page; the summary of each web page is extracted from the body of the web page based on the search term and a first text generated based on the search term, the first text being text associated with the search term;

[0072] The search result display module is used to display the search results in the second interface.

[0073] In some possible embodiments, the search result display module is further configured to: display a second control in the second interface; and display the first text in the third interface when a selection operation on the second control is detected.

[0074] In some possible embodiments, the search result display module is further used to: obtain a modified first text in response to a modification operation on the first text; obtain updated search results based on the search term and the modified first text; the updated search results include: a summary of at least one web page; the summary of each web page is extracted from the main text of the web page based on the search term and the modified first text; and display the updated search results in a second interface.

[0075] In some possible embodiments, the search result display module is specifically used to: display the first keyword in the abstract corresponding to each web page as a first priority in the second interface; and, display the second keyword in the abstract corresponding to each web page as a second priority; the first keyword is a keyword obtained by segmenting the search term, and the second keyword is a keyword obtained by segmenting the first text.

[0076] In some possible embodiments, the first emphasis display includes: red display, highlighted display, bottom wavy line display, and / or bold display; the second emphasis display includes: red display, highlighted display, bottom wavy line display, and / or bold display; the first emphasis display is different from the second emphasis display.

[0077] In some possible embodiments, the search term acquisition module is also used to: determine the target keyword in response to a selection operation on the search results; update the search term based on the target keyword to obtain an updated search term; obtain updated search results based on the updated search term; the updated search results include: a summary of at least one web page; the summary of each web page is extracted from the body of the web page based on the updated search term and a second text, the second text is generated based on the updated search term, and the second text is a text associated with the updated search term; and display the updated search results in a second interface.

[0078] In the fifth aspect, another embodiment of the present application also provides an electronic device, comprising at least one processor; and a memory communicatively connected to the at least one processor; wherein the memory stores instructions executable by the at least one processor, and the instructions are executed by the at least one processor so that the at least one processor can execute any method provided in the embodiment of the first aspect of the present application.

[0079] In the sixth aspect, another embodiment of the present application further provides a computer-readable storage medium, wherein the computer-readable storage medium stores a computer program, and the computer program is used to enable a computer to execute any method provided in the embodiment of the first aspect of the present application.

[0080] Other features and advantages of the present application will be described in the following description, and in part will become apparent from the description, or will be understood by practicing the present application. The purposes and other advantages of the present application can be realized and obtained by the structures particularly pointed out in the written description, claims, and drawings. BRIEF DESCRIPTION OF THE DRAWINGS

[0081] FIG1 is a schematic diagram of a process for extracting a summary in a related art embodiment of the present application;

[0082] FIG2 is a schematic diagram of the overall process of a search result generation method provided in an embodiment of the present application;

[0083] FIG3 is a schematic diagram of the internal structure of an AIGC system of a search result generation method provided in an embodiment of the present application;

[0084] FIG4 is a schematic diagram of a method for generating search results provided by an embodiment of the present application, in which a user enters a search term in a mobile browser search box;

[0085] FIG5 is a schematic diagram of a first text output by an AIGC system of a search result generation method provided in an embodiment of the present application;

[0086] FIG6A is a flow chart of a method for generating search results according to an embodiment of the present application for extracting a summary of a web page from the main text of the web page based on a search term and a first text;

[0087] FIG6B is a flow chart of a search result generation method according to an embodiment of the present application for determining a summary of a web page based on the relevance between each body sentence and keywords in a text keyword set;

[0088] FIG7 is a schematic diagram of a web page of a search result generating method provided by an embodiment of the present application;

[0089] FIG8 is a schematic diagram of a flow chart of obtaining a text keyword set in a search result generation method provided in an embodiment of the present application;

[0090] FIG9 is a schematic diagram of keywords and weights of a search result generation method provided in an embodiment of the present application;

[0091] FIG10 is a method for generating search results according to an embodiment of the present application for normalizing a set of text keywords after deduplication.

[0092] FIG11 is a flow chart of determining the relevance score between a text sentence and a search term in a search result generation method provided in an embodiment of the present application;

[0093] FIG12 is a schematic diagram of the overall process of determining the abstract corresponding to each web page in a search result generation method provided by an embodiment of the present application;

[0094] FIG13 is a flow chart of a method for generating a search result based on a search term and a first text, according to an embodiment of the present application;

[0095] FIG14 is a schematic diagram of a secondary search based on keywords in a search result generation method provided in an embodiment of the present application;

[0096] FIG15 is a schematic diagram of determining target keywords in a search result generation method provided in an embodiment of the present application;

[0097] FIG16 is a schematic diagram showing a first text in a method for generating search results provided by an embodiment of the present application;

[0098] FIG17 is a schematic diagram of the overall process of a display method provided in an embodiment of the present application;

[0099] FIG18 is a schematic diagram of a first control of a display method provided in an embodiment of the present application;

[0100] FIG19 is a schematic diagram of a display interface of a display method provided in an embodiment of the present application;

[0101] FIG20 is a schematic diagram of a flow chart of modifying search terms in a display method provided in an embodiment of the present application;

[0102] FIG21 is a schematic diagram of a flow chart of modifying a first text in a display method provided in an embodiment of the present application;

[0103] FIG22 is a schematic diagram of a system framework of a search result generation method provided in an embodiment of the present application;

[0104] FIG23 is an interactive diagram of a search result generation method provided in an embodiment of the present application;

[0105] FIG24 is a schematic diagram of a device for generating a search result according to an embodiment of the present application;

[0106] FIG25 is a schematic diagram of a device for a display method provided in an embodiment of the present application;

[0107] FIG26 is a schematic diagram of an electronic device for a search result generating method provided in an embodiment of the present application. DETAILED DESCRIPTION

[0108] The embodiments described below with reference to the accompanying drawings are exemplary and are only used to explain the present invention, and are not to be construed as limiting the present invention.

[0109] In order to better understand the technical solution of the present application, the embodiments of the present application are described in detail below with reference to the accompanying drawings.

[0110] It should be clear that the embodiments described are only part of the embodiments of this application, not all of the embodiments. Based on the embodiments in this application, all other embodiments obtained by ordinary technicians in this field without making creative work are within the scope of protection of this application.

[0111] The terms used in the embodiments of the present application are for the purpose of describing specific embodiments only and are not intended to limit the present application. The singular forms "a", "an", "the" and "the" used in the embodiments of the present application and the appended claims are also intended to include plural forms unless the context clearly indicates otherwise.

[0112] It should be understood that the term "and / or" as used herein simply describes a relationship between associated objects, indicating that three possible relationships exist. For example, "A and / or B" can represent: A alone, A and B together, or B alone. Furthermore, the character " / " in this document generally indicates an "or" relationship between the associated objects.

[0113] The inventors' research has revealed that in today's internet age, search engines are one of the most important and commonly used tools for people to obtain targeted information in a vast ocean of information. After a user enters a search term into a search engine, the search engine displays multiple results. Each result displays a title, URL, and summary. The summary allows users to quickly understand the relevant snippets of the result's webpage content related to the search term, helping them determine whether the result is relevant to the search term, useful to them, and whether they should click on it.

[0114] Each web page contains a large number of main text sentences, and the summary is determined from the main text sentences. The current summary calculation method is mainly based on the keywords of the search term and the keywords in the main text sentences of the web page to calculate the relevance between the main text sentences and the search term, and finally select the most relevant sentences from the web page content as the summary. For example, as shown in FIG1 , the following steps are included: (1) a query term preprocessing step uses a segmentation tool to segment the query term into several sub-query terms; (2) a text preprocessing step divides the preprocessed web page text into several sentences, uses a segmentation tool to segment each sentence and saves the distribution information of each sub-query term in the web page text; (3) a word frequency and word distribution information statistics step counts the number of sub-query terms contained in each sentence, the number of consecutive sub-query terms contained and the type of sub-query terms contained, and marks the sub-query terms that appear in the sentence and indicates which sub-query term it is; (4) a scoring step uses a scoring tool to score each sentence based on the statistical data of step (3) and calculates the score of each sentence; (5) a sorting and summary generation step sorts the sentences from high to low according to the score, and selects the top K sentences to generate a summary, where 0≦K≦3.

[0115] When users enter complex search terms, the summary calculated using this method is not ideal, resulting in search results that do not meet user intent. This primarily involves the following issues: When a user's search term is incomplete, the keywords and semantic information derived from the original search term cannot fully represent the user's intent, resulting in an unsatisfactory summary. When a user's search term involves a question or answer, the summary selected based on the keywords and semantics of the search term cannot directly answer the question.

[0116] To address the above issues, embodiments of the present application provide a search result generation method, apparatus, device, and storage medium for resolving the above issues. The inventive concept of this application can be summarized as follows: obtaining at least one webpage corresponding to a search term; generating a first text corresponding to the search term, where the first text is text associated with the search term; for each webpage, extracting a summary of the webpage from the main body of the webpage based on the search term and the first text; and generating search results for the search term based on the webpage summary.

[0117] In the present application, by obtaining the first text based on the search term, the search term is extended, making the search term more complete, and the summary is jointly determined based on the search term and the first text, which solves the problem of poor summary effect when the search term is incomplete or the search term has a question-and-answer intention. In the present application, when displaying the search results, the keywords corresponding to the search term in the summary and the keywords corresponding to the first text can be highlighted, so that users can intuitively understand the search results.

[0118] First, the application scenario of a search result generation method provided in an embodiment of the present application is described. The application scenario of the present application exists in various electronic devices with graphical user interfaces, including but not limited to mobile phones, tablets, notebooks, electronic watches and other terminal devices, and the application objects include all applications with search engines on the terminal devices.

[0119] To facilitate further understanding of a search result generation method provided in an embodiment of the present application, the following is an explanation with reference to the accompanying drawings:

[0120] FIG2 is a flow chart of a method for generating search results according to an embodiment of the present application, wherein:

[0121] In step 201: obtain at least one web page corresponding to the search term.

[0122] In this application, when obtaining web pages based on search terms, any one of the following technologies can be used: web crawler technology, retrieval and ranking technology, web page processing technology, big data processing technology, natural language processing technology, etc.; this application does not limit this.

[0123] In step 202: a first text corresponding to the search term is generated, where the first text is text associated with the search term.

[0124] In this application, considering that the selected summary may be unsatisfactory when the search term entered by the user is incomplete or has a question-and-answer intention, after obtaining the search term, this application will obtain the first text based on the search term, and determine the summary based on the first text in subsequent steps, thereby ensuring the accuracy of the obtained summary.

[0125] In some possible embodiments, the first text is text associated with a search term, and obtaining the first text corresponding to the search term based on the acquired search term includes: inputting the search term into a generative artificial intelligence AIGC system to obtain the first text output by the AIGC system.

[0126] In some possible embodiments, the number of characters in the first text is greater than the number of characters in the search term.

[0127] Among them, the internal structure of the AIGC system is shown in Figure 3. The search term is input into the AIGC system, and the output of the AIGC system is the first text. The AIGC system can be Huawei Pangu Big Model, ChatGPT, etc., and this application does not limit this.

[0128] For example, as shown in Figure 4, the search term "China's first Olympic gold medalist" entered by the user in the mobile browser search box is obtained. As shown in Figure 5, the search term is entered into the AIGC system, and the first text output by the AIGC system is "China's first Olympic gold medalist is Xu Haifeng. At the 1984 Los Angeles Olympics, Xu Haifeng won the men's XX championship, becoming the first gold medalist of the Olympics and China's first Olympic champion, breaking the zero gold medal record in China's Olympic history."

[0129] In some other possible embodiments, obtaining the first text corresponding to the search term based on the acquired search term may also be implemented as: inputting the search term into a pre-trained semantic extension model to obtain the first text output by the semantic extension model.

[0130] In other possible embodiments, obtaining the first text corresponding to the search term based on the acquired search term may also be implemented as: extending the search term by any other method (such as semantic extension) to obtain the first text.

[0131] It should be noted that this application does not limit the method for obtaining the first text. The three methods given above are only two embodiments. Other methods that can obtain the first text associated with the search term based on the search term are also applicable to this application.

[0132] In step 203: for each web page, extract a summary of the web page from the body of the web page according to the search term and the first text.

[0133] In this application, the first text is obtained by semantically extending the search term, making the search term more complete, and then when an abstract is extracted from the main text of the web page based on the first text and the search term, the obtained abstract is more complete.

[0134] In step 204: generating search results for the search term based on the abstract of the web page.

[0135] In this application, by obtaining the first text based on the search term, the semantic extension of the search term is achieved, making the search term more complete, and the summary is jointly determined based on the search term and the first text, which solves the problem of poor summary effect when the search term is incomplete or the search term has a question-and-answer intention.

[0136] In some possible embodiments, the above step 203 may be specifically implemented as the steps shown in FIG6A , wherein:

[0137] In step 601 : the search term and the first text are segmented to obtain a text keyword set; the text keyword set includes: at least one keyword and a weight corresponding to each keyword.

[0138] In this application, the search terms are segmented to obtain the first keyword and the weight corresponding to the first keyword; the first text is segmented to obtain the second keyword and the weight corresponding to the second keyword; the first keyword, the weight corresponding to the first keyword, the second keyword and the weight corresponding to the second keyword are combined to obtain a text keyword set.

[0139] The keywords in the keyword set obtained by this method are richer and more complete, avoiding the problem that the extracted summary does not meet the user's expectations due to incomplete search terms entered by the user, thereby ensuring that the subsequent summary is more in line with the user's expectations.

[0140] In step 602: the body of the web page is segmented into sentences to obtain at least one body sentence.

[0141] In this application, after obtaining the web page corresponding to the search term, the text semantic segmentation technology can be used to segment the web page into sentences, and other sentence segmentation methods can also be used. This application does not limit this. The existing word segmentation tools can be used to segment each text sentence.

[0142] For example, as shown in Figure 7, the webpage reads: "my country's first Olympic gold medalist is Xu Haifeng." At the 23rd Los Angeles Olympics, Xu Haifeng won the 60-shot slow-fire championship, becoming the first gold medalist of the Games. He was also the first gold medalist in Chinese Olympic history, breaking China's zero gold medal streak in the history of the Olympic Games.

[0143] The text is divided into sentences, and the sentences obtained include: "The winner of my country's first Olympic gold medal is Xu Haifeng", "At the 23rd Los Angeles Olympic Games, Xu Haifeng won the XX60-shot slow-fire championship, becoming the first gold medal winner of this Olympic Games. At the same time, he was also the first gold medal winner in the history of the Chinese Olympic Games, breaking the zero gold medal record since China personally participated in the Olympic Games."

[0144] In step 603: the summary of the web page is determined based on the correlation between each body sentence and the keywords in the text keyword set.

[0145] In some possible embodiments, the summary of the web page is determined based on the correlation between each body sentence and the keywords in the text keyword set, which can be specifically implemented as shown in FIG6B , wherein

[0146] In step 611 : each text sentence is segmented to obtain a text keyword set corresponding to each text sentence; the text keyword set includes: at least one keyword and a weight corresponding to each keyword.

[0147] Continuing with Figure 7 as an example, the word segmentation of "The winner of my country's first Olympic gold medal is Xu Haifeng" is: I / 0.975, country / 1.504, first / 0.147, piece / 0.258, Olympics / 0.369, gold medal / 0.123, winner / 0.456, is 0.789, Xu Haifeng / 0.159. The word segmentation of "At the 23rd Los Angeles Olympic Games, Xu Haifeng won the XX60-shot slow-fire championship and became the first gold medalist of this Olympic Games. He was also the first gold medalist in the history of the Chinese Olympic Games, breaking the breakthrough of zero gold medals since China personally participated in the Olympic Games" is as follows: in / 0.123, 23 / 0.456, session / 0.789, Los Angeles / 0.147, Olympic Games / 0.258, on / 0.369, Xu Haifeng / 0.159, won / 0.357, won / 0.432, XX / 0.324, 60 / 0.621, shot / 0.235, slow-fire / 0.532, champion / 0.325, became / 0.253, this / 0.145, Olympic Games / 0.541, first gold / 0.415, winner / 0.451, at the same time / 0.235, he / 0.356, also / 0.356, is / 0.325, China / 0.478, Olympics / 0.748, history / 0.874, on / 0.578, first / 0.758, pieces / 0.875, gold medal / 0.857, of / 0.233, winner / 0.322, broke / 0.124, won / 0.324, China / 0.745, personally / 0.126, participated in / 0.472, Olympics / 0.462, since / 0.348, zero / 0.364, gold medal / 0.621, of / 0.364, breakthrough / 0.322.

[0148] In step 612: based on the text keyword set and the text keyword set corresponding to each text sentence, a relevance score between each text sentence and the search term is obtained.

[0149] In the present application, the relevance between the main text sentence and the search term, ie, the relevance score, can be determined based on each main text keyword set and the text keyword set.

[0150] In step 613: a summary corresponding to the web page is obtained based on the relevance score corresponding to each body sentence.

[0151] In some possible embodiments, after determining the relevance score between each text sentence and the search term, the text sentences can be sorted according to the relevance score corresponding to each text sentence to obtain a text sentence sequence, and a specified number of text sentences can be selected from the text sentence sequence in order from high to low scores as summaries.

[0152] In order to ensure the display effect and the coherence of the main text sentences, in some embodiments, the main text sentence with the highest score can be selected from the main text sentence sequence as the summary; if the main text sentence is too short to completely fill the corresponding summary display area, the main text sentences adjacent to the main text sentence in the web page can be used as summaries at the same time to ensure the display effect.

[0153] In some possible embodiments, the search term and the first text are segmented to obtain a text keyword set. That is, step 601 may be specifically implemented as shown in FIG8 , wherein:

[0154] In step 801: the search term is segmented to obtain at least one first keyword and a weight corresponding to each first keyword.

[0155] In this application, in order to ensure that the obtained abstract is more consistent with the search terms, the search terms will be segmented; in this application, the search terms can be segmented based on a dictionary segmentation algorithm or a statistical machine learning algorithm, and this application does not limit this.

[0156] For example, as shown in FIG9 , the search term is “China’s first Olympic gold medalist”. The first keywords and weights obtained by segmenting the search term include: China / 0.178, first place / 0.097, Olympics / 0.169, gold medal / 0.158, winner / 0.289.

[0157] In step 802: word segmentation processing is performed on the first text to obtain at least one second keyword and a weight corresponding to each second keyword.

[0158] In this application, when performing word segmentation processing on the first text, the same word segmentation processing method as the search term can be used, or a word segmentation processing method different from the search term can be used. The word segmentation processing of the first text can be based on a dictionary segmentation algorithm or a statistical machine learning algorithm. This application does not limit this.

[0159] For example, as shown in Figure 9, the first text is "China's first Olympic gold medalist is Xu Haifeng. At the 1984 Los Angeles Olympics, Xu Haifeng won the men's XX championship, becoming the first gold medalist of the Olympics. He was also China's first Olympic champion, breaking the zero gold medal record in China's Olympic history." The first text is segmented, and the second keywords and weights obtained include: China / 0.178, first place / 0.097, Olympics / 0.169, gold medal / 0.158, winner / 0.289, Xu Haifeng / 0.126, 1984 / 0.049, year / 0.019, Los Angeles / 0.112, Olympics / 0.125, Xu Haifeng / 0.022, won / 0.112, men / 0.045, XX / 0.124, champion / 0.234, became / 0.123, That session / 0.147, Olympics / 0.258, first gold / 0.159, winner / 0.269, at the same time / 0.236, also / 0.124, China / 0.369, first / 0.425, Olympics / 0.321, champion / 0.014, achieved / 0.023, achieved / 0.023, China / 0.142, Olympics / 0.123, history / 0.015, above / 0.014, gold medal / 0.023, zero / 0.156, of / 0.233, breakthrough / 0.173.

[0160] In step 803 : a text keyword set is obtained based on the first keyword, the weight corresponding to each first keyword, the second keyword, and the weight corresponding to the second keyword.

[0161] In this application, the first keyword, the second keyword and their corresponding weights are combined to obtain a text keyword set, which expands the search terms, makes the search terms richer and more complete, and ensures that the subsequent summary is more in line with user expectations.

[0162] For example, if the search term is "China's first Olympic gold medalist", the first keywords and weights obtained by segmenting the search term include: China / 0.178, first place / 0.097, Olympics / 0.169, gold medal / 0.158, winner / 0.289; the first text is "China's first Olympic gold medalist is Xu Haifeng. At the 1984 Los Angeles Olympics, Xu Haifeng won the men's XX championship, becoming the first gold medalist of the Olympics and also China's first Olympic champion, achieving a breakthrough in the zero gold medal record in China's Olympic history." The first text is segmented and the second keywords and weights obtained include: China / 0.178, first place / 0.097, Olympics Meeting / 0.169, gold medal / 0.158, winner / 0.289, Xu Haifeng / 0.126, 1984 / 0.049, year / 0.019, Los Angeles / 0.112, Olympics / 0.125, Xu Haifeng / 0.022, won / 0.112, men's / 0.045, XX / 0.124, champion / 0.234, became / 0.123, the same / 0.147, Olympics / 0.258, first gold / 0.159, winner / 0.269, at the same time / 0.236, also / 0.124, China / 0.369, first / 0.425, Olympics / 0.321, champion / 0.014, achieved / 0.023, 0.023, China / 0.142, Olympic Games / 0.123, history / 0.015, on / 0.014, gold medal / 0.023, zero / 0.156, of / 0.233, breakthrough / 0.173; the text keyword set includes: China / 0.178, first / 0.097, Olympic Games / 0.169, gold medal / 0.158, winner / 0.289, China / 0.178, first / 0.097, Olympic Games / 0.169, gold medal / 0.158, winner / 0.289, Xu Haifeng / 0.126, 1984 / 0.049, year / 0.019, Los Angeles / 0.112, Olympic Games / 0.125, Xu Haifeng / 0.022, won / 0.112, men's / 0.045, XX / 0.124, champion / 0.234, became / 0.123, the same / 0.147, Olympics / 0.258, first gold / 0.159, winner / 0.269, at the same time / 0.236, also / 0.124, China / 0.369, first / 0.425, Olympics / 0.321, champion / 0.014, achieved / 0.023, achieved / 0.023, China / 0.142, Olympics / 0.123, history / 0.015, above / 0.014, gold medal / 0.023, zero / 0.156, / 0.233, breakthrough / 0.173.

[0163] It can be seen that there are many repeated keywords in the obtained text keyword set. Therefore, in order to reduce the amount of calculation and improve the search efficiency in this application, after constructing the text keyword set based on the steps shown in Figure 8, the following can be implemented: deduplication of the text keyword set to obtain the deduplicated text keyword set.

[0164] When deduplicating a text keyword set, only one identical keyword is retained, and the weight of the retained keyword is the average of the identical keywords.

[0165] For example, as shown in FIG9 , the text keyword set includes: China / 0.178, first place / 0.097, Olympic Games / 0.169, gold medal / 0.158, winner / 0.289, China / 0.178, first place / 0.097, Olympic Games / 0.169, gold medal / 0.158, winner / 0.289, Xu Haifeng / 0.126, 1984 / 0.049, year / 0.019, Los Angeles / 0.112, Olympic Games / 0.125, Xu Haifeng / 0.022, won / 0.112, men's / 0.045, XX / 0.124, champion / 0.234, became / 0.123, the session / 0.147, Olympics / 0.258, first gold / 0.159, winner / 0.269, at the same time / 0.236, also / 0.124, China / 0.369, first / 0.425, Olympics / 0.321, champion / 0.014, achieved / 0.023, achieved / 0.023, China Country / 0.142, Olympics / 0.123, History / 0.015, Above / 0.014, Gold Medal / 0.023, Zero / 0.156, Of / 0.233, Breakthrough,0.173; the text keyword set after deduplication includes: China / 0.217, First / 0.097, Olympics / 0.169, Gold Medal / 0.113, Winner / 0.282, Xu Haifeng / 0.074, 1984 / 0.049, Year / 0.019, Los Angeles / 0.112, won / 0.112, men's / 0.045, XX / 0.124, champion / 0.124, became / 0.123, the session / 0.147, the first gold / 0.159, at the same time / 0.236, also / 0.124, the first / 0.425, Olympics / 0.321, achieved / 0.023, achieved / 0.023, history / 0.015, above / 0.014, zero / 0.156, / 0.233, breakthrough / 0.173.

[0166] In some possible embodiments, in order to further improve the accuracy of the data, after obtaining the text keyword set, the weights of the keywords in the text keyword set may be normalized to increase the proportion of the first keyword in the text keyword set. Specifically, the steps shown in FIG. 10 may be implemented, wherein:

[0167] In step 1001: based on the preset weight ratio corresponding to the first keyword, the third keyword in the text keyword set is normalized to obtain the normalized weight corresponding to each third keyword in the text keyword set; the third keyword is the keyword corresponding to the search term in the text keyword set.

[0168] For example: the text keyword set includes: China / 0.217, first / 0.097, Olympic Games / 0.169, gold medal / 0.113, winner / 0.282, Xu Haifeng / 0.074, 1984 / 0.049, year / 0.019, Los Angeles / 0.112, won / 0.112, men / 0.045, XX / 0.124, champion / 0.124, become / 0.123, that session / 0.147, first gold / 0.159, at the same time / 0.236, also / 0.124, first / 0.425, Olympic Games / 0.321, achieved / 0.023, achieved / 0.023, History / 0.015, up / 0.014, zero / 0.156, of / 0.233, breakthrough / 0.173; the first keywords include: China, first, Olympic Games, gold medal, winner; the third keywords are: China, first, Olympic Games, gold medal, winner.

[0169] In some possible embodiments, the third keyword in the text keyword set is normalized based on the preset weight ratio corresponding to the first keyword. Specifically, the normalization process can be implemented as follows: the third keyword in the text keyword set is normalized based on the preset weight ratio corresponding to the first keyword and a first normalization formula; wherein the first normalization formula is shown in Formula 1:

[0170] Among them, qw′ i is the normalized weight of the third keyword in the text keyword set, α is the preset weight ratio corresponding to the first keyword, qw i is the weight corresponding to the i-th third keyword in the text keyword set, and m is the number of third keywords in the text keyword set.

[0171] In the present application, considering that the first keyword is obtained based on the search term input by the user and can better represent the user's intention, when setting the preset weight ratio of the first keyword, the preset weight ratio of the first keyword can be set to be greater than the preset weight ratio of the second keyword.

[0172] For example, as shown in FIG9 , the preset weight ratio corresponding to the first keyword is 0.6, and the text keyword set includes: China / 0.217, first / 0.097, Olympic Games / 0.169, gold medal / 0.113, winner / 0.282, Xu Haifeng / 0.074, 1984 / 0.049, year / 0.019, Los Angeles / 0.112, won / 0.112, men / 0.045, XX / 0.124, champion / 0.124, become / 0.123, that session / 0.147, first gold / 0.159, at the same time / 0.236, also / 0.124, first / 0.425, Olympic Games / 0 .321, achieved / 0.023, achieved / 0.023, history / 0.015, up / 0.014, zero / 0.156, of / 0.233, breakthrough / 0.173; the first keywords include: China, first, Olympic Games, gold medal, winner; the third keywords are: China, first, Olympic Games, gold medal, winner; the first third keyword in the text keyword set is "China", the second third keyword is "first", the third third keyword is "Olympic Games", the fourth third keyword is "gold medal", and the fifth third keyword is "winner"; the normalized weight of the first third keyword "China" is:

[0173] Similarly, the normalized weights of other third keywords can be obtained.

[0174] In step 1002: based on the preset weight ratio corresponding to the second keyword, the fourth keyword included in the text keyword set is normalized to obtain the normalized weight corresponding to each fourth keyword in the text keyword set; the fourth keyword is the keyword in the text keyword set other than the third keyword.

[0175] For example, the text keyword set includes: China / 0.217, first / 0.097, Olympic Games / 0.169, gold medal / 0.113, winner / 0.282, Xu Haifeng / 0.074, 1984 / 0.049, year / 0.019, Los Angeles / 0.112, won / 0.112, men / 0.045, XX / 0.124, champion / 0.124, become / 0.123, that session / 0.147, first gold / 0.159, at the same time / 0.236, also / 0.124, first / 0.425, Olympic Yun / 0.321, achieved / 0.023, achieved / 0.023, history / 0.015, above / 0.014, zero / 0.156, of / 0.233, breakthrough / 0.173; the first keyword includes: China, the first, Olympic, gold medalist; the third keyword is determined to be: China, the first, Olympic, gold medalist; the fourth keyword is: Xu Haifeng, 1984, Los Angeles, won the men's XX championship, becoming the first gold medal of the session, and also the first Olympic to achieve a breakthrough in history above zero.

[0176] In some possible embodiments, the fourth keyword included in the text keyword set is normalized based on the preset weight ratio corresponding to the second keyword. Specifically, the normalization process can be implemented as follows: the fourth keyword in the text keyword set is normalized based on the preset weight ratio corresponding to the second keyword and a second normalization formula; wherein the second normalization formula is shown in Formula 2:

[0177] Among them, qw′ i is the normalized weight of the fourth keyword in the text keyword set, 1-α is the preset weight ratio corresponding to the second keyword, tw i is the weight corresponding to the i-th fourth keyword in the text keyword set, and n is the number of fourth keywords in the text keyword set.

[0178] For example, as shown in FIG9 , the preset weight ratio corresponding to the second keyword is 0.4, and the text keyword set includes: China / 0.217, first / 0.097, Olympic Games / 0.169, gold medal / 0.113, winner / 0.282, Xu Haifeng / 0.074, 1984 / 0.049, year / 0.019, Los Angeles / 0.112, won / 0.112, men / 0.045, XX / 0.124, champion / 0.124, become / 0.123, that session / 0.147, first gold / 0.159, at the same time / 0.236, also / 0.124, first / 0.425, Olympic Games / 0.3 21, achieved / 0.023, achieved / 0.023, history / 0.015, above / 0.014, zero / 0.156, of / 0.233, breakthrough / 0.173; the fourth keywords include: Xu Haifeng, 1984, Los Angeles, won the men's XX championship, became the first gold medal of the session, and also the first Olympic champion, achieved a breakthrough in history above zero; the first fourth keyword in the text keyword set is Xu Haifeng, the second fourth keyword is 1984, the third fourth keyword is Los Angeles, ..., and the twenty-second fourth keyword is breakthrough; the normalized weight of the first fourth keyword is:

[0179] Similarly, the normalized weights of other fourth keywords can be obtained.

[0180] In step 1003 : the normalized weight corresponding to each third keyword is used to update the weight corresponding to each third keyword in the text keyword set.

[0181] In step 1004 : the normalized weight corresponding to each fourth keyword is used to update the weight corresponding to each fourth keyword in the text keyword set.

[0182] That is, in the present application, after the text keyword set is normalized, the normalized weight is used to replace the weight corresponding to each keyword in the text keyword set.

[0183] It should be noted that this application does not limit the order in which the steps in Figure 10 are executed, as long as step 1001 is executed before step 1003 and step 1002 is executed before step 1004. For example, in addition to that shown in Figure 10, the execution order in Figure 10 can also be: step 1001-step 1003-step 1002-step 1004; or, step 1002-step 1004-step 1001-step 1003, etc.

[0184] In some possible embodiments, for each text sentence corresponding to a text keyword set, a relevance score between each text sentence and the search term is obtained based on the text keyword set and the text keyword set corresponding to each text sentence. That is, step 604 may be specifically implemented as shown in FIG. 11 , wherein:

[0185] In step 1101 : determine the number of identical keywords in the main text keyword set and the text keyword set.

[0186] In step 1102 : based on the number of identical keywords, the relevance score between the text sentence corresponding to the text keyword set and the search term is obtained.

[0187] In some possible embodiments, the relevance score between the text sentences corresponding to the text keyword set and the search term is obtained based on the number of identical keywords. Specifically, the relevance score between the text sentences corresponding to the text keyword set and the search term is obtained based on the number of identical keywords and a relevance formula, where the relevance formula is shown in Formula 3:

[0188] Among them: revelance s is the relevance score between the text sentence and the search term corresponding to the text keyword set, l is the number of identical keywords in the text keyword set and the text keyword set, m i is the weight corresponding to the i-th keyword in the text keyword set, s i is the weight corresponding to the i-th keyword in the text keyword set.

[0189] For example, the text keyword set includes: China / 0.217, first / 0.097, Olympic Games / 0.169, gold medal / 0.113, winner / 0.282, Xu Haifeng / 0.074, 1984 / 0.049, year / 0.019, Los Angeles / 0.112, won / 0.112, men / 0.045, XX / 0.124, champion / 0.124, become / 0.123, that session / 0.147, first gold / 0.159, at the same time / 0.236, also / 0.124, first / 0.425, Olympic Games / 0.321, achieved / 0.023, achieved / 0.023, history / 0.015, on / 0.014, zero / 0.156, of / 0.233, breakthrough / 0.173;

[0190] The key words in the text include: in / 0.123, 23 / 0.456, session / 0.789, Los Angeles / 0.147, Olympics / 0.258, on / 0.369, Xu Haifeng / 0.159, obtained / 0.357, achieved / 0.432, XX / 0.324, 60 / 0.621, served / 0.235, slow shot / 0.532, champion / 0.325, became / 0.253, this session / 0.145, Olympics / 0.541, first gold / 0.415, winner / 0.451, at the same time / 0.235, he / 0.356, also / 0. 356, is / 0.325, China / 0.478, Olympics / 0.748, history / 0.874, on / 0.578, first / 0.758, pieces / 0.875, gold medal / 0.857, of / 0.233, winner / 0.322, break / 0.124, / 0.324, China / 0.745, personally / 0.126, participate / 0.472, Olympics / 0.462, since / 0.348, zero / 0.364, gold medal / 0.621, of / 0.364, breakthrough / 0.322; the number of identical keywords is 15. The correlation score between the main text keyword set and the text keyword set can be obtained through the above correlation formula.

[0191] In some possible embodiments, to facilitate users' intuitive understanding of keywords in the abstract, before generating search results for a search term based on the webpage abstract, the following steps may be performed: determining a first keyword and a second keyword in the webpage abstract based on a text keyword set; the first keyword being a keyword obtained by segmenting the search term, and the second keyword being a keyword obtained by segmenting the first text; and transmitting the first keyword and the second keyword in the webpage abstract to a terminal device. This method allows the terminal device to use different display methods based on different keywords.

[0192] To facilitate further understanding, the overall process of determining the summary corresponding to each web page is described below, as shown in Figure 12:

[0193] In step 1201: the search words are segmented to obtain first keywords and the weight corresponding to each first keyword.

[0194] In step 1202: word segmentation is performed on the first text to obtain second keywords and weights corresponding to each second keyword.

[0195] In step 1203 : a text keyword set is obtained based on the first keyword and the weight corresponding to each first keyword, the second keyword and the weight corresponding to each second keyword.

[0196] In step 1204: deduplication processing is performed on the text keyword set to obtain a deduplication text keyword set.

[0197] In step 1205: the text keyword set after deduplication is normalized to obtain a normalized text keyword set.

[0198] In step 1206: obtain at least one web page corresponding to the search term.

[0199] In step 1207: determine the text keyword set of each text sentence of the web page.

[0200] In step 1208: a summary corresponding to the web page is obtained based on the main text keyword set and the text keyword set.

[0201] In this application, the summary corresponding to each web page can be obtained through the steps shown in FIG12 .

[0202] For example, as shown in Figure 13, the search term is "China's first Olympic gold medalist", the first text is "China's first Olympic gold medalist is Xu Haifeng. At the 1984 Los Angeles Olympics, Xu Haifeng won the men's XX championship, becoming the first gold medalist of that Olympics. He was also China's first Olympic champion, breaking the zero gold medal record in China's Olympic history", and the webpage is "my country's first Olympic gold medalist is Xu Haifeng. At the 23rd Los Angeles Olympics, Xu Haifeng won the XX60-shot slow-fire championship, becoming the first gold medalist of this Olympics. He was also the first gold medalist in China's Olympic history, breaking the zero gold medal record since China personally participated in the Olympics". The determined abstract is "At the 23rd Los Angeles Olympics, Xu Haifeng won the XX60-shot slow-fire championship, becoming the first gold medalist of this Olympics. He was also the first gold medalist in China's Olympic history, breaking the zero gold medal record since China personally participated in the Olympics".

[0203] In some possible embodiments, the user may be dissatisfied with the displayed abstract or first text. Therefore, in order to ensure user experience and further ensure the accuracy of abstract extraction, a secondary search may be performed through the steps shown in FIG12 , wherein:

[0204] In step 1401 : in response to a selection operation on a search result, a target keyword is determined.

[0205] In step 1402: the search term is updated based on the target keyword to obtain an updated search term.

[0206] In step 1403: based on the updated search term, return to the step of obtaining at least one web page corresponding to the search term.

[0207] That is, in this application, after executing step 1403, execution starts again from step 201.

[0208] For example, as shown in FIG15 , when the user clicks on “gold medal” in the summary, the target keyword is determined to be “gold medal”, and “gold medal” is used to replace “China’s first Olympic gold medalist” for a secondary search, i.e., the steps shown in FIG2 are executed again.

[0209] In some possible embodiments, the user may feel that the first text is not accurate. Therefore, in order to ensure user experience and to further ensure the accuracy of the search, in this application, after the first text is displayed in the display interface, the following steps can be implemented to perform a secondary search based on the first text modified by the user: in response to the modification operation on the first text, a modified first text is obtained; based on the modified text, the step of extracting a summary of the web page from the main text of the web page according to the search term and the first text is returned.

[0210] For example, as shown in FIG16 , the first text is modified by the user from “China’s first Olympic gold medalist is Xu Haifeng. At the 1984 Los Angeles Olympics, Xu Haifeng won the men’s XX championship, becoming the first gold medalist of that Olympics and also China’s first Olympic champion, breaking the zero gold medal record in China’s Olympic history” to “China’s first Olympic gold medalist is Xu Haifeng. An introduction to Xu Haifeng’s life.” Based on the search term “China’s first Olympic gold medalist” and the modified text “China’s first Olympic gold medalist is Xu Haifeng. An introduction to Xu Haifeng’s life,” the user returns to the step of performing word segmentation on the search term and the first text respectively.

[0211] It should be noted that, in order to ensure search efficiency, the search result generation method proposed in the present application can be executed in the cloud. It can also be executed in the terminal device when the terminal device has high computing power, and this application does not limit this.

[0212] The present application also provides a display method, as shown in FIG17 , which is a schematic diagram of the overall flow of a display method provided by the present application, wherein:

[0213] In step 1701: a first interface is displayed, wherein the first interface includes a first control.

[0214] In this application, the user can enter a search term in the first control. The first control can be located above the first interface, in the middle of the first interface, or elsewhere on the first interface. This application does not limit the location of the first control in the first interface. For example, as shown in Figure 18, the first control is located above the first interface.

[0215] In step 1702: obtain the search term entered by the user in the first control.

[0216] For example, as shown in FIG4 , the search term entered by the user in the first control (the mobile browser search box) is “China's first Olympic gold medalist”.

[0217] In step 1703: search results are obtained based on the search terms, and the search results include: a summary of at least one web page; the summary of each web page is extracted from the body of the web page based on the search terms and a first text, the first text is generated based on the search terms, and the first text is text associated with the search terms.

[0218] In step 1704: the search results are displayed in the second interface.

[0219] In the present application, when displaying search results in the second interface, it can be specifically implemented as follows: in the second interface, the first keyword in the abstract corresponding to each web page is displayed as a first priority; and, the second keyword in the abstract corresponding to each web page is displayed as a second priority; the first keyword is obtained by segmenting the search term, and the second keyword is obtained by segmenting the first text.

[0220] It should be noted that when the execution subject of the search result generation method is the cloud, the first keyword and the second keyword are received from the cloud. When the execution subject of the search result generation method is the terminal device, the first keyword is obtained by directly segmenting the search term, and the second keyword is obtained by directly segmenting the first text.

[0221] Among them: the first key display includes: red display, highlighted display, bottom wavy line display, and bold display; the second key display includes: red display, highlighted display, bottom wavy line display, and bold display; the first key display is different from the second key display.

[0222] For example: the text keyword set includes: China, first, Olympic, gold medalist, Xu Haifeng, Los Angeles, won, men, XX, champion, became, the, first, gold, and also, a, breakthrough, in, history; among them, the first keyword includes: China, first, Olympic, gold medalist, Xu Haifeng; the second keyword includes: Los Angeles, won, men, XX, champion, became, the, first, gold, and also, a, breakthrough, in, history; and the first key display method is bold, and the second key display method is underlined at the bottom, then the search results in the second interface are shown in Figure 19.

[0223] In the following possible embodiment, in order to ensure the user's viewing experience, when the first text is displayed in the display interface, the second control can be displayed in the second interface; when the selection operation of the second control is detected, the first text is displayed in the third interface.

[0224] For example: The second control is shown in FIG19 . In response to the triggering operation on the second control, as shown in FIG16 , the first text “China’s first Olympic gold medalist is Xu Haifeng. At the 1984 Los Angeles Olympics, Xu Haifeng won the men’s XX championship, becoming the first gold medalist of the Olympics. He was also China’s first Olympic champion, breaking the zero gold medal record in China’s Olympic history” is displayed.

[0225] It should be noted that the icon in FIG19 is only an embodiment, and this application does not limit the specific style of the first text icon. This application also does not limit the display position of the first text icon.

[0226] In some possible embodiments, the user may be dissatisfied with the displayed summary or first text. Therefore, in order to ensure user experience and further ensure the accuracy of summary extraction, this application provides a display method that supports modification of the first text and search terms. When the user wants to modify the search terms, the steps shown in FIG. 20 may be specifically implemented, wherein:

[0227] In step 2001: in response to a selection operation on a search result, a target keyword is determined.

[0228] In step 2002: the search term is updated based on the target keyword to obtain an updated search term.

[0229] In step 2003: updated search results are obtained based on the updated search terms; the updated search results include: a summary of at least one web page; the summary of each web page is extracted from the body of the web page based on the updated search terms and a second text, the second text is generated based on the updated search terms, and the second text is text associated with the updated search terms.

[0230] In step 2004: the updated search results are displayed on the second interface.

[0231] For example, as shown in FIG15 , when the user clicks on “gold medal” in the summary, the target keyword is determined to be “gold medal”, “gold medal” is used to replace “China’s first Olympic gold medalist”, and the updated search results are obtained for display.

[0232] In some possible embodiments, the user may feel that the first text is not accurate. Therefore, in this application, in order to ensure user experience and to further ensure search accuracy, after the first text is displayed in the display interface, the steps shown in FIG. 21 may be implemented to perform a secondary search based on the first text modified by the user:

[0233] In step 2101: in response to the modification operation on the first text, a modified first text is obtained.

[0234] In step 2102: updated search results are obtained based on the search term and the modified first text; the updated search results include: a summary of at least one web page; the summary of each web page is extracted from the body of the web page based on the search term and the modified first text.

[0235] In step 2103: the updated search results are displayed in the second interface.

[0236] For example, as shown in FIG16 , the first text is modified by the user from “China’s first Olympic gold medalist is Xu Haifeng. At the 1984 Los Angeles Olympics, Xu Haifeng won the men’s XX championship, becoming the first gold medalist of that Olympics and also China’s first Olympic champion, breaking the zero gold medal record in China’s Olympic history” to “China’s first Olympic gold medalist is Xu Haifeng. An introduction to Xu Haifeng’s life.” Based on the search term “China’s first Olympic gold medalist” and the modified text “China’s first Olympic gold medalist is Xu Haifeng. An introduction to Xu Haifeng’s life,” the user returns to the step of performing word segmentation on the search term and the first text respectively.

[0237] The following describes a search system framework corresponding to a search result generation method provided in an embodiment of the present application. The search system framework in the present application includes: one or more cloud servers and multiple terminal devices. The cloud server includes an AIGC system, a search engine retrieval ranking module, and a summary extraction module. The terminal device includes a search engine terminal module, as shown in FIG22 :

[0238] Cloud server: A cloud service that hosts various applications and provides AIGC system, search engine retrieval ranking, and summary extraction services to terminal devices through network request service interfaces.

[0239] Terminal device 1 (mobile phone, tablet, watch, computer, etc.) with a search engine APP installed: Users use functions such as search, viewing search results, and secondary search by interacting with the terminal device.

[0240] Terminal device 2 (mobile phone, tablet, watch, computer, etc.) with a search engine APP installed: Users use functions such as search, viewing search results, and secondary search by interacting with the terminal device.

[0241] Terminal device N (mobile phone, tablet, watch, computer, etc.) with a search engine app installed: Users interact with the terminal device to use functions such as search, view search results, and perform secondary search.

[0242] AIGC system: provides AIGC services and generates corresponding first text based on the input search term;

[0243] Search engine retrieval and ranking module: retrieve and sort the relevant web pages according to the input search terms;

[0244] Abstract extraction module: calculates the abstract corresponding to each web page based on the input search term;

[0245] Search engine client module: displays search results and summary information obtained from the cloud side, and allows users to interact with keywords in the first text and / or summary.

[0246] The following describes a search result generation method provided by this application in conjunction with a search system, as shown in FIG23 :

[0247] In step 2301: the terminal device obtains the search term input by the user and sends the search term to the AIGC system in the cloud.

[0248] In step 2302: the AIGC system generates a corresponding first text according to the search term.

[0249] In step 2303: the AIGC system sends the search term and the first text to the summary extraction module.

[0250] In step 2304: the summary extraction module obtains a text keyword set based on the search term and the first text.

[0251] In step 2305: the summary extraction module obtains the web page and generates a set of text keywords corresponding to each text sentence.

[0252] In step 2306: a summary corresponding to each web page is obtained based on the text keyword set and the body keyword set.

[0253] In step 2307: the summary corresponding to each web page is sent to the terminal device.

[0254] In step 2308: the terminal device displays the summary corresponding to each web page and the first text icon.

[0255] In step 2309: in response to the modification operation on the first text, a modified first text is obtained.

[0256] In step 2310: the modified first text and the search term are sent to the summary extraction module.

[0257] In step 2311 : in response to a click operation on a keyword in the abstract, a target keyword is determined, and a search term is updated based on the target keyword to obtain an updated search term.

[0258] In step 2312: the updated search term is sent to the AIGC system.

[0259] Based on the same inventive concept, after introducing a search result generating method provided in an embodiment of the present application, as shown in FIG. 24 , a search result generating device 2400 provided in an embodiment of the present application is described below. The device includes:

[0260] Web page acquisition module 24001, used to obtain at least one web page corresponding to the search term;

[0261] A text determination module 24002 is configured to generate a first text corresponding to the search term, where the first text is text associated with the search term;

[0262] A summary acquisition module 24003 is configured to extract a summary of each web page from the main body of the web page according to the search term and the first text;

[0263] The search result generating module 24004 is used to generate search results of the search term according to the abstract of the web page.

[0264] Based on the same inventive concept, after introducing a display method provided by an embodiment of the present application, as shown in FIG25 , a display device 2500 provided by an embodiment of the present application is described below. The device includes:

[0265] A first interface display module 25001 is used to display a first interface, which includes a first control;

[0266] Search term acquisition module 25002, used to obtain the search term entered by the user in the first control;

[0267] Search result acquisition module 25003, configured to obtain search results based on the search term, the search results including: a summary of at least one web page; the summary of each web page is extracted from the body of the web page based on the search term and a first text generated based on the search term, the first text being text associated with the search term;

[0268] The search result display module 25004 is used to display the search results in the second interface.

[0269] Corresponding to the above embodiments, the present application also provides an electronic device. Figure 26 is a schematic diagram of the structure of an electronic device provided in an embodiment of the present invention. The electronic device 2600 may include: a processor 2601, a memory 2602, and a communication unit 2603. These components communicate through one or more buses. Those skilled in the art will understand that the structure of the electronic device shown in the figure does not constitute a limitation on the embodiments of the present invention. It can be a bus structure or a star structure, and can also include more or fewer components than shown in the figure, or combine certain components, or arrange the components differently.

[0270] The communication unit 2603 is configured to establish a communication channel so that the electronic device can communicate with other devices, receive user data sent by other devices, or send user data to other devices.

[0271] The processor 2601 is the control center of the electronic device. It uses various interfaces and lines to connect various parts of the entire electronic device. It runs or executes software programs and / or modules stored in the memory 2602, and calls data stored in the memory to perform various functions of the electronic device and / or process data. The processor can be composed of an integrated circuit (IC), for example, it can be composed of a single packaged IC, or it can be composed of multiple packaged ICs with the same or different functions. For example, the processor 2601 can only include a central processing unit (CPU). In an embodiment of the present invention, the CPU can be a single computing core or multiple computing cores.

[0272] The memory 2602 is used to store the execution instructions of the processor 2601. The memory 2602 can be implemented by any type of volatile or non-volatile storage device or a combination thereof, such as static random access memory (SRAM), electrically erasable programmable read-only memory (EEPROM), erasable programmable read-only memory (EPROM), programmable read-only memory (PROM), read-only memory (ROM), magnetic memory, flash memory, magnetic disk or optical disk.

[0273] When the execution instructions in the memory 2602 are executed by the processor 2601, the electronic device 2600 is enabled to execute part or all of the steps in the embodiment shown in FIG. 2 .

[0274] In a specific implementation, the present invention further provides a computer storage medium, wherein the computer storage medium may store a program that, when executed, may include some or all of the steps of each embodiment of the search result generation method provided by the present invention. The storage medium may be a magnetic disk, an optical disk, a read-only memory (ROM), or a random access memory (RAM).

[0275] Those skilled in the art will clearly understand that the techniques in the embodiments of the present invention can be implemented using software plus a necessary general-purpose hardware platform. Based on this understanding, the technical solutions in the embodiments of the present invention, or the portion that contributes to the prior art, can be embodied in the form of a software product. This computer software product can be stored in a storage medium such as ROM / RAM, a magnetic disk, or an optical disk, and includes a number of instructions for enabling a computer device (which can be a personal computer, server, or network device, etc.) to execute the methods described in various embodiments of the present invention or certain portions of the embodiments.

[0276] In this specification, reference can be made to the same or similar parts between the various embodiments. In particular, for the device embodiment and the terminal embodiment, since they are basically similar to the method embodiment, the description is relatively simple, and the relevant parts can be referred to the description in the method embodiment.

Claims

1. A search result generating method, characterized in that: The method comprises: Obtain at least one web page corresponding to the search term; generating a first text corresponding to the search term, where the first text is text associated with the search term; For each web page, extracting a summary of the web page from the body of the web page according to the search term and the first text; Generate search results for the search term based on the abstract of the web page.

2. The method according to claim 1, characterized in that Extracting the abstract of the webpage from the body of the webpage according to the search term and the first text includes: Performing word segmentation processing on the search term and the first text respectively to obtain a text keyword set; the text keyword set includes: at least one keyword and a weight corresponding to each keyword; Segmenting the body of the webpage to obtain at least one body sentence; The summary of the web page is determined according to the relevance between each of the body text sentences and the keywords in the text keyword set.

3. The method according to claim 2, characterized in that Determining the summary of the webpage according to the relevance between each of the body sentences and the keywords in the text keyword set includes: Performing word segmentation processing on each of the body sentences to obtain a body keyword set corresponding to each body sentence; the body keyword set includes: at least one keyword and a weight corresponding to each of the keywords; Based on the text keyword set and the text keyword set corresponding to each of the text sentences, obtaining a relevance score between each of the text sentences and the search term; A summary corresponding to the webpage is obtained based on the relevance score corresponding to each of the body sentences.

4. The method according to claim 2, characterized in that The segmentation processing is performed on the search term and the first text to obtain a text keyword set, including: Performing word segmentation on the search term to obtain at least one first keyword and a weight corresponding to each first keyword; Performing word segmentation on the first text to obtain at least one second keyword and a weight corresponding to each second keyword; The text keyword set is obtained based on the first keyword, the weight corresponding to each of the first keywords, the second keyword, and the weight corresponding to the second keyword.

5. The method according to claim 2, characterized in that Before determining the summary of the webpage based on the relevance between each of the body sentences and the keywords in the text keyword set, the method further includes: Deduplication processing is performed on the text keyword set to obtain a deduplicated text keyword set.

6. The method according to claim 2, characterized in that Before determining the summary of the webpage based on the relevance between each of the body sentences and the keywords in the text keyword set, the method further includes: The weights of the keywords in the text keyword set are normalized.

7. The method according to claim 6, characterized in that The normalizing of the weights of the keywords in the text keyword set includes: Normalizing the third keywords in the text keyword set based on a preset weight ratio corresponding to the first keyword to obtain a normalized weight corresponding to each third keyword in the text keyword set; the third keyword is a keyword corresponding to the search term in the text keyword set; Based on a preset weight ratio corresponding to the second keyword, normalizing the fourth keyword included in the text keyword set to obtain a normalized weight corresponding to each fourth keyword in the text keyword set; the fourth keyword is a keyword in the text keyword set other than the third keyword; Using the normalized weight corresponding to each third keyword to update the weight corresponding to each third keyword in the text keyword set; The normalized weight corresponding to each fourth keyword is used to update the weight corresponding to each fourth keyword in the text keyword set.

8. The method according to claim 7, characterized in that The normalizing of the third keyword in the text keyword set based on the preset weight ratio corresponding to the first keyword includes: Normalizing the third keyword in the text keyword set based on a preset weight ratio corresponding to the first keyword and a first normalization formula; The first normalization formula includes: Among them, qw′ i is the normalized weight of the third keyword in the text keyword set, α is the preset weight ratio corresponding to the first keyword, qw i is the weight corresponding to the i-th third keyword in the text keyword set, and m is the number of third keywords in the text keyword set.

9. The method according to claim 7, characterized in that The normalizing of the fourth keyword included in the text keyword set based on the preset weight ratio corresponding to the second keyword includes: Normalizing the fourth keyword in the text keyword set based on a preset weight ratio corresponding to the second keyword and a second normalization formula; The second normalization formula includes: Among them, tw′ i is the normalized weight of the ith fourth keyword in the text keyword set, 1-α is the preset weight ratio corresponding to the second keyword, tw i is the weight corresponding to the i-th fourth keyword in the text keyword set, and n is the number of fourth keywords in the text keyword set.

10. The method according to claim 3, characterized in that The obtaining, based on the text keyword set and the text keyword set corresponding to each of the text sentences, a relevance score between each text sentence and the search term includes: For each text keyword set corresponding to the text sentence, execute: Determining the number of identical keywords in the body keyword set and the text keyword set; A relevance score between the text sentence corresponding to the text keyword set and the search term is obtained based on the number of the same keywords.

11. The method according to claim 10, characterized in that Obtaining a relevance score between the text sentence corresponding to the text keyword set and the search term based on the number of the same keywords includes: Obtaining a relevance score between the text sentence corresponding to the text keyword set and the search term based on the number of the same keywords and a relevance formula; The correlation formula includes: Among them: revelance s is the relevance score between the text sentence corresponding to the text keyword set and the search term, l is the number of identical keywords in the text keyword set and the text keyword set, m i is the weight of the i-th keyword in the text keyword set, s i is the weight of the ith keyword in the text keyword set.

12. The method according to claim 2, characterized in that Before generating search results for the search term based on the abstract of the webpage, the method further includes: Determining a first keyword and a second keyword in the abstract of the webpage based on the text keyword set; the first keyword is a keyword obtained by segmenting the search term, and the second keyword is a keyword obtained by segmenting the first text; The first keyword and the second keyword in the abstract of the webpage are sent to a terminal device.

13. The method according to claim 1, wherein The method further comprises: In response to a selection operation on the search result, determining a target keyword; updating the search term based on the target keyword to obtain an updated search term; Based on the updated search term, return to the step of obtaining at least one web page corresponding to the search term.

14. The method according to claim 1, wherein The method further comprises: In response to the modification operation on the first text, obtaining a modified first text; Based on the modified text, the step of extracting the abstract of the web page from the body of the web page according to the search term and the first text is returned to.

15. The method according to claim 1, wherein Generating the first text corresponding to the search term includes: The search term is input into a generative artificial intelligence (AIGC) system to obtain a first text output by the AIGC system.

16. A display method, characterized in that: The method comprises: Displaying a first interface, wherein the first interface includes a first control; Obtaining a search term entered by the user in the first control; Obtaining search results based on the search term, the search results including: a summary of at least one web page; the summary of each web page is extracted from the body of the web page based on the search term and a first text, the first text being generated based on the search term and being text associated with the search term; The search results are displayed in the second interface.

17. The method according to claim 16, characterized in that The method further comprises: Displaying a second control in the second interface; A selection operation on the second control is detected, and the first text is displayed in a third interface.

18. The method according to claim 17, characterized in that After displaying the first text on the third interface, the method further includes: In response to the modification operation on the first text, obtaining a modified first text; Obtaining updated search results based on the search term and the modified first text; the updated search results include: a summary of at least one web page; the summary of each web page is extracted from the body of the web page based on the search term and the modified first text; The updated search results are displayed in the second interface.

19. The method according to claim 16, wherein The displaying of the search results on the second interface includes: In the second interface, the first keyword in the abstract corresponding to each web page is displayed with first emphasis; and the second keyword in the abstract corresponding to each web page is displayed with second emphasis; the first keyword is a keyword obtained by segmenting the search term, and the second keyword is a keyword obtained by segmenting the first text.

20. The method according to claim 19, characterized in that The first key display includes: red display, highlighted display, bottom wavy line display, and / or bold display; the second key display includes: red display, highlighted display, bottom wavy line display, and / or bold display; the first key display is different from the second key display.

21. The method according to claim 16, wherein The method further comprises: In response to a selection operation on the search result, determining a target keyword; updating the search term based on the target keyword to obtain an updated search term; Obtaining updated search results based on the updated search term; the updated search results include: a summary of at least one web page; the summary of each web page is extracted from the body of the web page based on the updated search term and a second text, the second text being generated based on the updated search term, and the second text being text associated with the updated search term; The updated search results are displayed in the second interface.

22. A search result generating device, characterized in that: The device comprises: A web page acquisition module, configured to obtain at least one web page corresponding to a search term; a text determination module, configured to generate a first text corresponding to the search term, wherein the first text is a text associated with the search term; a summary acquisition module, configured to extract a summary of each web page from the main body of the web page according to the search term and the first text; The search result generating module is used to generate the search results of the search term according to the abstract of the web page.

23. A display device, characterized in that: The device comprises: A first interface display module, configured to display a first interface, wherein the first interface includes a first control; A search term acquisition module, configured to obtain the search term input by the user in the first control; a search result acquisition module, configured to obtain search results based on the search term, the search results including: a summary of at least one web page; the summary of each web page is extracted from the body of the web page based on the search term and a first text generated based on the search term, the first text being text associated with the search term; The search result display module is used to display the search results in the second interface.

24. An electronic device, characterized in that: The electronic device comprises a memory for storing computer program instructions and a processor for executing the program instructions, wherein when the computer program instructions are executed by the processor, the electronic device is triggered to execute the method according to any one of claims 1-15 and 16-21.

25. A computer-readable storage medium, characterized in that The computer-readable storage medium includes a stored program, wherein when the program is executed, the device where the computer-readable storage medium is located is controlled to execute the method according to any one of claims 1 to 15 and 16 to 21.

Images