Search result page generation method and device, equipment and storage medium

Abstract

The invention discloses a search result page generation method and device, equipment and a storage medium, and belongs to the technical field of artificial intelligence. The method comprises the following steps: when a search instruction is received, determining a search keyword according to the search instruction; performing page search according to the search keyword to obtain a page search result, and generating response information through a preset language model according to the search keyword; and generating a search result page based on the page search result and the response information. Through the mode, not only is the traditional page search result provided when the user performs search, but also the response information generated through the preset language model is provided, and the finally generated search result page comprises the response information and the page search result, so that the accuracy of the search result is improved, the search result is closer to the demand of the user, and the user experience is improved. And the search result page better conforms to the expectation of the user.

Description

Technical Field

[0001] The present invention relates to the field of artificial intelligence technology, and in particular to a search result page generation method, device, equipment and storage medium. Background Art

[0002] Currently, when users use search engines or search pages to search, the results displayed are often based on keyword extraction and mechanical search results, and are sorted by relevance. However, due to the text style and other irregularities, the keyword extraction may not be accurate when the user searches, which ultimately results in the user's search results being completely inconsistent with their needs, increasing the user's search time cost.

[0003] The above contents are only used to assist in understanding the technical solution of the present invention and do not constitute an admission that the above contents are prior art. Summary of the invention

[0004] The main purpose of the present invention is to provide a search result page generation method, device, equipment and storage medium, aiming to solve the technical problem that the results displayed by traditional search engines in the prior art are deviated from the search results and do not meet the needs of users.

[0005] To achieve the above object, the present invention provides a method for generating a search result page, the method comprising the following steps:

[0006] When receiving a search instruction, determining a search keyword according to the search instruction;

[0007] Performing a page search according to the search keyword to obtain page search results, and generating response information according to the search keyword through a preset language model;

[0008] A search result page is generated based on the page search result and the response information.

[0009] Optionally, when receiving a search instruction, determining a search keyword according to the search instruction includes:

[0010] When receiving a search instruction, determining search text information according to the search instruction;

[0011] Extracting text word information according to the search text information;

[0012] Determine a search keyword based on the text word information.

[0013] Optionally, performing a page search according to the search keyword to obtain page search results, and generating response information according to the search keyword through a preset language model includes:

[0014] Importing the search keyword into a preset search engine to perform a page search and obtain page search results;

[0015] The search keyword is imported into a preset language model, and output response information is obtained.

[0016] Optionally, generating a search result page based on the page search result and the response information includes:

[0017] Generate a response summary answer according to the response information;

[0018] Determining a target number of screening display results based on the page search results;

[0019] A search result page is generated according to the response summary answer and the screening display result.

[0020] Optionally, generating a response summary answer according to the response information includes:

[0021] Determine a response result text according to the response information;

[0022] Importing the response result text into the preset language model to obtain output response summary information;

[0023] A response summary answer is generated according to the response summary information.

[0024] Optionally, the determining a target number of screening and displaying results based on the page search results includes:

[0025] Determine the search result ranking according to the page search results;

[0026] A screening display result is selected from the page search results according to the search ranking result, and the screening display result is a target number of page search results selected according to the search ranking result.

[0027] Optionally, generating a search result page according to the response summary answer and the screening display result includes:

[0028] Creating a blank display page, and dividing the blank display page into a priority display area and a reference display area;

[0029] Fill the response summary answer into the priority display area, and fill the screening display result into the reference display area to obtain an updated display area;

[0030] A search result page is generated according to the updated display area and the blank display page.

[0031] Optionally, after generating a search result page based on the page search result and the response information, the method further includes:

[0032] Determine search question information according to the search instruction;

[0033] Determining relevant recommended questions based on the search question information;

[0034] Generate guidance and follow-up information according to the relevant recommended questions;

[0035] The guiding question display information is filled into the search result display page.

[0036] Optionally, after filling the guided question display area into the search result display page, the method further includes:

[0037] When receiving the feedback follow-up search instruction, determining the follow-up keyword according to the follow-up search instruction;

[0038] Performing a page search according to the query keyword to obtain query search results, and generating query response information according to the query keyword through a preset language model;

[0039] Generate follow-up question display information based on the follow-up question search results and the follow-up question response information;

[0040] The follow-up question display information is updated to the search result display page.

[0041] Optionally, after generating a search result page based on the page search result and the response information, the method further includes:

[0042] Determining semantic information of each search display result in the search result page;

[0043] Determine, based on the semantic information, abnormal search results with content conflicts in each search display result;

[0044] Generate and display authenticity analysis results based on the abnormal search results.

[0045] Optionally, generating and displaying authenticity analysis results based on the abnormal search results includes:

[0046] Get the channel information corresponding to each abnormal search result;

[0047] Determining the channel credibility corresponding to each abnormal search result according to the channel information;

[0048] Generate and display authenticity analysis results of the abnormal search results based on the channel credibility.

[0049] Optionally, after generating and displaying the authenticity analysis result of the abnormal search result according to the channel credibility, the method further includes:

[0050] Determining the authenticity score of each abnormal search result according to the authenticity analysis result;

[0051] The best display result among the abnormal search results is selected according to the authenticity score, and the best display result is the abnormal search result with the highest corresponding authenticity score.

[0052] In addition, to achieve the above-mentioned purpose, the present invention also provides a search result page generation device, the search result page generation device comprising:

[0053] A keyword determination module, configured to determine a search keyword according to the search instruction when a search instruction is received;

[0054] A keyword search module, used to perform page search according to the search keyword, obtain page search results, and generate response information according to the search keyword through a preset language model;

[0055] The page generation module is used to generate a search result page based on the page search result and response information.

[0056] Optionally, the page generation module is further used to determine search question information according to the search instruction; determine related recommended questions based on the search question information; generate guided question display information according to the related recommended questions; and fill the guided question display information into the search result display page.

[0057] Optionally, the page generation module is also used to, when receiving a feedback follow-up search instruction, determine a follow-up keyword according to the follow-up search instruction; perform a page search according to the follow-up keyword to obtain a follow-up search result, and generate follow-up response information according to the follow-up keyword through a preset language model; generate follow-up display information based on the follow-up search result and the follow-up response information; and update the follow-up display information to the search result display page.

[0058] Optionally, the page generation module is also used to determine the semantic information of each search display result in the search result page; determine abnormal search results with content conflicts in each search display result based on the semantic information; and generate and display authenticity analysis results based on the abnormal search results.

[0059] Optionally, the page generation module is further used to obtain channel information corresponding to each abnormal search result; determine the channel credibility corresponding to each abnormal search result based on the channel information; and generate and display the authenticity analysis results of the abnormal search results based on the channel credibility.

[0060] Optionally, the page generation module is also used to determine the authenticity score of each abnormal search result based on the authenticity analysis result; select the best display result among the abnormal search results based on the authenticity score, and the best display result is the abnormal search result with the highest corresponding authenticity score.

[0061] In addition, to achieve the above-mentioned purpose, the present invention also proposes a search result page generation device, which includes: a memory, a processor, and a search result page generation program stored in the memory and running on the processor, and the search result page generation program is configured to implement the search result page generation method described above.

[0062] In addition, to achieve the above-mentioned purpose, the present invention further proposes a storage medium, on which a search result page generation program is stored, and when the search result page generation program is executed by a processor, the search result page generation method as described above is implemented.

[0063] When receiving a search instruction, the present invention determines a search keyword according to the search instruction; performs a page search according to the search keyword to obtain a page search result, and generates response information according to the search keyword through a preset language model; and generates a search result page based on the page search result and the response information. In this way, when a user performs a search, not only the traditional page search result is provided, but also the response information generated through a preset language model is provided. The search result page generated finally includes two results, the response information and the page search result, which improves the accuracy of the search result, is closer to the needs of the user, and makes the search result page more in line with the user's expectations. BRIEF DESCRIPTION OF THE DRAWINGS

[0064] Figure 1 It is a structural diagram of a search result page generating device in a hardware operating environment involved in an embodiment of the present invention;

[0065] Figure 2 This is a flow chart of a first embodiment of a method for generating a search result page according to the present invention;

[0066] Figure 3 This is a flow chart of a second embodiment of a method for generating a search result page according to the present invention;

[0067] Figure 4 A schematic diagram of a search result page in an embodiment of a method for generating a search result page of the present invention;

[0068] Figure 5 A schematic diagram of a flow chart of a third embodiment of a method for generating a search result page according to the present invention;

[0069] Figure 6This is a structural block diagram of the first embodiment of the search result page generating device of the present invention.

[0070] The realization of the purpose, functional features and advantages of the present invention will be further explained in conjunction with embodiments and with reference to the accompanying drawings. DETAILED DESCRIPTION

[0071] It should be understood that the specific embodiments described herein are only used to explain the present invention, and are not used to limit the present invention.

[0072] Reference Figure 1 , Figure 1 A schematic diagram of a device structure for generating a search result page for a hardware operating environment according to an embodiment of the present invention.

[0073] like Figure 1 As shown, the search result page generating device may include: a processor 1001, such as a central processing unit (CPU), a communication bus 1002, a user interface 1003, a network interface 1004, and a memory 1005. Among them, the communication bus 1002 is used to realize the connection and communication between these components. The user interface 1003 may include a display screen (Display), an input unit such as a keyboard (Keyboard), and the optional user interface 1003 may also include a standard wired interface and a wireless interface. The network interface 1004 may optionally include a standard wired interface and a wireless interface (such as a wireless fidelity (Wireless-Fidelity, Wi-Fi) interface). The memory 1005 may be a high-speed random access memory (Random Access Memory, RAM) memory, or a stable non-volatile memory (Non-Volatile Memory, NVM), such as a disk memory. The memory 1005 may also be a storage device independent of the aforementioned processor 1001.

[0074] Those skilled in the art will understand that Figure 1 The structure shown in the figure does not constitute a limitation on the search result page generating device, and may include more or less components than those shown in the figure, or combine certain components, or arrange the components differently.

[0075] like Figure 1 As shown, the memory 1005 as a storage medium may include an operating system, a network communication module, a user interface module, and a search result page generation program.

[0076] exist Figure 1In the search result page generating device shown, the network interface 1004 is mainly used for data communication with the network server; the user interface 1003 is mainly used for data interaction with the user; the processor 1001 and the memory 1005 in the search result page generating device of the present invention can be set in the search result page generating device, and the search result page generating device calls the search result page generating program stored in the memory 1005 through the processor 1001, and executes the search result page generating method provided by the embodiment of the present invention.

[0077] The embodiment of the present invention provides a method for generating a search result page, referring to Figure 2 , Figure 2 The figure is a flow chart of a first embodiment of a method for generating a search result page according to the present invention.

[0078] In this embodiment, the search result page generation method includes the following steps:

[0079] Step S10: upon receiving a search instruction, determining a search keyword according to the search instruction.

[0080] In this embodiment, the execution subject of this embodiment can be the search result page generation device, which has functions such as data processing, data communication and program running, and the search result page generation device can be a server, an intelligent terminal or a computer. Of course, it can also be other devices with similar functions, and this implementation condition does not limit this. For the convenience of explanation, this implementation is described by taking the search result page generation device as an example.

[0081] It should be noted that in order to improve information processing capabilities, in some embodiments, an intelligent assistant can be installed on the terminal device. The intelligent assistant can be a program built based on the natural language processing technology of the Generative Pre-trained Transformer (GPT) model. The intelligent assistant can receive the user's input text, parse the input text into a computer-understandable data format, and then use the pre-trained neural network model to generate response text.

[0082] In a specific implementation, the intelligent assistant utilizes a neural network in deep learning technology and is trained with a large amount of corpus data, enabling it to learn the structure, grammar rules, and semantic information of language. After the user inputs text, the intelligent assistant first performs word segmentation and parsing on the text, converts it into a computer-readable vector form, and then sends it into a pre-trained neural network model for inference. During the inference process, the intelligent assistant predicts the most likely next response based on the text input by the user and the previous context. This prediction process is based on the model's learning from a large amount of corpus data, so it can largely restore the way of human natural language expression and generate highly coherent and natural response text. Finally, the intelligent assistant returns the generated response text to the user to complete a conversation interaction. Throughout the process, the intelligent assistant will continuously learn and optimize to provide more accurate and user-appropriate answers.

[0083] It should be understood that the search instruction refers to an instruction for searching content triggered by the user through a keyboard, touch screen, voice, or other forms, and can be in any form. This embodiment does not limit it.

[0084] In a specific implementation, after receiving the search instruction, the search information in the search instruction is extracted to obtain search keywords.

[0085] Further, in order to accurately extract search keywords, step S10 includes: when receiving the search instruction, determining search text information according to the search instruction; extracting text word information according to the search text information; and determining search keywords according to the text word information.

[0086] It should be noted that after receiving the search instruction, first determine the search text information according to the search instruction, that is, the search content when the user conducts a search is converted into text information.

[0087] It should be understood that after obtaining the search text information, the search text is then split into multiple words, and the multiple words are aggregated to obtain text word information.

[0088] In a specific implementation, after obtaining the text word information, all the words in the text word information are screened, and words without practical meaning are removed, such as modal particles and auxiliary words like "ne" and "le", and finally the search keywords are obtained through screening.

[0089] It should be noted that the search keywords are the fragments of words and sentences obtained by combining the user's search text information with the neural network model in the intelligent assistant. The search text information can be standard, concise proper nouns or words, or more colloquial, fragmented information, or records of user conversations and questions and answers. Since the trained neural network model of the voice assistant can better process and understand sentences and semantics, compared with the existing processing methods, it can split and extract more accurate search keywords from the user's spoken language, fragmented conversation records and other sentences, thereby obtaining more optimized and high-quality answers.

[0090] In this way, it is possible to accurately obtain search keywords based on search instructions, making the search more in line with user needs.

[0091] Step S20: performing a page search according to the search keyword, obtaining page search results, and generating response information according to the search keyword through a preset language model.

[0092] It should be noted that after the search keyword is obtained, the search keyword is used to perform a page search. The page search is to search through traditional search methods to obtain page search results.

[0093] It should be understood that while performing the page search, the search keywords are imported into the preset language model, and the response information automatically output by the preset language model is obtained.

[0094] Furthermore, in order to accurately remember the two types of searches and responses, step S20 includes: importing the search keywords into a preset search engine to perform a page search and obtain page search results; importing the search keywords into a preset language model and obtaining output response information.

[0095] In a specific implementation, after obtaining the search keyword, the means of performing a page search using the search keyword is to import the search keyword into a preset search engine, thereby obtaining page search results, wherein the preset search engine can be any desktop search engine, or any publicly used search engine on the market, and this embodiment does not impose any restrictions on this.

[0096] It should be noted that by importing the search keywords into the preset language model, the response information output by the preset language model can be directly obtained, wherein the response information includes the search and optimization result information for the search keywords output by the preset language model in the intelligent assistant.

[0097] It should be understood that the preset language model refers to the built-in deep learning neural network model in the smart assistant, and the preset language model can also combine contextual context and semantic information when searching for keywords.

[0098] In this way, it is possible to search and analyze search keywords through traditional page search and intelligent analysis of preset language models to obtain search results that better meet user needs.

[0099] Step S30: Generate a search result page based on the page search result and response information.

[0100] In the specific implementation, after obtaining the page search results and response information, a summary answer is generated based on the response information for priority display, and then the page search results are filtered and displayed at the same time. The final search result page takes into account both the results of traditional searches and the analysis results of the intelligent assistant, so that the display of search results can more accurately meet user needs.

[0101] Furthermore, in order to facilitate users to conduct multiple searches in succession, after step S30, it also includes: determining search question information according to the search instruction; determining related recommended questions based on the search question information; generating guided question display information according to the related recommended questions; and filling the guided question display information into the search result display page.

[0102] It should be noted that the search question information refers to the user's search question determined according to the search instruction, that is, information related to the search topic. Then, based on the search question information, retrieval is performed to obtain related recommended questions similar to the user's search question in the search question information. For example, when the search question is "the director of a certain movie", the related recommended questions may be "the starring actor of the movie" and "the release date of the movie".

[0103] It should be understood that after determining the relevant recommended questions, display information for guiding follow-up questions is generated. Specifically, a new area is created on the search results display page, and then the text and arrangement information corresponding to the relevant recommended questions are filled into the newly created area, thereby guiding users to ask follow-up questions on the search results page, thereby optimizing the search results.

[0104] In this way, it is achieved that the user is facilitated to further search and obtain information by guiding the user to ask further questions and recommending related questions.

[0105] Furthermore, in order to update the search result display page when the user asks a follow-up question, after the step of filling the guide follow-up question display area into the search result display page, it also includes: when receiving a feedback follow-up question search instruction, determining a follow-up question keyword according to the follow-up question search instruction; performing a page search according to the follow-up question keyword to obtain a follow-up question search result, and generating follow-up question response information through a preset language model according to the follow-up question keyword; generating follow-up question display information based on the follow-up question search result and the follow-up question response information; and updating the follow-up question display information to the search result display page.

[0106] In a specific implementation, after the guiding follow-up question display information is filled into the search result display page and displayed to the user, the follow-up question search instruction fed back by the user is received, so that the search keyword corresponding to the follow-up question search instruction supplemented by the user can be determined according to the follow-up question search instruction as the follow-up question keyword.

[0107] It should be understood that after obtaining the follow-up keyword, the follow-up keyword is brought into the page search and the preset language model, so that the follow-up search results and follow-up response information for the follow-up keyword can be obtained, and finally the follow-up display information is generated based on the follow-up search results and the follow-up response information, and finally the follow-up display information is updated to a new area of ​​the search result display page, thereby prompting the user with the updated search results.

[0108] In this way, the search result display page can be updated in real time according to the follow-up questions of users, thus improving the search efficiency of users.

[0109] This embodiment determines the search keyword according to the search instruction when receiving the search instruction; performs a page search according to the search keyword to obtain the page search results, and generates response information according to the search keyword through a preset language model; and generates a search result page based on the page search results and the response information. In this way, when the user searches, not only the traditional page search results are provided, but also the response information generated by the preset language model is provided. The search result page finally generated includes two results, the response information and the page search results, which improves the accuracy of the search results, is closer to the needs of the user, and makes the search result page more in line with the user's expectations.

[0110] refer to Figure 3 , Figure 3 It is a flowchart diagram of a second embodiment of a method for generating a search result page according to the present invention.

[0111] Based on the first embodiment described above, the search result page generation method of this embodiment includes in step S30:

[0112] Step S301: Generate a response summary answer according to the response information.

[0113] It should be noted that first, a corresponding summary answer is obtained based on the response information, that is, content summary and key points are extracted based on the search results and search optimized content in the response information, so as to obtain a response summary answer corresponding to the response information.

[0114] Furthermore, in order to generate an easy-to-read response summary answer based on the response information, step S301 includes: determining the response result text according to the response information; importing the response result text into the preset language model to obtain output response summary information; and generating a response summary answer according to the response summary information.

[0115] It should be understood that the response result text refers to text information corresponding to the search answer included in each response information extracted from the response information.

[0116] In a specific implementation, after obtaining the response result text, the response result text is imported into a preset language model, so that the response result text can be analyzed by the preset language model to obtain response summary information corresponding to the response result text split by semantics and context, and then the response summary information is matched to each answer to obtain a response summary answer.

[0117] In this way, a corresponding response summary answer is generated based on the response information, which is convenient for users to read and browse quickly.

[0118] Step S302: Determine a target number of screening display results based on the page search results.

[0119] It should be noted that the target number refers to the preset number of displayed page search results, which can be any integer and is not limited in this embodiment.

[0120] It should be understood that the filtered display results refer to the results displayed after filtering and selecting the page search results.

[0121] Furthermore, in order to accurately filter and display the page search results, step S302 includes: determining the search result ranking based on the page search results; selecting filtered display results from the page search results based on the search ranking results, and the filtered display results are the target number of page search results selected according to the search ranking results.

[0122] In a specific implementation, the search result ranking is first determined based on the page search results. After searching through a traditional search engine, there are several page search results. Then the search result ranking is first determined, that is, the order in which the search engine automatically sorts according to relevance.

[0123] It should be noted that after determining the search result ranking, the target number of results are selected from the page search results according to the search result ranking as the filtered display results. For example, when the target number is 3, the top three page search results are selected according to the search result ranking as the filtered display results.

[0124] In this way, the page search results are filtered and more relevant results are displayed to users first, making it easier for users to read.

[0125] Step S303: Generate a search result page according to the response summary answer and the screening display result.

[0126] It should be understood that if Figure 4 The following is a schematic diagram of the search results page, which includes response summary answers and filtered display results. This structure makes it easier for users to read and obtain information.

[0127] Furthermore, in order to generate a search result page that is convenient for users to obtain information, step S303 includes: creating a blank display page, and dividing the blank display page into a priority display area and a reference display area; filling the response summary answer into the priority display area, and filling the screened display result into the reference display area to obtain an updated display area; generating a search result page based on the updated display area and the blank display page.

[0128] In the specific implementation, first create a blank display page, in which operations such as editing, inserting, and filling in information can be performed.

[0129] It should be noted that after the blank display page is created, a priority display area and a reference display area are created and divided on the blank display page, so that response summary answers and screened display results can be performed in the priority display area and the reference display area at the same time.

[0130] It should be understood that since the response information fed back by the intelligent assistant is smarter and closer to user needs, the response summary answer is filled in the priority display area, and then the results searched by the traditional search engine are filled in the reference display area as an auxiliary reference to obtain an updated display area. The updated display area includes the filled-in priority display area and reference display area.

[0131] In a specific implementation, after the update display area is obtained, the update display area is updated to a blank display page, thereby obtaining a search result page containing information of the search results.

[0132] In this way, a blank display page is created, and a display area is created in it, with the response summary answers output by the smart assistant as priority display, assisted by the search results of traditional search engines as a reference, thereby improving the efficiency of users in obtaining information and reducing the difficulty.

[0133] This embodiment generates a response summary answer based on the response information; determines a target number of screening display results based on the page search results; and generates a search result page based on the response summary answer and the screening display results. In this way, further screening and selection of the response information and the page search results are achieved, so that the search result page finally displayed to the user can more easily obtain the information most relevant to the user's question.

[0134] refer to Figure 5 , Figure 5 It is a flowchart diagram of a third embodiment of a method for generating a search result page according to the present invention.

[0135] Based on the first embodiment, the search result page generation method of this embodiment further includes, after step S30:

[0136] Step S31: Determine the semantic information of each search display result in the search result page.

[0137] It should be noted that semantic information refers to the specific semantics of each search display result output by the intelligent assistant, among which the search display result refers to the search results and response answer summary displayed on the search results page.

[0138] Step S32: determining abnormal search results with content conflicts in each search display result according to the semantic information.

[0139] It should be understood that after obtaining the semantic information of each search display result, the specific content of each search display result can be judged based on the semantic information, so that abnormal search results with conflicting content and meaning can be determined. For example, when two search display results show that the release date of movie A is different, these two search display results are determined to be abnormal search results.

[0140] Step S33: Generate and display authenticity analysis results based on the abnormal search results.

[0141] In a specific implementation, after the abnormal search results are determined, the authenticity of each abnormal search result is judged based on the channel information of the abnormal search results to obtain an authenticity analysis result.

[0142] Furthermore, in order to accurately determine the authenticity, step S33 includes: obtaining channel information corresponding to each abnormal search result; determining the channel credibility corresponding to each abnormal search result based on the channel information; generating and displaying the authenticity analysis results of the abnormal search results based on the channel credibility.

[0143] It should be noted that the channel information of each abnormal search result is first obtained. The channel information refers to the source information corresponding to each abnormal search result. Then, the channel credibility corresponding to each abnormal search result is verified based on the channel information. The channel credibility refers to the officiality and timeliness of the channel source. For example, the official website of an official and authoritative organization has a high credibility. The closer the release time is to the current moment, the higher the credibility.

[0144] It should be understood that after the channel credibility is determined, the authenticity analysis results of each abnormal search result are obtained based on the channel credibility summary and output in the form of a list or report.

[0145] In this way, authenticity analysis of each abnormal search result is achieved through the source channel, which improves the credibility of answer screening.

[0146] Furthermore, in order to prevent erroneous information from interfering with users, after the step of generating and displaying the authenticity analysis results of the abnormal search results according to the channel credibility, it also includes: determining the authenticity score of each abnormal search result according to the authenticity analysis results; selecting the best display result among each abnormal search result according to the authenticity score, and the best display result is the abnormal search result with the highest corresponding authenticity score.

[0147] It should be noted that after obtaining the authenticity analysis results, you can import them into the smart assistant to get the authenticity score of each abnormal search result. The authenticity score refers to the timeliness and channel source reliability score of each abnormal search result obtained after analyzing the authenticity analysis results through the smart assistant. The more reliable the timeliness and channel source, the higher the authenticity score.

[0148] It should be understood that, finally, according to the authenticity score, the one with the highest authenticity score in each group of abnormal search results is selected as the best display result, and the best display result is displayed on the search result page.

[0149] In this way, conflicting search results can be screened and evaluated, and only the best results can be displayed, thereby reducing the possibility of users being disturbed by erroneous information.

[0150] This embodiment determines the semantic information of each search display result in the search result page; determines abnormal search results with content conflicts in each search display result according to the semantic information; and generates and displays authenticity analysis results based on the abnormal search results. In this way, it is possible to assist users in authenticity analysis of search display results, thereby preventing users from being disturbed by erroneous information.

[0151] In addition, an embodiment of the present invention further provides a storage medium, on which a search result page generation program is stored. When the search result page generation program is executed by a processor, the steps of the search result page generation method described above are implemented.

[0152] Since the storage medium adopts all the technical solutions of all the above embodiments, it has at least all the beneficial effects brought by the technical solutions of the above embodiments, which will not be described one by one here.

[0153] Reference Figure 6 , Figure 6 This is a structural block diagram of the first embodiment of the search result page generating device of the present invention.

[0154] like Figure 6 As shown, the search result page generating device proposed in the embodiment of the present invention includes:

[0155] The keyword determination module 10 is used to determine the search keyword according to the search instruction when receiving the search instruction.

[0156] The keyword search module 20 is used to perform a page search according to the search keyword, obtain page search results, and generate response information according to the search keyword through a preset language model.

[0157] The page generation module 30 is used to generate a search result page based on the page search result and response information.

[0158] This embodiment determines the search keyword according to the search instruction when receiving the search instruction; performs a page search according to the search keyword to obtain the page search results, and generates response information according to the search keyword through a preset language model; and generates a search result page based on the page search results and the response information. In this way, when the user searches, not only the traditional page search results are provided, but also the response information generated by the preset language model is provided. The search result page finally generated includes two results, the response information and the page search results, which improves the accuracy of the search results, is closer to the needs of the user, and makes the search result page more in line with the user's expectations.

[0159] In one embodiment, the keyword determination module 10 is further used to, when receiving a search instruction, determine search text information according to the search instruction; extract text word information according to the search text information; and determine search keywords according to the text word information.

[0160] In one embodiment, the keyword search module 20 is further used to import the search keyword into a preset search engine to perform page search and obtain page search results; import the search keyword into a preset language model and obtain output response information.

[0161] In one embodiment, the page generation module 30 is further used to generate a response summary answer based on the response information; determine a target number of filtered display results based on the page search results; and generate a search result page based on the response summary answer and the filtered display results.

[0162] In one embodiment, the page generation module 30 is further used to determine a response result text according to the response information; import the response result text into the preset language model to obtain output response summary information; and generate a response summary answer according to the response summary information.

[0163] In one embodiment, the page generation module 30 is further used to determine the search result ranking based on the page search results; select filtered display results from the page search results based on the search ranking results, and the filtered display results are the target number of page search results selected according to the search ranking results.

[0164] In one embodiment, the page generation module 30 is also used to create a blank display page, and divide the blank display page into a priority display area and a reference display area; fill the response summary answer into the priority display area, and fill the screening display result into the reference display area to obtain an updated display area; generate a search result page based on the updated display area and the blank display page.

[0165] In one embodiment, the page generation module 30 is further used to determine relevant recommended questions based on the search question information; generate guided question display information according to the relevant recommended questions; and fill the guided question display information into the search result display page.

[0166] In one embodiment, the page generation module 30 is further used to determine a follow-up question keyword according to the follow-up question search instruction when a feedback follow-up question search instruction is received; perform a page search according to the follow-up question keyword to obtain a follow-up question search result, and generate a follow-up question response information according to the follow-up question keyword through a preset language model; generate a follow-up question display information based on the follow-up question search result and the follow-up question response information; and update the follow-up question display information to the search result display page.

[0167] In one embodiment, the page generation module 30 is also used to determine the semantic information of each search display result in the search result page; determine abnormal search results with content conflicts in each search display result based on the semantic information; and generate and display authenticity analysis results based on the abnormal search results.

[0168] In one embodiment, the page generation module 30 is further used to obtain channel information corresponding to each abnormal search result; determine the channel credibility corresponding to each abnormal search result based on the channel information; and generate and display the authenticity analysis results of the abnormal search results based on the channel credibility.

[0169] In one embodiment, the page generation module 30 is also used to determine the authenticity score of each abnormal search result based on the authenticity analysis result; select the best display result among the abnormal search results based on the authenticity score, and the best display result is the abnormal search result with the highest authenticity score.

[0170] It should be understood that the above is only an example and does not constitute any limitation on the technical solution of the present invention. In specific applications, technicians in this field can make settings as needed, and the present invention does not limit this.

[0171] It should be noted that the workflow described above is merely illustrative and does not limit the scope of protection of the present invention. In practical applications, technicians in this field can select part or all of them according to actual needs to achieve the purpose of the present embodiment, and no limitation is made here.

[0172] In addition, for technical details not fully described in this embodiment, reference may be made to the search result page generation method provided in any embodiment of the present invention, and will not be repeated here.

[0173] In addition, it should be noted that, in this article, the terms "include", "comprises" or any other variations thereof are intended to cover non-exclusive inclusion, so that a process, method, article or system including a series of elements includes not only those elements, but also includes other elements not explicitly listed, or also includes elements inherent to such process, method, article or system. In the absence of further restrictions, an element defined by the sentence "comprises a ..." does not exclude the existence of other identical elements in the process, method, article or system including the element.

[0174] The serial numbers of the above embodiments of the present invention are only for description and do not represent the advantages or disadvantages of the embodiments.

[0175] Through the description of the above implementation methods, those skilled in the art can clearly understand that the above-mentioned embodiment methods can be implemented by means of software plus a necessary general hardware platform, and of course by hardware, but in many cases the former is a better implementation method. Based on such an understanding, the technical solution of the present invention, or the part that contributes to the prior art, can be embodied in the form of a software product, which is stored in a storage medium (such as a read-only memory (ROM) / RAM, a magnetic disk, or an optical disk), and includes a number of instructions for a terminal device (which can be a mobile phone, a computer, a server, or a network device, etc.) to execute the methods described in each embodiment of the present invention.

[0176] The above are only preferred embodiments of the present invention, and are not intended to limit the patent scope of the present invention. Any equivalent structure or equivalent process transformation made using the contents of the present invention specification and drawings, or directly or indirectly applied in other related technical fields, are also included in the patent protection scope of the present invention.

[0177] The present invention discloses A1, a method for generating a search result page, the method for generating a search result page comprising:

[0178] When receiving a search instruction, determining a search keyword according to the search instruction;

[0179] Performing a page search according to the search keyword to obtain page search results, and generating response information according to the search keyword through a preset language model;

[0180] A search result page is generated based on the page search result and the response information.

[0181] A2. The method for generating a search result page as described in A1, wherein upon receiving a search instruction, determining a search keyword according to the search instruction comprises:

[0182] When receiving a search instruction, determining search text information according to the search instruction;

[0183] Extracting text word information according to the search text information;

[0184] Determine a search keyword based on the text word information.

[0185] A3. The method for generating a search result page as described in A1, wherein the page search is performed according to the search keyword to obtain the page search result, and response information is generated according to the search keyword through a preset language model, including:

[0186] Importing the search keyword into a preset search engine to perform a page search and obtain page search results;

[0187] The search keyword is imported into a preset language model, and output response information is obtained.

[0188] A4. The method for generating a search result page as described in A1, wherein generating a search result page based on the page search result and the response information comprises:

[0189] Generate a response summary answer according to the response information;

[0190] Determining a target number of screening display results based on the page search results;

[0191] A search result page is generated according to the response summary answer and the screening display result.

[0192] A5. The method for generating a search result page as described in A4, wherein generating a response summary answer according to the response information comprises:

[0193] Determine a response result text according to the response information;

[0194] Importing the response result text into the preset language model to obtain output response summary information;

[0195] A response summary answer is generated according to the response summary information.

[0196] A6. The method for generating a search result page as described in A4, wherein the target number of screening and displaying results is determined based on the page search results, comprising:

[0197] Determine the search result ranking according to the page search results;

[0198] A screening display result is selected from the page search results according to the search ranking result, and the screening display result is a target number of page search results selected according to the search ranking result.

[0199] A7. The method for generating a search result page as described in A4, wherein generating a search result page according to the response summary answer and the screening display result comprises:

[0200] Creating a blank display page, and dividing the blank display page into a priority display area and a reference display area;

[0201] Fill the response summary answer into the priority display area, and fill the screening display result into the reference display area to obtain an updated display area;

[0202] A search result page is generated according to the updated display area and the blank display page.

[0203] A8. The method for generating a search result page as described in any one of A1 to A7, after generating the search result page based on the page search result and the response information, further comprising:

[0204] Determine search question information according to the search instruction;

[0205] Determining relevant recommended questions based on the search question information;

[0206] Generate guidance and follow-up information according to the relevant recommended questions;

[0207] The guiding question display information is filled into the search result display page.

[0208] A9. The method for generating a search result page as described in A8, after filling the guiding question display area into the search result display page, further comprising:

[0209] When receiving the feedback follow-up search instruction, determining the follow-up keyword according to the follow-up search instruction;

[0210] Performing a page search according to the query keyword to obtain query search results, and generating query response information according to the query keyword through a preset language model;

[0211] Generate follow-up question display information based on the follow-up question search results and the follow-up question response information;

[0212] The follow-up question display information is updated to the search result display page.

[0213] A10. The method as described in any one of A1 to A7, after generating a search result page based on the page search results and the response information, further comprising:

[0214] Determining semantic information of each search display result in the search result page;

[0215] Determine, based on the semantic information, abnormal search results with content conflicts in each search display result;

[0216] Generate and display authenticity analysis results based on the abnormal search results.

[0217] A11. The method as described in A10, wherein generating and displaying authenticity analysis results based on the abnormal search results comprises:

[0218] Get the channel information corresponding to each abnormal search result;

[0219] Determining the channel credibility corresponding to each abnormal search result according to the channel information;

[0220] Generate and display authenticity analysis results of the abnormal search results based on the channel credibility.

[0221] A12. The method as described in A11, after generating and displaying the authenticity analysis result of the abnormal search result according to the channel credibility, further comprising:

[0222] Determining the authenticity score of each abnormal search result according to the authenticity analysis result;

[0223] The best display result among the abnormal search results is selected according to the authenticity score, and the best display result is the abnormal search result with the highest corresponding authenticity score.

[0224] The present invention also discloses B13, a search result page generating device, the search result page generating device comprising:

[0225] A keyword determination module, configured to determine a search keyword according to the search instruction when a search instruction is received;

[0226] A keyword search module, used to perform page search according to the search keyword, obtain page search results, and generate response information according to the search keyword through a preset language model;

[0227] The page generation module is used to generate a search result page based on the page search result and response information.

[0228] B14. In the search result page generation device as described in B13, the page generation module is further used to determine search question information according to the search instruction; determine related recommended questions based on the search question information; generate guided question display information according to the related recommended questions; and fill the guided question display information into the search result display page.

[0229] B15. The search result page generation device as described in B14, wherein the page generation module is further used to determine a follow-up keyword according to the follow-up search instruction when a feedback follow-up search instruction is received; perform a page search according to the follow-up keyword to obtain a follow-up search result, and generate follow-up response information according to the follow-up keyword through a preset language model; generate follow-up display information based on the follow-up search result and the follow-up response information; and update the follow-up display information to the search result display page.

[0230] B16. In the search result page generation device as described in B13, the page generation module is further used to determine the semantic information of each search display result in the search result page; determine abnormal search results with content conflicts in each search display result based on the semantic information; and generate and display authenticity analysis results based on the abnormal search results.

[0231] B17. In the search result page generation device as described in B16, the page generation module is further used to obtain channel information corresponding to each abnormal search result; determine the channel credibility corresponding to each abnormal search result based on the channel information; and generate and display the authenticity analysis results of the abnormal search results based on the channel credibility.

[0232] B18. In the search result page generation device as described in B17, the page generation module is further used to determine the authenticity score of each abnormal search result based on the authenticity analysis result; and select the best display result among the abnormal search results based on the authenticity score, and the best display result is the abnormal search result with the highest authenticity score.

[0233] The present invention also discloses C19, a search result page generating device, the search result page generating device comprising: a memory, a processor and a search result page generating program stored in the memory and running on the processor, the search result page generating program is configured to implement the search result page generating method as described above.

[0234] The present invention also discloses D20, a storage medium, on which a search result page generation program is stored, and when the search result page generation program is executed by a processor, the search result page generation method as described above is implemented.

Claims

1. A method for generating a search result page, characterized in that: The search result page generation method comprises: When receiving a search instruction, determining a search keyword according to the search instruction; Performing a page search according to the search keyword to obtain page search results, and generating response information according to the search keyword through a preset language model; A search result page is generated based on the page search result and the response information.

2. The method for generating a search result page according to claim 1, wherein: The step of determining a search keyword according to the search instruction upon receiving the search instruction includes: When receiving a search instruction, determining search text information according to the search instruction; Extracting text word information according to the search text information; Determine a search keyword based on the text word information.

3. The method for generating a search result page according to claim 1, wherein: The performing of page search according to the search keyword to obtain page search results, and generating response information according to the search keyword through a preset language model includes: Importing the search keyword into a preset search engine to perform a page search and obtain page search results; The search keyword is imported into a preset language model, and output response information is obtained.

4. The method for generating a search result page according to claim 1, wherein: The generating a search result page based on the page search result and the response information includes: Generate a response summary answer according to the response information; Determining a target number of screening display results based on the page search results; A search result page is generated according to the response summary answer and the screening display result.

5. The method for generating a search result page according to claim 4, wherein: The step of generating a response summary answer according to the response information includes: Determine a response result text according to the response information; Importing the response result text into the preset language model to obtain output response summary information; A response summary answer is generated according to the response summary information.

6. The method for generating a search result page according to claim 4, wherein: The determining of the target number of screening and displaying results based on the page search results includes: Determine the search result ranking according to the page search results; A screening display result is selected from the page search results according to the search ranking result, and the screening display result is a target number of page search results selected according to the search ranking result.

7. The method for generating a search result page according to claim 4, wherein: Generating a search result page according to the response summary answer and the screening display result includes: Creating a blank display page, and dividing the blank display page into a priority display area and a reference display area; Fill the response summary answer into the priority display area, and fill the screening display result into the reference display area to obtain an updated display area; A search result page is generated according to the updated display area and the blank display page.

8. A search result page generating device, characterized in that: The search result page generating device comprises: A keyword determination module, configured to determine a search keyword according to the search instruction when a search instruction is received; A keyword search module, used to perform page search according to the search keyword, obtain page search results, and generate response information according to the search keyword through a preset language model; The page generation module is used to generate a search result page based on the page search result and response information.

9. A search result page generating device, characterized in that: The search result page generating device comprises: a memory, a processor, and a search result page generating program stored in the memory and running on the processor, wherein the search result page generating program is configured to implement the search result page generating method according to any one of claims 1 to 7.

10. A storage medium, characterized in that: The storage medium stores a search result page generation program, and when the search result page generation program is executed by the processor, the search result page generation method according to any one of claims 1 to 7 is implemented.

Images