Information generation method and device, information interaction method and device, electronic equipment and medium

Abstract

The invention provides an information generation method and device, an information interaction method and device, electronic equipment and a storage medium, and relates to the technical field of artificial intelligence, in particular to the technical field of natural language processing, knowledge questions and answers, large models and generation models. According to the specific implementation scheme, received query information is responded, intention analysis is conducted on the query information, intention analysis results are obtained, and the intention analysis results are used for triggering different retrieval modes; triggering a professional retrieval mode in response to the intention analysis result, and retrieving to obtain a reference resource comprising at least one professional resource; generating professional reply information according to the reference resources; and outputting professional reply information.

Description

Technical Field

[0001] The present disclosure relates to the field of artificial intelligence technology, and in particular to the field of natural language processing, knowledge question answering, large models and generative model technology. More specifically, the present disclosure provides an information generation method, an information interaction method, an apparatus, an electronic device, a storage medium and a computer program product. Background Art

[0002] With the development of artificial intelligence technology, the application of knowledge question answering based on big models is becoming more and more widespread. At present, the question answering system based on retrieval augmented generation (Retrieval Augmented Generation) can retrieve knowledge from the knowledge base according to the user's questions, and then send the retrieved knowledge into the big model to generate reply information, so as to improve the accuracy of the reply information. Summary of the invention

[0003] The present disclosure provides an information generation method, an apparatus, an electronic device, a storage medium, and a computer program product.

[0004] According to a first aspect, an information generation method is provided, the method comprising: in response to receiving query information, performing intent analysis on the query information to obtain intent analysis results, wherein the intent analysis results are used to trigger different retrieval modes; in response to the intent analysis results, triggering a professional retrieval mode to retrieve reference resources including at least one professional resource; generating professional reply information based on the reference resources; and outputting the professional reply information.

[0005] According to the second aspect, an information interaction method is provided, the method comprising: in response to receiving query information, calling an intent analysis model to perform intent analysis on the query information to obtain an intent analysis result, wherein the intent analysis result is used to trigger different retrieval modes; displaying first reference resource information, wherein the first reference resource includes at least one professional resource, and the first reference resource is retrieved when the intent analysis result triggers a professional retrieval mode; and displaying professional reply information, wherein the professional reply information is generated by extracting a reference fragment from the first reference resource.

[0006] According to a third aspect, an information generating device is provided, which includes: an intention analysis module, which is used to perform intention analysis on the query information in response to receiving the query information, and obtain the intention analysis result, wherein the intention analysis result is used to trigger different retrieval modes; a professional retrieval module, which is used to trigger the professional retrieval mode in response to the intention analysis result, and retrieve reference resources containing at least one professional resource; a professional reply information generating module, which is used to generate professional reply information based on the reference resources; and a professional reply information output module, which is used to output professional reply information.

[0007] According to a fourth aspect, an information interaction device is provided, which includes: an information receiving module, which is used to call an intent analysis model to perform intent analysis on the query information in response to receiving query information, and obtain an intent analysis result, wherein the intent analysis result is used to trigger different retrieval modes; a first display module, which is used to display first reference resource information, wherein the first reference resource includes at least one professional resource, and the first reference resource is retrieved when the intent analysis result triggers a professional retrieval mode; and a second display module, which is used to display professional reply information, wherein the professional reply information is generated by extracting reference fragments from the first reference resource.

[0008] According to a fifth aspect, an electronic device is provided, 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 the method provided according to the present disclosure.

[0009] According to a sixth aspect, a non-transitory computer-readable storage medium storing computer instructions is provided, wherein the computer instructions are used to cause a computer to execute the method provided according to the present disclosure.

[0010] According to a seventh aspect, a computer program product is provided, comprising a computer program, wherein the computer program is stored on at least one of a readable storage medium and an electronic device, and wherein the computer program implements the method provided according to the present disclosure when executed by a processor.

[0011] It should be understood that the content described in this section is not intended to identify the key or important features of the embodiments of the present disclosure, nor is it intended to limit the scope of the present disclosure. Other features of the present disclosure will become easily understood through the following description. BRIEF DESCRIPTION OF THE DRAWINGS

[0012] The accompanying drawings are used to better understand the present solution and do not constitute a limitation of the present disclosure.

[0013] Figure 1 is a schematic diagram of an exemplary system architecture to which the information generation method and the information interaction method can be applied according to an embodiment of the present disclosure;

[0014] Figure 2 is a flow chart of an information generation method according to an embodiment of the present disclosure;

[0015] Figure 3 is a system schematic diagram of an information generation method according to an embodiment of the present disclosure;

[0016] Figure 4is a flow chart of an information generation method in a professional retrieval mode according to an embodiment of the present disclosure;

[0017] Figure 5 is a flow chart of an information interaction method according to an embodiment of the present disclosure;

[0018] Figure 6 is a schematic diagram of an information interaction method according to an embodiment of the present disclosure;

[0019] Figure 7 is a schematic diagram of an information interaction method according to another embodiment of the present disclosure;

[0020] Figure 8 is a block diagram of an information generating device according to an embodiment of the present disclosure;

[0021] Fig. 9 is a block diagram of an information interaction device according to an embodiment of the present disclosure; and

[0022] Fig.10 It is a block diagram of an electronic device according to at least one of the information generation method and the information interaction method of an embodiment of the present disclosure. DETAILED DESCRIPTION

[0023] The following is a description of exemplary embodiments of the present disclosure in conjunction with the accompanying drawings, including various details of the embodiments of the present disclosure to facilitate understanding, which should be considered as merely exemplary. Therefore, it should be recognized by those of ordinary skill in the art that various changes and modifications may be made to the embodiments described herein without departing from the scope and spirit of the present disclosure. Similarly, for the sake of clarity and conciseness, descriptions of well-known functions and structures are omitted in the following description.

[0024] At present, most of the question-answering systems based on search enhancement generation rely on general search sources, such as various web pages publicly available on the Internet. However, this general search mode based on general search sources cannot meet the user's demand for obtaining professional information.

[0025] In view of this, the embodiments of the present disclosure provide an information generation method, which adds a professional search mode based on the current general search mode. In the professional search mode, professional resources in professional search sources are intelligently recommended to users. Professional search sources include, for example, paper journal databases or websites, various academic official websites, professional sites, etc. Professional resources in professional search sources include, for example, journal papers, conference papers, books and other professional documents, which can meet the needs of users to quickly obtain academic information and vertical information, give full play to the information collection and collation capabilities of retrieval enhancement, provide more authoritative and professional answers, and enhance user trust.

[0026] The embodiment of the present disclosure also provides an information interaction method for professional reply information generated based on the information generation method, which can display the source information of professional resources and improve the user experience.

[0027] In the technical solution of the present disclosure, the collection, storage, use, processing, transmission, provision and disclosure of user personal information involved are in compliance with the provisions of relevant laws and regulations and do not violate public order and good morals.

[0028] In the technical solution of the present disclosure, the user's authorization or consent is obtained before obtaining or collecting the user's personal information.

[0029] Figure 1 is a schematic diagram of an exemplary system architecture to which the information generation method and the information interaction method can be applied according to an embodiment of the present disclosure. It should be noted that: Figure 1 What is shown is merely an example of a system architecture to which the embodiments of the present disclosure can be applied, in order to help those skilled in the art understand the technical content of the present disclosure, but it does not mean that the embodiments of the present disclosure cannot be used in other devices, systems, environments or scenarios.

[0030] like Figure 1 As shown, the system architecture 100 according to this embodiment may include terminal devices 101, 102, 103, a network 104 and a server 105. The network 104 is used to provide a medium for communication links between the terminal devices 101, 102, 103 and the server 105. The network 104 may include various connection types, such as wired and / or wireless communication links, etc.

[0031] Users can use terminal devices 101, 102, 103 to interact with server 105 through network 104 to receive or send messages, etc. Terminal devices 101, 102, 103 can be various electronic devices, including but not limited to smart phones, tablet computers, laptops, etc.

[0032] The server 105 may be a server that provides various services, such as a background management server that provides support for websites browsed by users using the terminal devices 101, 102, and 103 (for example only). The background management server may analyze and process the received data such as user requests, and feed back the processing results (such as generated professional reply information) to the terminal device. The server 102 may be deployed with a trained predetermined model, which may be a large language model (LLM).

[0033] The information generation method provided in the embodiment of the present disclosure can generally be executed by the server 105. Accordingly, the information retrieval device provided in the embodiment of the present disclosure can generally be set in the server 105.

[0034] The information interaction method provided in the embodiment of the present disclosure can generally be executed by the terminal devices 101, 102, and 103. Accordingly, the information interaction apparatus provided in the embodiment of the present disclosure can generally be set in the terminal devices 101, 102, and 103.

[0035] Figure 2 is a flow chart of an information generating method according to an embodiment of the present disclosure.

[0036] like Figure 2 As shown, the information generation method 200 includes operations S210 to S240. The execution subject of this embodiment may be a server.

[0037] In operation S210, in response to receiving the query information, intent analysis is performed on the query information to obtain an intent analysis result.

[0038] The query information can be a question (Query) input by the user. The intent analysis model can be used to perform intent analysis on the query information to determine the user's search intent. Based on the user's search intent, it can be determined which retrieval mode is suitable for the user's search needs.

[0039] The search mode can include general search mode (also called ordinary search mode) and professional search mode. General search mode can refer to searching in the web pages open to the whole network, and professional search mode can refer to searching in professional databases (such as papers, journals, etc.), academic official websites, professional sites and other professional search sources.

[0040] For example, if it is determined that the user's search intention is to obtain information related to professional documents such as papers and books, it can be determined that the user's search needs are suitable for professional search mode. If it is determined that the user's search intention does not include obtaining information related to professional documents such as papers and books, it can be determined that the user's search needs are suitable for general search mode.

[0041] If it is determined that the user's search needs are suitable for the professional search mode, the professional search mode can be triggered for search. If it is determined that the user's search needs are suitable for the ordinary search mode, the ordinary search mode can be triggered for search.

[0042] For example, if the query entered by the user is "provide some papers on XX", the intent analysis model can analyze that the user's intention is to obtain papers on the topic of XX, and this intention can trigger the professional retrieval mode.

[0043] For example, if the query entered by the user is "Which documents mention XX?", the intent analysis model can analyze that the user's intention is to find specific documents containing the topic of XX, and this intention can trigger the professional search mode.

[0044] For example, if the query entered by the user is "effects of American ginseng", the intent analysis model can analyze that the user's intention is to understand the effects of the common medicinal material American ginseng. This type of information can generally be found in general resources, so the normal search mode can be triggered.

[0045] The above-mentioned intent analysis model may be a fine-tuned large model, which may be trained using a sample query data set and the actual intent of the sample query.

[0046] In operation S220, in response to triggering a professional retrieval mode based on the intention analysis result, reference resources including at least one professional resource are retrieved.

[0047] When the user's search needs trigger the professional search mode, at least one professional resource can be retrieved from the professional search source by searching the professional search source. The professional search source may include a paper journal database, an academic official website, a professional website, etc. The retrieved professional resources may include journal papers, conference papers, books and other professional documents.

[0048] In operation S230, professional reply information is generated according to the reference resource.

[0049] In operation S240, professional reply information is output.

[0050] For example, the professional resources that the user is looking for can be returned to the user, and key information and content can be extracted from the professional resources according to the user's questions to generate a reply text, and the reply text can be returned to the user. The reply text can also be used as the main text, and the retrieved professional resources can be used as references, and the main text and references can be combined and returned to the user.

[0051] The embodiments of the present disclosure determine the user's search intent by performing intent analysis on the user's query information. When the user's search intent triggers a professional search mode, at least one professional resource is retrieved and professional reply information is generated based on the professional resources, which can improve the authority of the reply information, enhance user trust, and improve user experience.

[0052] According to an embodiment of the present disclosure, operation S220 also includes triggering a professional retrieval mode in response to the intent analysis result, retrieving at least one professional resource from a professional retrieval source; and retrieving at least one first general resource from a general retrieval source.

[0053] When the user's search demand triggers the professional search mode, the professional search source can be searched to obtain at least one professional resource.

[0054] In addition, when the user's search needs trigger the professional search mode, the general search mode can also be triggered at the same time to search the general search source, that is, the general resources on the entire network, to obtain at least one first general resource. The search results of the general search source can supplement the professional search results.

[0055] The at least one professional resource and the at least one first general resource may be returned to the user as reference resources. Key information may be extracted and content summarized for the at least one professional resource and the at least one first general resource to generate a reply text, and the reply text may be returned to the user. The reply text may be used as the main text, the at least one professional resource and the at least one first general resource as references, and the main text and references may be returned to the user in combination.

[0056] According to an embodiment of the present disclosure, when the user's search intention triggers a professional search mode, a general search mode can also be triggered at the same time, which can supplement the professional search results, making the search results more comprehensive and facilitating the generation of reply information.

[0057] Figure 3 is a system diagram of an information generating method according to an embodiment of the present disclosure.

[0058] like Figure 3 As shown, the system of this embodiment includes an intention analysis module 310 , a professional retrieval module 320 and a general retrieval module 330 .

[0059] The user inputs a question Query, and the intention analysis module 310 performs an intention analysis on the question Query to determine the user's search intention. Different search intentions can trigger different search modes. For example, if it is determined that the user's search intention is to obtain professional literature, the professional search mode can be triggered. In the professional search mode, the professional search module 320 searches the professional search source to obtain at least one professional resource. Professional search sources may include paper journal databases or websites, academic official websites, and professional sites, and professional resources may include papers, books and other documents.

[0060] In addition, the general search mode can also be triggered in the professional search mode. In the general search mode, the general search module 330 searches the general search source to obtain at least one first general resource. For example, the general search source can be a website or URL open to the whole network, and the first general resource can be webpage content.

[0061] When it is determined that the user's search intention does not include obtaining professional documents, only the general search mode can be triggered. In the general search mode, the general search module 330 searches the general search source to obtain at least one second general resource. For example, the general search source can be a website or URL publicly available on the Internet, and the second general resource can be web page content.

[0062] Table 1 below shows examples of different retrieval modes triggered by the intent analysis module 310 and retrieval source information under different retrieval modes.

[0063] Table 1

[0064]

[0065] As shown in Table 1, the intention analysis module 310 can determine whether the user's intention involves obtaining professional resources such as documents and papers. If it is determined that professional resources are to be obtained, the professional search mode can be triggered, otherwise the general search mode is triggered.

[0066] In the professional search mode, at least one professional resource can be retrieved from the professional search source, and at least one first general resource can be retrieved from the general search source. For the professional resource and the first general resource, the big model can be used to extract key information, summarize it, and generate professional reply information 321.

[0067] In the common search mode, at least one second common resource can be retrieved from the common search source. For the second common resource, the big model can also be used to extract key information and summarize the content to generate common reply information 331.

[0068] The above professional reply information or general reply information can be returned to the user, and this round of process ends. The above process can be repeated for the next Query.

[0069] The embodiments of the present disclosure identify user intentions. Different user intentions trigger different retrieval modes. At least one of professional retrieval resources and general retrieval resources is retrieved under different retrieval modes and used as a reference resource. Professional reply information or general reply information is generated based on the reference resource, which can meet the user's needs for obtaining professional information and improve user experience.

[0070] Figure 4 It is a flow chart of an information generation method in a professional retrieval mode according to an embodiment of the present disclosure.

[0071] like Figure 4 As shown, this embodiment includes operations S410 to S460.

[0072] In operation S410, at least one professional resource is retrieved from a professional search source, and at least one first general resource is retrieved from a general search source.

[0073] In the professional search mode, professional search sources can be searched, and in order to obtain more comprehensive information, general search sources can also be searched at the same time. The search results may include at least one professional resource and at least one first general resource.

[0074] After obtaining the search results, professional response information may be generated, which may include the main text content and reference content.

[0075] The following operations S420 to S440 are steps for generating the main content, and operation S450 is a step for generating the reference content.

[0076] In operation 420 , a plurality of reference segments are extracted from at least one professional resource and at least one general resource.

[0077] In operation 430 , body content is generated based on the plurality of reference segments.

[0078] The big model can be used to extract key information from professional resources and first general resources to obtain multiple reference fragments, and then the multiple reference fragments can be integrated and summarized to obtain the main text content.

[0079] At operation 440 , a reference icon is inserted into the text, wherein the reference icon is associated with a reference resource referenced by the reference snippet.

[0080] According to an embodiment of the present disclosure, a reference icon corresponding to a reference segment is generated; the reference icon is associated with a reference resource referenced by the reference segment; and the reference icon is inserted into the body content.

[0081] For example, the generated text content may include multiple paragraphs. In a paragraph containing a reference fragment, a reference icon may be inserted. The reference icon may be associated with the reference resource cited by the reference fragment in the paragraph, thereby realizing document traceability.

[0082] For example, a reference icon can be inserted at the location of a reference fragment in a paragraph. When a user clicks the reference icon or hovers the mouse over the reference icon, information about the reference resource cited by the reference fragment can be displayed, such as the name of the reference. The user can also jump to the source webpage of the reference by clicking the name of the reference.

[0083] For example, a reference icon may be inserted at the end of a paragraph. If a reference fragment in the paragraph cites multiple reference resources, only one reference icon may be inserted at the end of the paragraph. The reference icon may be associated with multiple reference resources cited by the reference mark in the paragraph. When a user clicks the reference icon or hovers the mouse over the reference icon, the multiple reference resource information may be displayed, such as the title of each reference. For each reference, the user may click the name of the reference to jump to the source webpage of the reference.

[0084] In operation 450 , reference content is generated based on at least one professional resource and at least one first general resource.

[0085] The reference content may include the name or source information of the reference resource. For example, the reference content may include: referring to 3 information sources, 2 of which are professional information sources such as "XX University Journal".

[0086] In operation 460, professional reply information is generated according to the main text content and the reference content, and the professional reply information is output.

[0087] The above main text content and reference content can be combined together as professional reply information and returned to the terminal used by the user.

[0088] The embodiments of the present disclosure insert reference icons into professional reply information, and the reference icons are associated with reference resources cited by reference fragments in the text, so as to achieve document traceability and improve user interaction experience.

[0089] According to an embodiment of the present disclosure, the present disclosure also provides an information interaction method.

[0090] Figure 5 It is a flowchart of an information interaction method according to an embodiment of the present disclosure.

[0091] like Figure 5 As shown, the information interaction method includes operations S510 to S530. The execution subject of this embodiment may be a terminal device used by a user.

[0092] In operation S510, in response to receiving query information, an intent analysis model is called to perform intent analysis on the query information to obtain an intent analysis result.

[0093] The query information can be a question (Query) input by the user. After receiving the query information, the terminal can call the intention analysis model of the server to perform intent analysis on the query information to determine the user's search intention. If it is determined that the user's search intention is to obtain information related to professional documents such as papers and books, the professional search mode can be triggered. If it is determined that the user's search intention does not include obtaining information related to professional documents such as papers and books, the general search mode can be triggered.

[0094] The search mode can include general search mode (also called ordinary search mode) and professional search mode. The general search mode can refer to searching among all publicly available web pages on the entire network, and the professional search mode can refer to searching professional databases (such as papers, journals, etc.), academic official websites, professional sites and other professional search sources.

[0095] If it is determined that the user's search intention is suitable for the professional search mode, the professional search mode can be triggered for search. If it is determined that the user's search intention is suitable for the ordinary search mode, the ordinary search mode can be triggered for search.

[0096] The above-mentioned intent recognition model can be a fine-tuned large model, and the large model can be trained using a sample query data set and the actual intent of the sample query.

[0097] In operation S520, first reference resource information is displayed.

[0098] The first reference resource information includes the title and source information of the first reference resource. The first reference resource may be obtained by searching a professional search source when the intention analysis result triggers a professional search mode.

[0099] The first reference resource may include at least one professional resource, such as journal articles, conference papers, books, and other documents.

[0100] After the user enters a query, the title and source information of at least one professional resource can be displayed to the user. For example, the display text can be: Reading professional material: xxxxx (document title). And the titles of multiple documents can be displayed in rotation.

[0101] In operation S530, professional reply information is displayed.

[0102] The professional reply information may be generated based on the first reference resource. It may include text content and reference content. The text part may be obtained by extracting multiple reference fragments from the first reference resource and integrating and summarizing the reference fragments. The reference content may include the title and source information of the first reference resource.

[0103] The embodiments of the present disclosure can trigger a professional search mode by performing intent analysis on the user's query information. In the professional search mode, professional resources can be displayed to the user, as well as professional reply information generated based on the professional resources, which can enhance user trust and improve user experience.

[0104] Figure 6 It is a schematic diagram of an information interaction method according to an embodiment of the present disclosure.

[0105] like Figure 6 As shown, pages 610-640 may be a series of display pages after the user enters a query. For example, after the user enters a query, if the professional search mode is triggered, the text in page 610 "Start professional search...", "In-depth analysis of professional information..." may be displayed to the user.

[0106] After the reference resource is retrieved, the text "Reading professional status: xxxxx" in page 620 may be displayed, where "xxxxx" may be the title of the document. In the case where there are multiple reference resources (reference documents) retrieved, the titles of the multiple reference documents may be displayed in rotation.

[0107] Next, the reference resource information can be displayed. For example, the text displayed on page 630 is "Reference to 3 information sources, 2 of which are from professional information sources such as "XX University Journal". The reference resource information can also be expanded, and after expansion, a reference resource list can be displayed. For example, the detailed information of "1. The role of XXX in the development process" displayed on page 630 can also include the source (such as XX University Journal) and the author (such as author A, author B).

[0108] The reference resource information displayed on the above page 630 can exist stably together with the main text content returned by the large model after it appears. For example, page 640 displays professional reply information, including the reference resource information displayed on page 630 and the main text content generated by the large model, and also includes an input box for interacting with the user. The user can enter a new query in the input box to execute the next round of question and answer.

[0109] The embodiments of the present disclosure display a series of display pages in the professional search mode, which can enhance user trust and improve user experience.

[0110] Figure 7 is a schematic diagram of an information interaction method according to another embodiment of the present disclosure.

[0111] like Figure 7As shown, page 710 may be a display page for professional reply information. The page may include reference content, text content, and an input box. The reference content may be reference resource information displayed before the text content is displayed, that is, after the reference resource information appears, it may be retained, and after the text content appears, the reference content and the text content are displayed together as professional reply information.

[0112] The text content of the embodiment of the present disclosure may include multiple paragraphs, such as the first paragraph content, the second paragraph content, and the third paragraph content. For each paragraph, if the paragraph contains a reference fragment, a reference icon may be displayed at the end of the paragraph. The reference fragment may be key information extracted from the reference resource.

[0113] According to an embodiment of the present disclosure, the professional reply information includes a body content, the body content includes a reference segment and a reference icon corresponding to the reference segment, and the reference icon is associated with a first reference resource cited by the reference segment. The information interaction method also includes, in response to receiving a selection operation on the reference icon, displaying the first reference resource information associated with the reference icon; and in response to receiving a selection operation on the first reference resource, displaying the source information of the first reference resource.

[0114] For example, the first paragraph content and the second paragraph content may both contain reference fragments, and reference icons are displayed at the end of the first paragraph and the third paragraph. The reference icon is associated with the reference resource cited by the reference fragment contained in the paragraph.

[0115] For example, the reference resource cited by the reference fragment included in the first paragraph is document A. When the user clicks the reference icon at the end of the first paragraph, or the user hovers the mouse over the reference icon at the end of the first paragraph, the title of document A can be displayed. When the user clicks the title of document A, it can jump to the source webpage of document A.

[0116] For example, the reference fragment included in the third paragraph contains multiple reference resources, such as multiple reference resources including document A, document B and document C. A reference icon may be displayed at the end of the third paragraph. When the user clicks the reference icon at the end of the third paragraph, or the user hovers the mouse over the reference icon at the end of the third paragraph, the titles of document A, document B, document C, etc. may be displayed. For each document title, when the user clicks the title of the document, the user may jump to the source webpage of the document.

[0117] According to an embodiment of the present disclosure, the reference resources displayed on page 710 may include at least one professional resource retrieved from a professional search source, and may also include at least one first ordinary resource retrieved from an ordinary search source. That is, when the user intends to trigger the professional search mode, the ordinary search mode may also be triggered at the same time, and the search results of the ordinary search mode may serve as a supplement to the professional search results. The reference fragment contained in the text content may be extracted from at least one professional resource and at least one ordinary resource. When the user clicks on the reference icon, the title of the professional resource referenced by the reference fragment may be displayed, or the title of the ordinary resource referenced by the reference fragment may be displayed. At the same time, the user may also jump to the source web page of the ordinary resource by clicking on the ordinary resource title.

[0118] According to an embodiment of the present disclosure, the information interaction method also includes displaying second reference resource information, wherein the second reference resource includes at least one second ordinary resource, and the second ordinary resource is retrieved when the intention analysis result triggers an ordinary retrieval mode; and displaying ordinary reply information, wherein the ordinary reply information is generated based on at least one second ordinary resource.

[0119] When the user intends to trigger the general search mode, only the general search source can be searched to obtain at least one second general resource, and a reference fragment can be extracted from the second general resource to generate general reply information. The display and interaction of the general reply information can be consistent with the display and interaction of the professional reply information.

[0120] For example, reference content may also be displayed, the reference content includes the title and source of the second common reference resource, the body content is displayed, and the end of the paragraph containing the reference segment in the body content includes a reference icon. When the user hovers the mouse or clicks the reference icon, the title of the reference resource may be displayed, and when the user clicks the title, the user may jump to the source webpage of the reference resource.

[0121] The display page of the embodiment of the present disclosure provides a user interaction function, through which the user can view the reference resource information cited in the text content, and can also jump to the source web page of the reference resource, thereby realizing document tracing and improving the user experience.

[0122] According to an embodiment of the present disclosure, the present disclosure also provides an information generating device and an information interacting device.

[0123] Figure 8 is a block diagram of an information generating device according to an embodiment of the present disclosure.

[0124] like Figure 8As shown, the information generating device 800 includes an intention analyzing module 810 , a professional retrieval module 820 , a professional reply information generating module 830 and a professional reply information output module 840 .

[0125] The intent analysis module 810 is used to perform intent analysis on the query information in response to receiving the query information, and obtain the intent analysis result, wherein the intent analysis result is used to trigger different retrieval modes.

[0126] The professional search module 820 is used to trigger a professional search mode in response to the intention analysis result, and retrieve reference resources including at least one professional resource.

[0127] The professional reply information generating module 830 is used to generate professional reply information according to reference resources.

[0128] The professional reply information output module 840 is used to output professional reply information.

[0129] According to an embodiment of the present disclosure, the reference resource further includes at least one first common resource. The professional search module 820 includes a first search unit and a second search unit.

[0130] The first retrieval unit is used to trigger a professional retrieval mode in response to the intention analysis result, and retrieve at least one professional resource from a professional retrieval source.

[0131] The second retrieval unit is used to retrieve at least one first common resource from a common retrieval source.

[0132] The professional reply information includes the body content. The professional reply information generation module 830 includes an extraction unit, a body generation unit and a correlation unit.

[0133] The extraction unit is used to extract a plurality of reference segments from at least one professional resource and at least one first common resource.

[0134] The body text generation unit is used to generate body text content of professional reply information according to multiple reference fragments.

[0135] The association unit is used to associate, for each reference fragment, the reference fragment with a reference resource referenced by the reference fragment.

[0136] The associating unit includes an icon generating subunit, an associating subunit and an inserting subunit.

[0137] The icon generation subunit is used to generate a reference icon corresponding to the reference segment.

[0138] The association subunit is used to associate the reference icon with the reference resource referenced by the reference fragment.

[0139] The Insert subunit is used to insert a reference icon into the main text content.

[0140] The professional reply information includes reference content. The professional reply information generation module 830 includes a reference content generation unit.

[0141] The reference content generating unit is used to generate reference content according to at least one professional resource and at least one first common resource.

[0142] The information generating device 800 further includes a general search module, a general reply information generating module and a general reply information output module.

[0143] The common search module is used to trigger the common search mode in response to the intention analysis result, and retrieve at least one second common resource from the common search source.

[0144] The common reply information generating module is used to generate common reply information according to at least one second common resource.

[0145] The general reply information output module is used to output general reply information.

[0146] Fig. 9 It is a block diagram of an information interaction device according to an embodiment of the present disclosure.

[0147] like Fig. 9 As shown, the information interaction device 900 includes an information receiving module 910 , a first display module 920 and a second display module 930 .

[0148] The information receiving module 910 is used to call the intention analysis model to perform intention analysis on the query information in response to receiving the query information, and obtain the intention analysis result, wherein the intention analysis result is used to trigger different retrieval modes.

[0149] The first display module 920 is used to display first reference resource information, wherein the first reference resource includes at least one professional resource, and the first reference resource is retrieved when the intention analysis result triggers a professional retrieval mode.

[0150] The second display module 930 is used to display professional reply information, wherein the professional reply information is generated by extracting reference fragments from the first reference resource.

[0151] The professional reply information includes a body content, the body content includes a reference segment and a reference icon corresponding to the reference segment, and the reference icon is associated with the first reference resource cited by the reference segment. The information interaction device 900 also includes a first interaction module and a second interaction module.

[0152] The first interaction module is used for displaying first reference resource information associated with the reference icon in response to receiving a selection operation on the reference icon.

[0153] The second interaction module is used to display source information of the first reference resource in response to receiving a selection operation on the first reference resource.

[0154] The first reference resource also includes at least one first general resource, and the professional reply information is generated by extracting reference fragments from the at least one professional resource and the at least one first general resource.

[0155] The information interaction device 900 further includes a third display module and a fourth display module.

[0156] The third display module is used to display second reference resource information, wherein the second reference resource includes at least one second common resource, and the second common resource is retrieved when the intent analysis result triggers a common retrieval mode.

[0157] The fourth display module is used to display common reply information, wherein the common reply information is generated according to at least one second common resource.

[0158] According to an embodiment of the present disclosure, the present disclosure also provides an electronic device, a readable storage medium and a computer program product.

[0159] Fig.10 A schematic block diagram of an example electronic device 1000 that can be used to implement an embodiment of the present disclosure is shown. The electronic device is intended to represent various forms of digital computers, such as laptop computers, desktop computers, workstations, personal digital assistants, servers, blade servers, mainframe computers, and other suitable computers. The electronic device can also represent various forms of mobile devices, such as personal digital processing, cellular phones, smart phones, wearable devices, and other similar computing devices. The components shown herein, their connections and relationships, and their functions are merely examples and are not intended to limit the implementation of the present disclosure described and / or required herein.

[0160] like Fig.10 As shown, the device 1000 includes a computing unit 1001, which can perform various appropriate actions and processes according to a computer program stored in a read-only memory (ROM) 1002 or a computer program loaded from a storage unit 1008 into a random access memory (RAM) 1003. In the RAM 1003, various programs and data required for the operation of the device 1000 can also be stored. The computing unit 1001, the ROM 1002, and the RAM 1003 are connected to each other via a bus 1004. An input / output (I / O) interface 1005 is also connected to the bus 1004.

[0161] A number of components in the device 1000 are connected to the I / O interface 1005, including: an input unit 1006, such as a keyboard, a mouse, etc.; an output unit 1007, such as various types of displays, speakers, etc.; a storage unit 1008, such as a disk, an optical disk, etc.; and a communication unit 1009, such as a network card, a modem, a wireless communication transceiver, etc. The communication unit 1009 allows the device 1000 to exchange information / data with other devices through a computer network such as the Internet and / or various telecommunication networks.

[0162] The computing unit 1001 may be a variety of general and / or special processing components with processing and computing capabilities. Some examples of the computing unit 1001 include, but are not limited to, a central processing unit (CPU), a graphics processing unit (GPU), various dedicated artificial intelligence (AI) computing chips, various computing units running machine learning model algorithms, digital signal processors (DSPs), and any appropriate processors, controllers, microcontrollers, etc. The computing unit 1001 performs the various methods and processes described above, such as at least one of the information generation method and the information interaction method. For example, in some embodiments, at least one of the information generation method and the information interaction method may be implemented as a computer software program, which is tangibly contained in a machine-readable medium, such as a storage unit 1008. In some embodiments, part or all of the computer program may be loaded and / or installed on the device 1000 via the ROM 1002 and / or the communication unit 1009. When the computer program is loaded into the RAM 1003 and executed by the computing unit 1001, one or more steps of at least one of the information generation method and the information interaction method described above may be performed. Alternatively, in other embodiments, the computing unit 1001 may be configured to execute at least one of the information generation method and the information interaction method in any other appropriate manner (eg, by means of firmware).

[0163] Various implementations of the systems and techniques described above herein can be implemented in digital electronic circuit systems, integrated circuit systems, field programmable gate arrays (FPGAs), application specific integrated circuits (ASICs), application specific standard products (ASSPs), systems on chips (SOCs), complex programmable logic devices (CPLDs), computer hardware, firmware, software, and / or combinations thereof. These various implementations can include: being implemented in one or more computer programs that can be executed and / or interpreted on a programmable system including at least one programmable processor, which can be a special purpose or general purpose programmable processor that can receive data and instructions from a storage system, at least one input device, and at least one output device, and transmit data and instructions to the storage system, the at least one input device, and the at least one output device.

[0164] The program code for implementing the method of the present disclosure may be written in any combination of one or more programming languages. These program codes may be provided to a processor or controller of a general-purpose computer, a special-purpose computer, or other programmable data processing device, so that the program code, when executed by the processor or controller, enables the functions / operations specified in the flow chart and / or block diagram to be implemented. The program code may be executed entirely on the machine, partially on the machine, partially on the machine and partially on a remote machine as a stand-alone software package, or entirely on a remote machine or server.

[0165] In the context of the present disclosure, a machine-readable medium may be a tangible medium that may contain or store a program for use by or in conjunction with an instruction execution system, device, or equipment. A machine-readable medium may be a machine-readable signal medium or a machine-readable storage medium. A machine-readable medium may include, but is not limited to, an electronic, magnetic, optical, electromagnetic, infrared, or semiconductor system, device, or device, or any suitable combination of the foregoing. A more specific example of a machine-readable storage medium may include an electrical connection based on one or more lines, a portable computer disk, a hard disk, a random access memory (RAM), a read-only memory (ROM), an erasable programmable read-only memory (EPROM or flash memory), an optical fiber, a portable compact disk read-only memory (CD-ROM), an optical storage device, a magnetic storage device, or any suitable combination of the foregoing.

[0166] To provide interaction with a user, the systems and techniques described herein can be implemented on a computer having: a display device (e.g., a CRT (cathode ray tube) or LCD (liquid crystal display) monitor) for displaying information to the user; and a keyboard and pointing device (e.g., a mouse or trackball) through which the user can provide input to the computer. Other types of devices can also be used to provide interaction with the user; for example, the feedback provided to the user can be any form of sensory feedback (e.g., visual feedback, auditory feedback, or tactile feedback); and input from the user can be received in any form (including acoustic input, voice input, or tactile input).

[0167] The systems and techniques described herein may be implemented in a computing system that includes back-end components (e.g., as a data server), or a computing system that includes middleware components (e.g., an application server), or a computing system that includes front-end components (e.g., a user computer with a graphical user interface or a web browser through which a user can interact with implementations of the systems and techniques described herein), or a computing system that includes any combination of such back-end components, middleware components, or front-end components. The components of the system may be interconnected by any form or medium of digital data communication (e.g., a communication network). Examples of communication networks include: a local area network (LAN), a wide area network (WAN), and the Internet.

[0168] A computer system may include clients and servers. Clients and servers are generally remote from each other and usually interact through a communication network. The relationship of client and server is generated by computer programs running on respective computers and having a client-server relationship to each other.

[0169] It should be understood that the various forms of processes shown above can be used to reorder, add or delete steps. For example, the steps recorded in this disclosure can be executed in parallel, sequentially or in different orders, as long as the desired results of the technical solutions disclosed in this disclosure can be achieved, and this document does not limit this.

[0170] The above specific implementations do not constitute a limitation on the protection scope of the present disclosure. It should be understood by those skilled in the art that various modifications, combinations, sub-combinations and substitutions can be made according to design requirements and other factors. Any modification, equivalent substitution and improvement made within the spirit and principle of the present disclosure shall be included in the protection scope of the present disclosure.

Claims

1. A method for generating information, comprising: In response to receiving the query information, performing intent analysis on the query information to obtain an intent analysis result, wherein the intent analysis result is used to trigger different retrieval modes; In response to the intention analysis result, a professional search mode is triggered to retrieve reference resources including at least one professional resource; Generate professional response information based on the reference resources; and Output the professional reply information.

2. The method according to claim 1, wherein: The reference resource also includes at least one first common resource; The triggering of the professional search mode in response to the intention analysis result, retrieving reference resources including at least one professional resource includes: In response to triggering a professional search mode based on the intention analysis result, the at least one professional resource is retrieved from a professional search source; as well as At least one first general resource is retrieved from a general search source.

3. The method according to claim 2, wherein: The professional reply information includes the text content; Generating professional reply information according to the reference resource includes: extracting a plurality of reference segments from the at least one professional resource and the at least one first general resource; Generating the body content of the professional reply information according to the multiple reference fragments; and For each reference fragment, the reference fragment is associated with a reference resource referenced by the reference fragment.

4. The method according to claim 3, wherein: For each reference segment, associating the reference segment with the reference resource referenced by the reference segment comprises: for each reference segment, generating a reference icon corresponding to the reference segment; Associating the reference icon with the reference resource referenced by the reference fragment; The reference icon is inserted into the body content.

5. The method according to claim 2, wherein: The professional response information includes reference content; Generating professional reply information according to the reference resource includes: Reference content is generated according to the at least one professional resource and the at least one first general resource.

6. The method according to claim 1, further comprising: In response to triggering a common search mode as a result of the intention analysis, at least one second common resource is retrieved from a common search source; generating common reply information according to the at least one second common resource; and Output the common reply information.

7. An information interaction method, comprising: In response to receiving the query information, calling the intention analysis model to perform intention analysis on the query information to obtain an intention analysis result, wherein the intention analysis result is used to trigger different retrieval modes; Displaying first reference resource information, wherein the first reference resource includes at least one professional resource, and the first reference resource is retrieved when the intention analysis result triggers a professional retrieval mode; and Professional response information is displayed, wherein the professional response information is generated by extracting reference fragments from the first reference resource.

8. The method according to claim 7, wherein: The professional reply information includes text content, the text content includes the reference segment and a reference icon corresponding to the reference segment, and the reference icon is associated with a first reference resource referenced by the reference segment; the method further includes: In response to receiving a selection operation on the reference icon, displaying first reference resource information associated with the reference icon; and In response to receiving a selection operation on the first reference resource, source information of the first reference resource is displayed.

9. The method according to claim 7 or 8, wherein: The first reference resource also includes at least one first general resource, and the professional reply information is generated by extracting reference fragments from the at least one professional resource and the at least one first general resource.

10. The method according to claim 7, further comprising: Displaying second reference resource information, wherein the second reference resource includes at least one second common resource, and the second common resource is retrieved when the intent analysis result triggers a common retrieval mode; and Displaying common reply information, wherein the common reply information is generated according to the at least one second common resource.

11. An information generating device, comprising: An intention analysis module, configured to, in response to receiving query information, perform intention analysis on the query information to obtain an intention analysis result, wherein the intention analysis result is used to trigger different retrieval modes; A professional search module, configured to trigger a professional search mode in response to the intention analysis result, and retrieve reference resources including at least one professional resource; A professional reply information generating module, used to generate professional reply information according to the reference resources; and The professional reply information output module is used to output the professional reply information.

12. The device according to claim 11, wherein The reference resource also includes at least one first general resource; the professional search module includes: A first retrieval unit, configured to trigger a professional retrieval mode in response to the intention analysis result, and retrieve the at least one professional resource from a professional retrieval source; and The second retrieval unit is used to retrieve at least one first common resource from a common retrieval source.

13. The device according to claim 12, wherein: The professional reply information includes text content; the professional reply information generation module includes: an extraction unit, configured to extract a plurality of reference segments from the at least one professional resource and the at least one first common resource; a body text generation unit, configured to generate body text content of the professional reply information according to the plurality of reference fragments; and The associating unit is used to associate, for each reference fragment, the reference fragment with a reference resource referenced by the reference fragment.

14. The device according to claim 13, wherein: The association unit comprises: An icon generation subunit, configured to generate a reference icon corresponding to the reference segment; an associating subunit, configured to associate the reference icon with a reference resource referenced by the reference fragment; The inserting subunit is used to insert the reference icon into the text content.

15. The device according to claim 12, wherein: The professional response information includes reference content; The professional reply information generating module comprises: The reference content generating unit is configured to generate reference content according to the at least one professional resource and the at least one first common resource.

16. The apparatus according to claim 11, further comprising: A common search module, configured to trigger a common search mode in response to the intention analysis result, and retrieve at least one second common resource from a common search source; a common reply information generating module, configured to generate common reply information according to the at least one second common resource; and The common reply information output module is used to output the common reply information.

17. An information interaction device, comprising: An information receiving module, configured to, in response to receiving query information, call an intention analysis model to perform intention analysis on the query information to obtain an intention analysis result, wherein the intention analysis result is used to trigger different retrieval modes; A first display module is used to display first reference resource information, wherein the first reference resource includes at least one professional resource, and the first reference resource is retrieved when the intention analysis result triggers a professional retrieval mode; and The second display module is used to display professional reply information, wherein the professional reply information is generated by extracting reference fragments from the first reference resource.

18. The device according to claim 17, wherein: The professional reply information includes text content, the text content includes the reference segment and a reference icon corresponding to the reference segment, and the reference icon is associated with a first reference resource cited by the reference segment; the device also includes: A first interaction module, configured to display first reference resource information associated with the reference icon in response to receiving a selection operation on the reference icon; and The second interaction module is configured to display source information of the first reference resource in response to receiving a selection operation on the first reference resource.

19. The device according to claim 17 or 18, wherein: The first reference resource also includes at least one first general resource, and the professional reply information is generated by extracting reference fragments from the at least one professional resource and the at least one first general resource.

20. The apparatus according to claim 19, further comprising: A third display module is used to display second reference resource information, wherein the second reference resource includes at least one second common resource, and the second common resource is retrieved when the intention analysis result triggers a common search mode; and The fourth display module is used to display common reply information, wherein the common reply information is generated according to the at least one second common resource.

21. An electronic device comprising: at least one processor; as well as a memory communicatively connected to the at least one processor; wherein, The memory stores instructions that can be executed by the at least one processor, and the instructions are executed by the at least one processor to enable the at least one processor to perform the method according to any one of claims 1 to 10.

22. A non-transitory computer-readable storage medium storing computer instructions, wherein: The computer instructions are used to cause the computer to execute the method according to any one of claims 1 to 10.

23. A computer program product, comprising a computer program, wherein the computer program is stored on at least one of a readable storage medium and an electronic device, and when the computer program is executed by a processor, the method according to any one of claims 1 to 10 is implemented.

Images