Business resource retrieval method and device based on AI large model, computer device and readable storage medium

Abstract

The application relates to an AI large model-based business resource retrieval method and device, computer equipment, a computer readable storage medium and a computer program product. The method comprises the following steps: acquiring a business file, and determining business file content in the business file; the business file content comprises at least two content elements with different content formats; the business file content is matched with preset business keywords according to the matching priority of the content format corresponding to each content element; key information matched with the business keywords in the business file content is extracted, and the key information is stored in a business resource question and answer database; in response to input business resource search text, a business resource search result for responding to the business resource search text is output based on the business file content corresponding to target key information; the target key information is key information matched with the business resource search text in the business resource question and answer database. The method can improve the accuracy of the retrieval result of the business resource.

Description

Technical Field

[0001] The present application relates to the field of artificial intelligence technology, and in particular to a business resource retrieval method, apparatus, computer equipment, computer-readable storage medium, and computer program product based on an AI big model. Background Art

[0002] As businesses continue to grow, they accumulate a vast amount of information within their business resource repositories, including reports, contracts, documentation, and training materials. Most companies organize these resources by file type. Searching for specific information often relies heavily on human organizers, which is time-consuming and labor-intensive. Furthermore, accurate searches based on document content are often difficult, and search results are often overly complex and require manual filtering.

[0003] Therefore, there is a problem in the related art that the accuracy of the retrieval results of business resources is low. Summary of the Invention

[0004] Based on this, it is necessary to provide a business resource retrieval method, device, computer equipment, computer-readable storage medium and computer program product based on an AI big model that can improve the accuracy of business resource retrieval results in response to the above technical problems.

[0005] In a first aspect, the present application provides a business resource retrieval method based on an AI big model, comprising:

[0006] Acquire a business file and determine the business file content in the business file; the business file content includes at least two content elements with different content formats;

[0007] Matching the business file content with preset business keywords according to the matching priority of the content format corresponding to each content element;

[0008] Extract key information from the business document content that matches the preset business keywords through the AI ​​big model, and store the key information in the business resource question and answer database;

[0009] In response to the input business resource search text, a business resource search result is outputted to respond to the business resource search text based on the business file content corresponding to the target key information; the target key information is the key information in the business resource question and answer database that matches the business resource search text.

[0010] In one embodiment, extracting key information from the business document content that matches the preset business keywords using the AI ​​big model includes:

[0011] Extracting target content from the business document content that matches the preset business keyword;

[0012] The key information in the target content is extracted through the AI ​​big model as the key information matching the preset business keywords.

[0013] In one embodiment, the content elements of the business document content include at least a title and a body content, and matching the business document content with preset business keywords according to the matching priority of the content format corresponding to each of the content elements includes:

[0014] Matching the title with the preset business keywords;

[0015] In the case that the preset business keyword is not matched in the title, the body content is matched with the preset business keyword.

[0016] In one embodiment, extracting target content matching the preset business keyword from the business document content includes:

[0017] When the preset business keyword is matched in the title, extracting the text content belonging to the title;

[0018] The extracted text content belonging to the title is used as the target content.

[0019] In one embodiment, when the preset business keyword is not matched in the title, extracting target content that matches the preset business keyword in the business file content includes:

[0020] When the preset business keyword is matched in the body content, the body content is extracted and the extracted body content is used as the target content.

[0021] In one embodiment, the content elements of the business document content include a title, a body content, and a file text identifier. When neither the title nor the body content matches the preset business keyword, the method further includes:

[0022] Extract key information from the document text identifier using the AI ​​large model;

[0023] The key information in the file text identifier is stored in the business resource question and answer database.

[0024] In a second aspect, the present application also provides a business resource retrieval device based on an AI large model, comprising:

[0025] An acquisition module is configured to acquire a business file and determine the business file content in the business file; the business file content includes at least two content elements with different content formats;

[0026] A matching module, configured to match the business file content with preset business keywords according to the matching priority of the content format corresponding to each content element;

[0027] An extraction module, configured to extract key information from the business document content that matches the preset business keywords, and store the key information in a business resource question-and-answer database;

[0028] An output module is used to respond to an input business resource search text and output a business resource search result that responds to the business resource search text based on the business file content corresponding to the target key information; the target key information is the key information in the business resource question and answer database that matches the business resource search text.

[0029] In a third aspect, the present application further provides a computer device comprising a memory and a processor, wherein the memory stores a computer program, and when the computer program is executed by the processor, the steps of the above method are implemented.

[0030] In a fourth aspect, the present application further provides a computer-readable storage medium having a computer program stored thereon, which implements the steps of the above method when executed by a processor.

[0031] In a fifth aspect, the present application further provides a computer program product, comprising a computer program, which implements the steps of the above method when executed by a processor.

[0032] The above-mentioned business resource retrieval method, device, computer equipment, computer-readable storage medium and computer program product based on the AI ​​big model determine the business file content in the business file by obtaining the business file; the business file content includes at least two content elements with different content formats; the business file content is matched with the preset business keywords according to the matching priority of the content format corresponding to each content element; the key information matching the preset business keywords in the business file content is extracted, and the key information is stored in the business resource question and answer database; in response to the input business resource search text, based on the business file content corresponding to the target key information, the business resource search result for responding to the business resource search text is output; the target key information is the key information matching the business resource search text in the business resource question and answer database.

[0033] In this way, the business document content is matched with the preset business keywords according to the matching priority of the content format corresponding to each content element in the business document content, and the key information of the business document content is extracted through the preset business keywords. The priority setting makes the matching of business keywords more accurate, so that when parsing the key information of the business document content based on the business keywords, the key information related to the preset business keywords in the business document content can be more accurately parsed to store the key information in the business resource question and answer database. When the user enters the business resource search text, the business resource search results related to the search text can be quickly and accurately output based on the key information stored in the database, which effectively improves the accuracy and reliability of the retrieval results of business resources. BRIEF DESCRIPTION OF THE DRAWINGS

[0034] In order to more clearly illustrate the technical solutions in the embodiments of the present application or related technologies, the following briefly introduces the drawings required for use in the embodiments of the present application or related technical descriptions. Obviously, the drawings described below are only some embodiments of the present application. For ordinary technicians in this field, other related drawings can be obtained based on these drawings without paying any creative work.

[0035] Figure 1 1. A flowchart of a business resource retrieval method based on an AI big model in one embodiment;

[0036] Figure 2 Schematic diagram of a process of a business resource retrieval method based on an AI big model in another embodiment;

[0037] Figure 3 1 is a flowchart of another method for retrieving business resources based on an AI big model in one embodiment;

[0038] Figure 4 This is a structural block diagram of a business resource retrieval device based on an AI large model in one embodiment;

[0039] Figure 5 FIG. 1 is a diagram showing the internal structure of a computer device in one embodiment. DETAILED DESCRIPTION

[0040] In order to make the purpose, technical solutions and advantages of this application more clear, the following further describes this application in detail with reference to the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are only used to explain this application and are not intended to limit this application.

[0041] It should be noted that the terms "first," "second," and the like in the specification and claims of the present disclosure and the above-mentioned drawings are used to distinguish similar objects and are not necessarily used to describe a specific order or precedence. It should be understood that the numbers used in this manner are interchangeable where appropriate so that the embodiments of the present disclosure described herein can be implemented in an order other than those illustrated or described herein. The embodiments described in the following exemplary embodiments do not represent all embodiments consistent with the present disclosure. Instead, they are merely examples of apparatus and methods consistent with certain aspects of the present disclosure as detailed in the appended claims.

[0042] In one embodiment, Figure 1 As shown, a business resource retrieval method based on an AI large model is provided. This embodiment uses the method applied to a computer device as an example. It is understandable that the computer device can be a terminal, a server, or a system including a terminal and a server. In this embodiment, the method includes the following steps:

[0043] Step S110: Acquire a business file and determine the business file content in the business file.

[0044] The business document content includes at least two content elements with different content formats.

[0045] Business documents can be documents that record business-related information. For example, business documents can be documents that record and manage internal and external information of an enterprise, including contracts, reports, plans, policies, manuals, standard operating procedures, etc. Business documents can be collected through file upload, API interfaces, etc.

[0046] After obtaining the business file, the computer device can determine the business file content in the business file by parsing the business file.

[0047] The file format of the business file may be determined first, and then the business file may be parsed according to a parsing method corresponding to the file format to determine the business file content in the business file.

[0048] The file format of the business document can be an editable format or a non-editable format. That is, the business document can be an editable file or a non-editable file. In the case where the business document is an editable file, such as an editable text file, the business document content can be directly extracted from the editable file from the computer device; in the case where the business document is a non-editable file, such as a scanned version of the business document (such as a file obtained by scanning a paper document), the business document content in the business document can be parsed through OCR (Optical Character Recognition) recognition technology.

[0049] Step S120 : matching the business file content with preset business keywords according to the matching priority of the content format corresponding to each content element.

[0050] Among them, content elements with different content formats have different matching priorities. For example, a content element may include a title and a body content with different content formats, and the matching priorities of the title and the body content are different.

[0051] The preset business keywords may be keywords associated with the business document.

[0052] In a specific implementation, the computer device may match the business file content with the preset business keywords according to the matching priority of the content format corresponding to each content element.

[0053] Step S130: extract key information matching preset business keywords from the business document content through the AI ​​big model, and store the key information in the business resource question and answer database.

[0054] In a specific implementation, if a computer device matches content in a business file that matches preset business keywords, the key information that matches the preset business keywords in the business file content can be extracted through the AI ​​big model to store the key information in the business resource question and answer database.

[0055] The extracted key information can be vectorized and stored in a business resource question and answer database to achieve vectorized storage of key information in the business document content.

[0056] In this way, for a large number of business files, the computer device can store key information in each business file into the business resource question and answer database according to the above steps S110 to S130.

[0057] Step S140 : In response to the input business resource search text, outputting business resource search results that respond to the business resource search text based on the business file content corresponding to the target key information.

[0058] Business resource search text is a business-related query text entered by the user to find specific information or resources. This text can contain keywords, questions, or topics, and the user hopes to obtain relevant business documents, data, or knowledge through the search.

[0059] The target key information is the key information in the business resource question and answer database that matches the business resource search text.

[0060] In the specific implementation, when the user needs to find a business file, he can input the business resource search text into the question-and-answer interactive interface. The computer device can query the business resource question-and-answer database for key information that matches the business resource search text as the target key information, and output the business resource search results that respond to the business resource search text based on the business file content corresponding to the target key information.

[0061] Here, according to the keywords of the input business resource search text, key information in the business resource question and answer database that matches the keywords of the business resource search text may be determined as the target key information.

[0062] In some embodiments, the business resource question-and-answer database may be a database of a business resource question-and-answer system, which runs on a computer device. A user may enter a business resource search text through a question-and-answer interactive interface provided by the business resource question-and-answer system. The business resource question-and-answer system then searches the business resource question-and-answer database for key information matching the key information based on the keywords in the business resource search text, and uses the key information as target key information. The system then outputs business resource search results based on the business document content corresponding to the target key information.

[0063] In other embodiments, the user can input business resource search text in the interactive interface of the question and answer big model, and the computer device queries the business resource question and answer database for key information that matches the keywords of the business resource search text based on the input business resource search text, and uses the key information as the target key information. According to the business file content corresponding to the target key information and the preset prompt word template, the target question and answer prompt is obtained. The target question and answer prompt is used to instruct the question and answer big model to output business resource search results for responding to the business resource search text by referring to the business file content corresponding to the target key information.

[0064] In this way, users can query business files through dialogue, which improves the flexibility of business resource query and makes query more convenient and quick.

[0065] In the above-mentioned business resource retrieval method based on the AI ​​big model, the business file content in the business file is determined by obtaining the business file; the business file content includes at least two content elements with different content formats; the business file content is matched with the preset business keywords according to the matching priority of the content format corresponding to each content element; the key information matching the preset business keywords in the business file content is extracted, and the key information is stored in the business resource question and answer database; in response to the input business resource search text, based on the business file content corresponding to the target key information, the business resource search result for responding to the business resource search text is output; the target key information is the key information matching the business resource search text in the business resource question and answer database.

[0066] In this way, the business document content is matched with the preset business keywords according to the matching priority of the content format corresponding to each content element in the business document content, and the key information of the business document content is extracted through the preset business keywords. The priority setting makes the matching of business keywords more accurate, so that when parsing the key information of the business document content based on the business keywords, the key information related to the preset business keywords in the business document content can be more accurately parsed to store the key information in the business resource question and answer database. When the user enters the business resource search text, the business resource search results related to the search text can be quickly and accurately output based on the key information stored in the database, which effectively improves the accuracy and reliability of the retrieval results of business resources.

[0067] In one embodiment, key information that matches preset business keywords in the business document content is extracted through the AI ​​big model, including: extracting target content that matches preset business keywords in the business document content; extracting key information in the target content through the AI ​​big model as key information that matches the preset business keywords.

[0068] In a specific implementation, when a computer device extracts key information from a business file that matches preset business keywords, the computer device can extract the content from the business file that matches the preset business keywords as the target content, and then extract the key information from the target content as the key information that matches the preset business keywords.

[0069] Among them, AI big models can be used to extract key information from the target content. The key information may include keywords, summaries, highlights and other elements.

[0070] In actual applications, computer devices can store the mapping relationship between key information and target content, so that in the process of outputting business resource search results for responding to business resource search text based on the business file content corresponding to the target key information, the computer device can determine the target content corresponding to the target key information based on the mapping relationship between the key information and the target content, and thus obtain the business file content corresponding to the target key information based on the target content corresponding to the target key information.

[0071] Among them, the business file content corresponding to the target key information may include a file link to the business file corresponding to the target business key information (the file link is used for users to access the corresponding business file), and may also include the business file content containing the target key information, such as the title containing the target key information and the text content under the title, as well as the text content containing the target key information, and the page numbers and specific locations of these titles and text contents in the corresponding business file.

[0072] The technical solution of this embodiment extracts target content from business documents that matches preset business keywords; then uses the AI ​​big model to extract key information from the target content as key information that matches the preset business keywords. In this way, target content from business documents that matches the business keywords is filtered out based on the business keywords, and then key information from the target content is extracted. This ensures that the extracted key information is strongly correlated with the preset business keywords, thereby enabling more accurate business resource retrieval based on the extracted key information.

[0073] In one embodiment, the content elements of the business document content include at least a title and body content. The business document content is matched with preset business keywords according to the matching priority of the content format corresponding to each content element, including: matching the title with the preset business keywords; if the preset business keywords are not matched in the title, matching the body content with the preset business keywords.

[0074] The content elements of the business document content include at least a title and a body of content.

[0075] The main content may include text, tables, etc.

[0076] Among them, the matching priority of the title is higher than the matching priority of the text content.

[0077] In a specific implementation, when a computer device matches the business file content with preset business keywords according to the matching priority of the content format corresponding to each content element, the computer device can match each title with the preset business keyword. If the preset business keyword is not matched in the title, the body content will be matched with the preset business keyword.

[0078] The process of matching the main text content with the preset business keywords refers to matching the main text content under all titles with the preset business keywords.

[0079] The technical solution of this embodiment is to match the title with the preset business keywords; when the preset business keywords are not matched in the title, the body content is matched with the preset business keywords. In this way, since the title is usually concise and clear and can quickly indicate the subject, the priority matching of the title can quickly filter out the content related to the preset business keywords, thereby improving the matching efficiency. If the title has matched the preset business keywords, further matching processing of the body content can be avoided, thereby saving computing resources and time. In addition, the title can usually reflect the business information of the file more accurately. Prioritizing the matching of the title can improve the relevance and accuracy of the final extraction result. When the business keywords are not matched in the title, the body content is matched. The body content matching can supplement the relevant information that may not be clearly stated in the title, forming a more comprehensive search result.

[0080] In one embodiment, target content that matches preset business keywords in the business file content is extracted, including: when the preset business keywords are matched in the title, the body content belonging to the title is extracted; and the extracted body content belonging to the title is used as the target content.

[0081] In a specific implementation, when a computer device is extracting target content that matches preset business keywords from the content of a business file, if the preset business keywords are matched in the title, the computer device can extract the body content under the title that matches the preset business keywords, and use the extracted body content under the title as the target content that matches the preset business keywords, so that the key information in the target content can be extracted as the key information that matches the preset business keywords.

[0082] In the process of determining the text content under a title that matches the preset business keyword, the computer device can use the title that matches the preset business keyword as the target title. The computer device can determine the next title at the same level as the target title based on the hierarchical structure of the outline directory of the business file, and the text content between the target title and the next title is used as the target content that matches the preset business keyword.

[0083] The technical solution of this embodiment is to extract the text content under the title when the preset business keywords are matched in the title; and use the text content under the title as the target content. In this way, since the title can usually more accurately reflect the business information of the text content under it, when the preset business keywords are matched in the title, the text content under the title that matches the preset business keywords is also content that has a strong correlation with the preset business keywords. The text content under the title that matches the preset business keywords is extracted and used as the target content that matches the preset business keywords for key information extraction, thereby accurately extracting key information that matches the preset business keywords in the business file, which is conducive to improving the accuracy of business resource retrieval results.

[0084] In one embodiment, when the preset business keywords are not matched in the title, the target content that matches the preset business keywords in the business file content is extracted, including: when the preset business keywords are matched in the body content, the body content is extracted and the extracted body content is used as the target content.

[0085] In a specific implementation, when the preset business keywords are not matched in the title, the computer device can match the text content under each title with the preset business keywords. In this way, when the computer device is extracting the target content that matches the preset business keywords in the business file content, the computer device can extract the text content that matches the preset business keywords when the preset business keywords are matched in the text content, and use the extracted text content as the target content that matches the preset business keywords, thereby extracting key information of the target content.

[0086] The technical solution of this embodiment is to extract the main body content when the preset business keywords are not matched in the title, and use the extracted main body content as the target content to extract the key information of the target content. In this way, when the preset business keywords are not matched in the title, since the main body content of the business document can also reflect the business information of the business document, when the preset business keywords are matched in the main body content, the main body content that matches the preset business keywords is extracted, and the extracted main body content is used as the target content that matches the preset business keywords, thereby extracting the key information of the target content, and accurately and comprehensively extracting the key information associated with the preset business keywords in the business document.

[0087] In one embodiment, the content elements of a business file include a title, body content, and a file text identifier. When neither the title nor the body content matches the preset business keywords, the method further includes: extracting key information from the file text identifier through an AI large model; and storing the key information in the file text identifier in a business resource question and answer database.

[0088] The file text identifier may be the file name of the business file.

[0089] In the specific implementation, when the preset business keywords are not matched in the title of the business file and the text content does not match the preset business keywords, since the file text identifier can also contain the business information of the business file, the key information in the file text identifier can be extracted through the AI ​​big model and stored in the business resource question and answer database.

[0090] In actual applications, taking the file text identifier as the file name as an example, the computer equipment can extract key information from the file name, such as keywords, summaries and highlights, through the AI ​​big model, and store it in the business resource question and answer database.

[0091] This step-by-step key information extraction method effectively reduces the omission of key information. Even if the title and body content fail to match the preset business keywords, key information can still be obtained from the file name.

[0092] In the technical solution of this embodiment, the content elements of the business file content also include a file text identifier. When the preset business keywords are not matched in the main text, the key information in the file text identifier is extracted through the AI ​​large model; the key information in the file text identifier is stored in the business resource question and answer database. In this way, when the title and main text of the business file do not match the preset business keywords, since the file text identifier of the business file can also reflect certain business information of the business file, by extracting the key information from the file text identifier and storing the key information in the file text identifier in the business resource question and answer database, the key information can be effectively supplemented to ensure that the key information of the business file is extracted.

[0093] In another embodiment, Figure 2 As shown, a business resource retrieval method based on an AI big model is provided. The method is applied to a computer device as an example, and includes the following steps:

[0094] Step S202: Acquire a business file and determine the business file content in the business file.

[0095] Step S204: Match the title with preset business keywords.

[0096] Step S206: When the preset business keyword is matched in the title, the body content belonging to the title is extracted as the target content.

[0097] Step S208: If the title does not match the preset business keywords, the body content is matched with the preset business keywords.

[0098] Step S210: When the preset business keywords are matched in the main content, the main content is used as the target content.

[0099] Step S212: extract key information from the target content through the AI ​​big model as key information that matches the preset business keywords.

[0100] Step S214: If the preset business keywords are not matched in the title and the body content, the key information in the file text identifier is extracted through the AI ​​large model.

[0101] Step S216: storing the key information in the business resource question and answer database.

[0102] Step S218 : In response to the inputted business resource search text, outputting business resource search results that respond to the business resource search text based on the business file content corresponding to the target key information.

[0103] It should be noted that the specific limitations of the above steps can be found in the specific limitations of a business resource retrieval method based on an AI big model above.

[0104] In one embodiment, Figure 3 As shown, a flowchart of another business resource retrieval method based on the AI ​​big model is provided. Take the business document as an example for explanation. In contract management, faced with a large number of existing contracts and new contracts, you can obtain scanned versions of non-editable contract files (such as contract file images). First, upload the contract files in batches, and then perform OCR parsing one by one. If the parsed file is a compressed package containing contract file images, you need to re-sort the content returned by OCR (after OCR parsing, rearrange the order according to the number in the footer of the page). If it is not a compressed package, there is no need to re-arrange. The content after OCR parsing will be divided into title, body content, contract name, etc. according to the format of the content. There are three different steps to obtain key information for different situations:

[0105] 1. Match the title with the preset business keywords. If a title that matches the business keyword is selected, extract the main content under the title and use the AI ​​model to extract keywords, summaries, and highlights from the extracted content.

[0106] 2. If no business keywords are found in the title, the business keywords are matched with the main content of all titles. If a business keyword is found, the main content is extracted and the extracted content is used by the AI ​​model to extract keywords, summaries, and highlights.

[0107] 3. When the target keywords are not matched in the title and body content, because the contract name also contains the business information of the contract document, the contract name will be used through the AI ​​big model to extract keywords, summaries and highlights.

[0108] After obtaining the key information of the contract document through the above steps, this key information will be stored in the vector database (business resource question and answer database). In this way, users can find the required contract documents through dialogue, making the file search method more diversified and intelligent.

[0109] It should be understood that, although the various steps in the flowcharts involved in the various embodiments described above are displayed in sequence according to the instructions of the arrows, these steps are not necessarily executed in sequence in the order indicated by the arrows. Unless otherwise specified herein, there is no strict order restriction on the execution of these steps, and these steps can be executed in other orders. Moreover, at least a portion of the steps in the flowcharts involved in the various embodiments described above can include multiple steps or multiple stages, and these steps or stages are not necessarily executed and completed at the same time, but can be executed at different times, and the execution order of these steps or stages is not necessarily to be carried out in sequence, but can be executed in turn or alternately with other steps or at least a portion of steps or stages in other steps.

[0110] Based on the same inventive concept, the embodiment of the present application also provides an AI big model-based business resource retrieval device for implementing the above-mentioned AI big model-based business resource retrieval method. The implementation solution provided by the device is similar to the implementation solution described in the above-mentioned method. Therefore, the specific limitations in the embodiments of one or more AI big model-based business resource retrieval devices provided below can be found in the above-mentioned limitations on the AI ​​big model-based business resource retrieval method, and will not be repeated here.

[0111] In an exemplary embodiment, Figure 4 As shown, a business resource retrieval device based on an AI large model is provided, comprising: an acquisition module 410, a matching module 420, an extraction module 430 and an output module 440, wherein:

[0112] The acquisition module 410 is configured to acquire a business file and determine the business file content in the business file; the business file content includes at least two content elements with different content formats.

[0113] The matching module 420 is configured to match the business file content with preset business keywords according to the matching priority of the content format corresponding to each content element.

[0114] The extraction module 430 is used to extract key information matching the preset business keywords from the business document content through the AI ​​big model, and store the key information in the business resource question and answer database.

[0115] The output module 440 is used to respond to the input business resource search text and output the business resource search results for responding to the business resource search text based on the business file content corresponding to the target key information; the target key information is the key information in the business resource question and answer database that matches the business resource search text.

[0116] In one embodiment, the extraction module 430 is specifically used to extract target content from the business document content that matches the preset business keywords; and extract key information from the target content through the AI ​​big model as key information that matches the preset business keywords.

[0117] In one embodiment, the content elements of the business document content include at least a title and a body content, and the matching module 420 is specifically used to match the title with the preset business keywords; if the preset business keywords are not matched in the title, the body content is matched with the preset business keywords.

[0118] In one embodiment, the extraction module 430 is specifically configured to extract the text content under the title when the preset business keyword is matched in the title; and use the extracted text content under the title as the target content.

[0119] In one embodiment, when the preset business keyword is not matched in the title, the extraction module 430 is specifically used to extract the body content when the preset business keyword is matched in the body content, and use the extracted body content as the target content.

[0120] In one embodiment, the content elements of the business file content include a title, body content, and a file text identifier. When the preset business keywords are not matched in the title and the body content, the extraction module 430 is further used to extract key information in the file text identifier through the AI ​​big model; and store the key information in the file text identifier in the business resource question and answer database.

[0121] Each module in the aforementioned AI-based business resource retrieval device can be implemented in whole or in part through software, hardware, or a combination thereof. Each module can be embedded in or independent of a processor in a computer device in hardware form, or can be stored in a computer device's memory in software form, so that the processor can call and execute the corresponding operations of each module.

[0122] In an exemplary embodiment, a computer device is provided. The computer device may be a terminal, and its internal structure diagram may be as shown in FIG. Figure 5 As shown. The computer device includes a processor, memory, an input / output interface, a communication interface, a display unit, and an input device. The processor, memory, and input / output interface are connected via a system bus, and the communication interface, display unit, and input device are connected to the system bus via the input / output interface. The processor of the computer device is used to provide computing and control capabilities. The memory of the computer device includes a non-volatile storage medium and internal memory. The non-volatile storage medium stores an operating system and computer programs. The internal memory provides an environment for the operation of the operating system and computer programs in the non-volatile storage medium. The input / output interface of the computer device is used to exchange information between the processor and external devices. The communication interface of the computer device is used to communicate with external terminals via wired or wireless means, and the wireless means can be implemented via Wi-Fi, mobile cellular networks, near-field communication (NFC), or other technologies. When executed by the processor, the computer program implements a business resource retrieval method based on an AI large model. The display unit of the computer device is used to form a visually visible image, and can be a display screen, a projection device, or a virtual reality imaging device. The display screen can be a liquid crystal display screen or an electronic ink display screen, and the input device of the computer device can be a touch layer covering the display screen, or a button, trackball or touchpad set on the computer device casing, or an external keyboard, touchpad or mouse.

[0123] Those skilled in the art will understand that Figure 5 The structure shown in the figure is only a block diagram of a part of the structure related to the solution of the present application, and does not constitute a limitation on the computer device to which the solution of the present application is applied. The specific computer device may include more or fewer components than shown in the figure, or combine certain components, or have a different component arrangement.

[0124] In one embodiment, a computer device is further provided, including a memory and a processor. The memory stores a computer program, and the processor implements the steps in the above method embodiments when executing the computer program.

[0125] In one embodiment, a computer-readable storage medium is provided, on which a computer program is stored. When the computer program is executed by a processor, the steps in the above-mentioned method embodiments are implemented.

[0126] In one embodiment, a computer program product is provided, including a computer program, which implements the steps in the above method embodiments when executed by a processor.

[0127] It should be noted that the user information (including but not limited to user device information, user personal information, etc.) and data (including but not limited to data used for analysis, stored data, displayed data, etc.) involved in this application are all information and data authorized by the user or fully authorized by all parties, and the collection, use and processing of relevant data must comply with relevant regulations.

[0128] Those skilled in the art will understand that all or part of the processes in the above-mentioned embodiments can be implemented by instructing the relevant hardware through a computer program. The computer program can be stored in a non-volatile computer-readable storage medium. When the computer program is executed, it can include the processes of the embodiments of the above-mentioned methods. In particular, any reference to memory, database, or other media used in the embodiments provided in this application can include at least one of non-volatile memory and volatile memory. Non-volatile memory can include read-only memory (ROM), magnetic tape, floppy disk, flash memory, optical memory, high-density embedded non-volatile memory, resistive random access memory (ReRAM), magnetic random access memory (MRAM), ferroelectric random access memory (FRAM), phase change memory (PCM), graphene memory, etc. Volatile memory can include random access memory (RAM) or external cache memory, etc. By way of illustration and not limitation, RAM can take various forms, such as static random access memory (SRAM) or dynamic random access memory (DRAM). The databases involved in the various embodiments provided herein may include at least one of a relational database and a non-relational database. Non-relational databases may include, but are not limited to, blockchain-based distributed databases. The processors involved in the various embodiments provided herein may be, but are not limited to, general-purpose processors, central processing units (CPUs), graphics processing units (GPUs), digital signal processors (DSPs), programmable logic devices (PLDs), quantum computing-based data processing logic devices, artificial intelligence (AI) processors, and the like.

[0129] The technical features of the above embodiments can be combined arbitrarily. In order to make the description concise, not all possible combinations of the technical features in the above embodiments are described. However, as long as there is no contradiction in the combination of these technical features, they should be considered to be within the scope of this application.

[0130] The above-described embodiments merely represent several implementation methods of the present application. While the descriptions are relatively specific and detailed, they should not be construed as limiting the scope of the present application. It should be noted that a person of ordinary skill in the art may make various modifications and improvements without departing from the spirit of the present application, and these modifications and improvements fall within the scope of protection of the present application. Therefore, the scope of protection of the present application shall be determined by the appended claims.

Claims

1. A business resource retrieval method based on AI big model, characterized in that: The method comprises: Acquire a business file and determine the business file content in the business file; the business file content includes at least two content elements with different content formats; Matching the business file content with preset business keywords according to the matching priority of the content format corresponding to each content element; Extract key information from the business document content that matches the preset business keywords through the AI ​​big model, and store the key information in the business resource question and answer database; In response to an input business resource search text, outputting a business resource search result that responds to the business resource search text based on the business file content corresponding to target key information; the target key information is key information in the business resource question and answer database that matches the business resource search text; Among them, when the content elements include a title, body content and a file text identifier, and the preset business keywords are not matched in the title and the body content, the key information in the file text identifier is extracted through the AI ​​big model; the key information in the file text identifier is stored in the business resource question and answer database.

2. The method according to claim 1, characterized in that The extracting of key information from the business document content that matches the preset business keywords using the AI ​​big model includes: Extracting target content from the business document content that matches the preset business keyword; The key information in the target content is extracted through the AI ​​big model as the key information matching the preset business keywords.

3. The method according to claim 2, characterized in that The matching of the business file content with preset business keywords according to the matching priority of the content format corresponding to each content element includes: Matching the title with the preset business keywords; In the case that the preset business keyword is not matched in the title, the body content is matched with the preset business keyword.

4. The method according to claim 3, characterized in that The extracting target content that matches the preset business keyword from the business file content includes: When the preset business keyword is matched in the title, extracting the text content belonging to the title; The extracted text content belonging to the title is used as the target content.

5. The method according to claim 3, characterized in that In the case that the preset business keyword is not matched in the title, extracting target content that matches the preset business keyword in the business file content includes: When the preset business keyword is matched in the body content, the body content is extracted and the extracted body content is used as the target content.

6. A business resource retrieval device based on AI big model, characterized in that: The device comprises: An acquisition module is configured to acquire a business file and determine the business file content in the business file; the business file content includes at least two content elements with different content formats; A matching module, configured to match the business file content with preset business keywords according to the matching priority of the content format corresponding to each content element; An extraction module is used to extract key information matching the preset business keywords from the business document content through an AI big model, and store the key information in a business resource question and answer database; An output module, configured to respond to an input business resource search text and output a business resource search result corresponding to the business resource search text based on business file content corresponding to target key information; the target key information being key information in the business resource question and answer database that matches the business resource search text; Among them, when the content elements include a title, body content and a file text identifier, and the preset business keywords are not matched in the title and the body content, the extraction module is also used to extract key information in the file text identifier through the AI ​​big model; and store the key information in the file text identifier in the business resource question and answer database.

7. The device according to claim 6, characterized in that The extraction module is specifically used to extract target content from the business document content that matches the preset business keywords; and extract key information from the target content through the AI ​​big model as key information that matches the preset business keywords.

8. A computer device comprising a memory and a processor, wherein the memory stores a computer program, wherein: When the processor executes the computer program, the steps of the method according to any one of claims 1 to 5 are implemented.

9. A computer-readable storage medium having a computer program stored thereon, characterized in that: When the computer program is executed by a processor, the steps of the method according to any one of claims 1 to 5 are implemented.

10. A computer program product comprising a computer program, characterized in that When the computer program is executed by a processor, the steps of the method according to any one of claims 1 to 5 are implemented.

Images