Knowledge retrieval method, device, equipment, medium and product

Abstract

The invention discloses a knowledge retrieval method, device and equipment, a medium and a product, relates to the technical field of data processing, in particular to the technical field of large models, and can be applied to the field of financial science and technology. The method comprises the following steps: acquiring a retrieval problem of a retrieval party; retrieving from a business knowledge base or out-of-row information according to the retrieval problem to obtain a target retrieval result; constructing a target cue word according to the target retrieval result; according to the target cue word, performing knowledge summarization through a retrieval model to obtain a target answer; wherein the retrieval model is obtained based on a large model. Through the technical scheme, the retrieval efficiency and accuracy of professional business knowledge can be improved.

Description

Technical Field

[0001] The present invention relates to the field of data processing technology, in particular to the field of large model technology, and can be applied to the field of financial technology. Specifically, it relates to a knowledge retrieval method, device, equipment, medium and product. Background Art

[0002] In the daily work process, risk line business personnel in the financial industry need to query and review a large amount of risk-related rules and regulations, business knowledge, experience summaries and other data. Existing technical solutions mainly achieve the acquisition of relevant data and content through file search, fuzzy content search, simple search and other methods, and assist in promoting the improvement of quality and efficiency of financial business processing and daily office work.

[0003] Existing technical solutions can meet the basic needs of daily office work, but the cost of acquiring business knowledge is high, especially for novice business personnel or staff who are not familiar with the scope of documents. Based on existing technical solutions, it takes a lot of time to locate documents, search documents, and acquire knowledge. In addition, existing solutions do not support fuzzy search of encrypted documents and cannot support semantic understanding search as required. Overall, the search time is long and the efficiency is low, which affects daily office efficiency. Summary of the invention

[0004] The present invention provides a knowledge retrieval method, device, equipment, medium and product to improve the efficiency of professional business knowledge retrieval.

[0005] According to one aspect of the present invention, a knowledge retrieval method is provided, the method comprising:

[0006] Obtaining the search question of the search party;

[0007] Search from the business knowledge base or external information according to the search question to obtain the target search result;

[0008] Constructing a target prompt word according to the target search result;

[0009] According to the target prompt word, knowledge is summarized through a retrieval model to obtain a target answer; wherein the retrieval model is obtained based on a large model.

[0010] According to another aspect of the present invention, there is provided a knowledge retrieval device, the device comprising:

[0011] A search question acquisition module is used to acquire the search question of the search party;

[0012] A search result determination module is used to search from the business knowledge base or off-line information according to the search question to obtain a target search result;

[0013] A target prompt word construction module, used to construct a target prompt word according to the target search result;

[0014] The target answer determination module is used to summarize knowledge based on the target prompt word through a retrieval model to obtain a target answer; wherein the retrieval model is obtained based on a large model.

[0015] According to another aspect of the present invention, an electronic device is provided, the electronic device comprising:

[0016] at least one processor; and

[0017] a memory communicatively connected to the at least one processor; wherein,

[0018] The memory stores a computer program that can be executed by the at least one processor, and the computer program is executed by the at least one processor so that the at least one processor can execute the knowledge retrieval method described in any embodiment of the present invention.

[0019] According to another aspect of the present invention, a computer-readable storage medium is provided, wherein the computer-readable storage medium stores computer instructions, and the computer instructions are used to enable a processor to implement the knowledge retrieval method described in any embodiment of the present invention when executed.

[0020] According to another aspect of the present invention, a computer program product is provided. The computer program product includes a computer program. When the computer program is executed by a processor, the computer program implements the knowledge retrieval method according to any embodiment of the present invention.

[0021] The technical solution of the embodiment of the present invention obtains the search question of the search party, then searches from the business knowledge base or off-line information according to the search question to obtain the target search result, constructs the target prompt word according to the target search result, and then summarizes the knowledge through the search model according to the target prompt word to obtain the target answer; wherein the search model is obtained based on the large model. The above technical solution can improve the efficiency and accuracy of knowledge retrieval by selecting and judging the search method.

[0022] It should be understood that the contents described in this section are not intended to identify the key or important features of the embodiments of the present invention, nor are they intended to limit the scope of the present invention. Other features of the present invention will become easily understood through the following description. BRIEF DESCRIPTION OF THE DRAWINGS

[0023] In order to more clearly illustrate the technical solutions in the embodiments of the present invention, the following briefly introduces the drawings required for use in the description of the embodiments. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without creative work.

[0024] Figure 1 is a flow chart of a knowledge retrieval method provided according to an embodiment of the present invention;

[0025] Figure 2 is a flow chart of a knowledge retrieval method provided according to an embodiment of the present invention;

[0026] Figure 3 is a schematic diagram of a knowledge retrieval interactive interface provided according to an embodiment of the present invention;

[0027] Figure 4 is a schematic diagram of the structure of a knowledge retrieval device provided according to an embodiment of the present invention;

[0028] Figure 5 It is a schematic diagram of the structure of an electronic device for implementing the knowledge retrieval method according to an embodiment of the present invention. DETAILED DESCRIPTION

[0029] In order to enable those skilled in the art to better understand the scheme of the present invention, the technical scheme in the embodiments of the present invention will be clearly and completely described below in conjunction with the drawings in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work should fall within the scope of protection of the present invention.

[0030] It should be noted that the terms "first", "second", "target", etc. in the specification and claims of the present invention and the above-mentioned drawings are used to distinguish similar objects, and are not necessarily used to describe a specific order or sequence. It should be understood that the data used in this way can be interchanged where appropriate, so that the embodiments of the present invention described herein can be implemented in an order other than those illustrated or described herein. In addition, the terms "including" and "having" and any variations thereof are intended to cover non-exclusive inclusions, for example, a process, method, system, product or device that includes a series of steps or units is not necessarily limited to those steps or units that are clearly listed, but may include other steps or units that are not clearly listed or inherent to these processes, methods, products or devices.

[0031] In addition, it should be noted that in the technical solution of the present invention, the collection, storage, use, processing, transmission, provision and disclosure of relevant data such as pending business documents are in compliance with the provisions of relevant laws and regulations and do not violate public order and good morals.

[0032] Figure 1is a flow chart of a knowledge retrieval method provided according to an embodiment of the present invention. This embodiment can be applied to situations where knowledge retrieval is performed, and is particularly applicable to situations where business knowledge retrieval is performed in the financial industry. The method can be performed by a knowledge retrieval device, which can be implemented in the form of hardware and / or software, and can be configured in an electronic device that carries a knowledge retrieval function, such as a server. Figure 1 As shown, the method includes:

[0033] S110, obtaining a search question of the search party;

[0034] The search party refers to the party that needs to search for business knowledge, such as a novice business person or a staff member who is not familiar with the scope of the document. The so-called search question refers to the relevant question that needs to be searched, such as the relevant question of bank risk business.

[0035] Specifically, the search question of the search party may be obtained through voice dialogue.

[0036] S120: Search from the business knowledge base or external information according to the search question to obtain target search results.

[0037] An optional method is to identify the intention of the search question based on a method judgment model to determine the search method of the search question; the search method includes in-line search or out-of-line search; if the search method is in-line search, then according to the search question, search from the business knowledge base through a vector search method to obtain a first search result, and use the first search result as the target search result; if the search method is out-of-line search, then according to the search question, search from out-of-line information based on a large model to obtain a second search result, and use the second search result as the target search result.

[0038] Among them, the method judgment model can be obtained by training the neural network based on historical retrieval problems and retrieval methods. It should be noted that the method judgment model is a small model, that is, its parameter quantity and network structure are less than those of the large model. The retrieval method includes in-line retrieval and out-of-line retrieval; among them, in-line retrieval refers to retrieval based on in-line professional business knowledge; out-of-line retrieval refers to retrieval based on information on the Internet.

[0039] The so-called business knowledge base refers to the knowledge base of financial industry risk-related knowledge. The so-called first search result refers to the search result obtained based on the business knowledge base, including at least one business data. The so-called second search result refers to the search result obtained from off-line information based on the big model. The so-called target search result refers to the search result used to finally construct the target prompt word.

[0040] Specifically, first input the search question into the mode judgment model for intent recognition, and process it through the mode judgment model to obtain the search mode of the search question. If the search mode is in-line search, extract keywords from the search question, use the keywords as indexes, search from the business knowledge base, obtain at least two candidate search results, and select the search result ranked first, such as top3, from the at least two candidate search results as the first search result; or, vectorize the search question to obtain a question vector, perform matching search from the business knowledge base based on the question vector to obtain at least two candidate search results, select the search result ranked first, such as top3, from the at least two candidate search results as the first search result, and use the first search result as the target search result. If the search mode is out-of-line search, input the search question into the big model, use the big model to obtain the second search result from out-of-line information such as the Internet, and use the second search result as the target search result.

[0041] It is understandable that by using the retrieval method to determine whether it is necessary to access the business knowledge base or to directly search for information outside the industry, it is possible to avoid directly using large model resources while ensuring the accuracy of answers to professional questions.

[0042] S130: construct a target prompt word according to the target search result.

[0043] Specifically, based on the prompt word template, the target search result is processed to obtain the target prompt word.

[0044] S140: Based on the target prompt word, knowledge is summarized through a retrieval model to obtain a target answer.

[0045] Among them, the retrieval model is obtained based on the large model; it should be noted that the large model is a neural network model with a large number of parameters and a complex network structure.

[0046] Specifically, the target prompt word is input into the retrieval model, and the knowledge summary and reasoning are performed through the retrieval model to obtain the target answer. At the same time, the source of the target answer can also be obtained through the retrieval model.

[0047] The technical solution of the embodiment of the present invention obtains the search question of the search party, then searches from the business knowledge base or off-line information according to the search question to obtain the target search result, constructs the target prompt word according to the target search result, and then summarizes the knowledge through the search model according to the target prompt word to obtain the target answer; wherein the search model is obtained based on the large model. The above technical solution can improve the efficiency and accuracy of knowledge retrieval by selecting and judging the search method.

[0048] On the basis of the above embodiment, as an optional method of the present invention, the target answer can also be fed back to the search party through an interactive interface; in response to the search party's evaluation operation on the target answer, evaluation feedback information is obtained.

[0049] Specifically, after obtaining the target answer, the target answer is fed back to the search party through the interactive interface. The interactive interface then provides an evaluation option, and the search party can evaluate the target answer in the interactive interface, such as selecting a star rating, giving thumbs up / down, and selecting an evaluation tag, where the evaluation tags include but are not limited to incorrect or incomplete reference documents and irrelevant answer content, and then obtain evaluation feedback information.

[0050] Furthermore, the large model can be iteratively optimized based on the evaluation feedback information.

[0051] It can be understood that by feeding back the target answer through the interactive interface and obtaining the feedback information of the search party, the feedback data can be refluxed, thereby iteratively optimizing the search capability.

[0052] Figure 2 FIG. 1 is a flow chart of a knowledge retrieval method provided according to an embodiment of the present invention. Based on the above embodiment, this embodiment further describes the process of constructing a business knowledge base. Figure 2 As shown, the method includes:

[0053] S210: Obtain the business document to be processed.

[0054] S220: Decompose the business document to be processed to obtain chapter contents.

[0055] S230, vectorizing the chapter content to obtain chapter vectors, and constructing a business knowledge base based on the chapter vectors.

[0056] Among them, the business documents to be processed refer to documents related to business knowledge that need to be processed, such as risk regulations, management methods and other documents.

[0057] Specifically, the business documents to be processed can be obtained from the bank, and then the business documents to be processed can be broken down according to chapters to obtain chapter content, and then vectorized processing can be used to vectorize the chapter content to obtain chapter vectors, and a business knowledge base can be constructed based on the chapter vectors.

[0058] S240: Obtain the search question of the search party.

[0059] S250: Search from the business knowledge base or external information according to the search question to obtain target search results.

[0060] S260: construct a target prompt word according to the target search result.

[0061] S270: Based on the target prompt word, knowledge is summarized through a retrieval model to obtain a target answer.

[0062] The technical solution provided by the embodiment of the present invention obtains the business document to be processed, then disassembles the business document to be processed to obtain the chapter content, and then vectorizes the chapter content to obtain the chapter vector, and constructs a business knowledge base based on the chapter vector, and finally obtains the search question of the search party, searches from the business knowledge base or off-line information according to the search question to obtain the target search result, constructs the target prompt word according to the target search result, and then summarizes the knowledge through the search model according to the target prompt word to obtain the target answer; wherein the search model is obtained based on the large model. The above technical solution improves the search efficiency by constructing a business knowledge base so that the business professional knowledge can be quickly retrieved by the search party.

[0063] Based on the above embodiment, as an optional method of the present invention, before the business document to be processed is disassembled to obtain the chapter content, it also includes: if the business document to be processed is an encrypted document, the business document to be processed is decrypted. Accordingly, the decrypted business document to be processed is disassembled to obtain the chapter content, the chapter content is vectorized to obtain the chapter vector, and the business knowledge base is constructed according to the chapter vector.

[0064] Specifically, if the business document to be processed is an encrypted document, determine whether the encrypted document can be decrypted. If it can be decrypted, decrypt the business document to be processed, disassemble the decrypted business document to be processed to obtain the chapter content, vectorize the chapter content to obtain the chapter vector, and build a business knowledge base based on the chapter vector.

[0065] If the decrypted document cannot be decrypted, the encrypted document is directly stored in the backup library.

[0066] It is understandable that encrypted documents can be processed so that they can also be searched, thereby improving the accuracy of search content.

[0067] Based on the above embodiment, as another optional method of the present invention, it also includes: associating the business document to be processed with the accessible organization to obtain the authority and access control list; mapping the authority and access control list to the business knowledge base.

[0068] Specifically, access to specific data is restricted by setting permissions and access control lists (ACLs). The ID of the business document to be processed and the organization number of the accessible organization that can access the document are stored in the ACL table, which is mapped to the vector database, i.e., the business knowledge base.

[0069] It is understandable that by setting access permissions for business documents to be processed, when the search party conducts knowledge search, by obtaining the search party's institution number, it is possible to determine the scope of documents that the search party can access in the business knowledge base, thereby improving search efficiency.

[0070] A specific example is that in the process of knowledge retrieval, based on engineering capabilities, query efficiency and accuracy are improved. First, if the TOP3 score of the search recall from the knowledge base is greater than 0.9, the search results are merged with the prompt words, the large model is called to generate the results, and the reference document sources are listed. Second, if the TOP3 score of the search recall is less than 0.9, it means that no relevant knowledge has been retrieved, and the default refusal to answer is returned (such as sorry that no relevant content was found). Figure 3 Example shown.

[0071] The present invention combines the generation capabilities of the knowledge base and the large model, and utilizes the semantic analysis and search generation capabilities of the large model to support front-line business personnel in quickly retrieving risk business knowledge, thereby improving the efficiency of business risk knowledge retrieval and solving pain points in daily office processes, such as the difficulty in retrieving risk knowledge documents, the rapid updating of professional knowledge, and the difficulty in answering risk professional knowledge questions.

[0072] By building a risk-specific knowledge base from massive documents such as risk regulations, management methods, etc., we can solve the problems of insufficient professional knowledge in large models and low model interpretability. At the same time, we can use database isolation capabilities to achieve fine-grained permission isolation for knowledge data, obtain retrieval data according to the principle of minimum authorization, and ensure the security of data access.

[0073] The present invention solves the pain points of traditional knowledge retrieval methods, such as the difficulty and time-consuming search of massive knowledge documents, the inability to search encrypted documents, and inaccurate search content, by utilizing the technology that combines large models with knowledge bases. Through a new conversational interactive retrieval method, the search and query efficiency of knowledge data of risk business line personnel is continuously improved, which is estimated to be improved by about 20%, thus assisting in improving the quality and efficiency of daily office work.

[0074] Figure 4 1 is a schematic diagram of the structure of a knowledge retrieval device provided according to an embodiment of the present invention. This embodiment can be applied to situations where knowledge retrieval is performed, and is particularly applicable to situations where business knowledge retrieval is performed in the financial industry. The device can be implemented in the form of hardware and / or software, and can be configured in an electronic device that carries a knowledge retrieval function, such as a server. Figure 4 As shown, the device comprises:

[0075] A search question acquisition module 310 is used to acquire the search question of the search party;

[0076] A search result determination module 320 is used to search from the business knowledge base or external information according to the search question to obtain a target search result;

[0077] A target prompt word construction module 330 is used to construct a target prompt word according to the target search result;

[0078] The target answer determination module 340 is used to perform knowledge summary based on the target prompt word through a retrieval model to obtain a target answer; wherein the retrieval model is obtained based on a large model.

[0079] The technical solution of the embodiment of the present invention obtains the search question of the search party, then searches from the business knowledge base or off-line information according to the search question to obtain the target search result, constructs the target prompt word according to the target search result, and then summarizes the knowledge through the search model according to the target prompt word to obtain the target answer; wherein the search model is obtained based on the large model. The above technical solution can improve the efficiency and accuracy of knowledge retrieval by selecting and judging the search method.

[0080] Optionally, the search result determination module 320 is specifically used for:

[0081] Based on the method judgment model, the intention of the search question is identified to determine the search method of the search question; the search method includes in-line search or out-of-line search;

[0082] If the search method is an in-line search, searching the business knowledge base by a vector search method according to the search question to obtain a first search result, and using the first search result as a target search result;

[0083] If the search method is an out-of-row search, then according to the search question, a search is performed from out-of-row information based on a large model to obtain a second search result, and the second search result is used as a target search result.

[0084] Optionally, the device further comprises a knowledge base building module, which is used to:

[0085] Get business documents to be processed;

[0086] Disassemble the business documents to be processed to obtain chapter contents;

[0087] The chapter content is vectorized to obtain chapter vectors, and a business knowledge base is constructed based on the chapter vectors.

[0088] Optionally, the knowledge base building module is also used to:

[0089] The business document to be processed is disassembled, and before obtaining the chapter content, if the business document to be processed is an encrypted document, the business document to be processed is decrypted.

[0090] Optionally, the device further includes a permission setting module, which is used to:

[0091] Associate the business documents to be processed with the accessible organizations to obtain the permissions and access control lists;

[0092] Map permissions and access control lists to the business knowledge base.

[0093] Optionally, the device further comprises a feedback module, for:

[0094] Feedback the target answer to the search party through an interactive interface;

[0095] In response to the search party's evaluation operation on the target answer, evaluation feedback information is obtained.

[0096] The knowledge retrieval device provided in the embodiment of the present invention can execute the knowledge retrieval method provided in any embodiment of the present invention, and has the corresponding functional modules and beneficial effects of the execution method.

[0097] According to an embodiment of the present invention, the present invention also provides an electronic device, a readable storage medium and a computer program product.

[0098] Figure 5 It is a schematic diagram of the structure of an electronic device for implementing the knowledge retrieval method according to an embodiment of the present invention. Figure 5 A schematic diagram of the structure of an electronic device 10 that can be used to implement an embodiment of the present invention 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 (such as helmets, glasses, watches, etc.) 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 invention described and / or required herein.

[0099] like Figure 5As shown, the electronic device 10 includes at least one processor 11, and a memory connected to the at least one processor 11, such as a read-only memory (ROM) 12, a random access memory (RAM) 13, etc., wherein the memory stores a computer program that can be executed by at least one processor, and the processor 11 can perform various appropriate actions and processes according to the computer program stored in the read-only memory (ROM) 12 or the computer program loaded from the storage unit 18 to the random access memory (RAM) 13. In the RAM 13, various programs and data required for the operation of the electronic device 10 can also be stored. The processor 11, the ROM 12, and the RAM 13 are connected to each other through a bus 14. An input / output (I / O) interface 15 is also connected to the bus 14.

[0100] A number of components in the electronic device 10 are connected to the I / O interface 15, including: an input unit 16, such as a keyboard, a mouse, etc.; an output unit 17, such as various types of displays, speakers, etc.; a storage unit 18, such as a disk, an optical disk, etc.; and a communication unit 19, such as a network card, a modem, a wireless communication transceiver, etc. The communication unit 19 allows the electronic device 10 to exchange information / data with other devices through a computer network such as the Internet and / or various telecommunication networks.

[0101] The processor 11 may be a variety of general and / or special processing components with processing and computing capabilities. Some examples of the processor 11 include, but are not limited to, a central processing unit (CPU), a graphics processing unit (GPU), various special artificial intelligence (AI) computing chips, various processors running machine learning model algorithms, a digital signal processor (DSP), and any appropriate processor, controller, microcontroller, etc. The processor 11 executes the various methods and processes described above, such as the knowledge retrieval method.

[0102] In some embodiments, the knowledge retrieval method may be implemented as a computer program, which is tangibly contained in a computer-readable storage medium, such as a storage unit 18. In some embodiments, part or all of the computer program may be loaded and / or installed on the electronic device 10 via the ROM 12 and / or the communication unit 19. When the computer program is loaded into the RAM 13 and executed by the processor 11, one or more steps of the knowledge retrieval method described above may be performed. Alternatively, in other embodiments, the processor 11 may be configured to perform the knowledge retrieval method in any other appropriate manner (e.g., by means of firmware).

[0103] 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), load 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.

[0104] Computer programs for implementing the methods of the present invention may be written in any combination of one or more programming languages. These computer programs may be provided to a processor of a general-purpose computer, a special-purpose computer, or other programmable data processing device, so that when the computer program is executed by the processor, the functions / operations specified in the flow chart and / or block diagram are implemented. The computer program 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.

[0105] In the context of the present invention, a computer-readable storage medium may be a tangible medium that may contain or store a computer program for use by or in combination with an instruction execution system, device or equipment. A computer-readable storage medium may include, but is not limited to, electronic, magnetic, optical, electromagnetic, infrared, or semiconductor systems, devices or equipment, or any suitable combination of the foregoing. Alternatively, a computer-readable storage medium may be a machine-readable signal medium. 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.

[0106] To provide interaction with a user, the systems and techniques described herein may be implemented on an electronic device 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 a pointing device (e.g., a mouse or trackball) through which the user can provide input to the electronic device. Other types of devices may also be used to provide interaction with the user; for example, the feedback provided to the user may be any form of sensory feedback (e.g., visual feedback, auditory feedback, or tactile feedback); and input from the user may be received in any form (including acoustic input, voice input, or tactile input).

[0107] The systems and techniques described herein may be implemented in a computing system that includes backend 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 frontend 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 backend components, middleware components, or frontend 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), a blockchain network, and the Internet.

[0108] A computing system may include a client and a server. The client and the server are generally remote from each other and usually interact through a communication network. The client and server relationship is generated by computer programs running on the corresponding computers and having a client-server relationship with each other. The server may be a cloud server, also known as a cloud computing server or cloud host, which is a host product in the cloud computing service system to solve the defects of difficult management and weak business scalability in traditional physical hosts and VPS services.

[0109] 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 described in the present invention can be executed in parallel, sequentially or in different orders, as long as the desired results of the technical solution of the present invention can be achieved, and this document does not limit this.

[0110] The above specific implementations do not constitute a limitation on the protection scope of the present invention. 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 invention should be included in the protection scope of the present invention.

Claims

1. A knowledge retrieval method, characterized in that: include: Obtaining the search question of the search party; Search from the business knowledge base or external information according to the search question to obtain the target search result; Constructing a target prompt word according to the target search result; According to the target prompt word, knowledge is summarized through a retrieval model to obtain a target answer; wherein the retrieval model is obtained based on a large model.

2. The method according to claim 1, characterized in that: Search the business knowledge base or external information according to the search question to obtain target search results, including: Based on the method judgment model, the intention of the search question is identified, and the search method of the search question is determined; the search method includes in-line search or out-of-line search; If the search method is an in-line search, searching the business knowledge base by a vector search method according to the search question to obtain a first search result, and using the first search result as a target search result; If the search method is an out-of-row search, then according to the search question, a search is performed from out-of-row information based on a large model to obtain a second search result, and the second search result is used as a target search result.

3. The method according to claim 1, characterized in that The method further comprises: Get business documents to be processed; Decomposing the business document to be processed to obtain chapter contents; The chapter content is vectorized to obtain a chapter vector, and a business knowledge base is constructed according to the chapter vector.

4. The method according to claim 3, characterized in that Before the business document to be processed is disassembled to obtain chapter contents, the following steps are also included: If the business document to be processed is an encrypted document, the business document to be processed is decrypted.

5. The method according to claim 3, characterized in that: The method further comprises: Associate the business documents to be processed with the accessible organizations to obtain the permissions and access control lists; The permissions and access control lists are mapped to a business knowledge base.

6. The method according to claim 1, characterized in that The method further comprises: Feedback the target answer to the search party through an interactive interface; In response to the search party's evaluation operation on the target answer, evaluation feedback information is obtained.

7. A knowledge retrieval device, characterized in that: include: A search question acquisition module is used to acquire the search question of the search party; A search result determination module is used to search from the business knowledge base or off-line information according to the search question to obtain a target search result; A target prompt word construction module, used to construct a target prompt word according to the target search result; The target answer determination module is used to summarize knowledge based on the target prompt word through a retrieval model to obtain a target answer; wherein the retrieval model is obtained based on a large model.

8. An electronic device, characterized in that: The electronic device comprises: at least one processor; and a memory communicatively connected to the at least one processor; wherein, The memory stores a computer program executable by the at least one processor, and the computer program is executed by the at least one processor so that the at least one processor can execute the knowledge retrieval method according to any one of claims 1 to 6.

9. A computer-readable storage medium, characterized in that: The computer-readable storage medium stores computer instructions, and the computer instructions are used to enable a processor to implement the knowledge retrieval method according to any one of claims 1 to 6 when executed.

10. A computer program product, characterized in that The computer program product comprises a computer program, and when the computer program is executed by a processor, the computer program implements the knowledge retrieval method according to any one of claims 1 to 6.

Images