Auxiliary method and device for system use, equipment, medium and program product

Abstract

The invention relates to the technical field of artificial intelligence, and discloses an auxiliary method and device for system use, equipment, a medium and a program product. The method comprises the steps of obtaining a current system position and a current user problem when it is detected that a triggering condition of system use assistance is met, and obtaining a current problem vector corresponding to the current user problem; according to the current system position and the current question vector, if at least one matched document entry is found in a vector database, inputting each matched document entry into a pre-trained target question and answer model, and obtaining first answer information output by the target question and answer model; and displaying the first answer information to the user. According to the scheme of the embodiment, accurate knowledge is searched according to the vector database, then the pre-trained large language model is used for question answering, an intelligent question answering mode can be adopted to assist a user in system use, the user can be effectively assisted in service system use, and user experience can be improved.

Description

Technical Field

[0001] The present invention relates to the field of artificial intelligence technology, and in particular to an auxiliary method, device, equipment, medium and program product used by a system. Background Art

[0002] There are many internal systems in banks, involving many professional terms and non-intuitive operation steps. Bank staff will encounter many stuck points and bottlenecks when using the system, and cannot smoothly complete the operation steps at one time. Some staff are involved in operating multiple systems at the same time during their work, which further exacerbates the problem.

[0003] At present, the auxiliary method used by existing systems is mainly to provide system operation manuals for users to read. However, the operation manuals are prone to incomplete coverage. Users cannot get answers to their questions without any guidance. Moreover, the operation manuals require a long time to read and digest, which increases the learning and usage costs of users. As a result, it is impossible to effectively assist users in using the business system. Summary of the invention

[0004] The present invention provides a method, device, equipment, medium and program product for assisting users in using the system, which can assist users in using the system in the form of intelligent questions and answers, and can effectively assist users in using the business system.

[0005] According to one aspect of the present invention, there is provided an auxiliary method used by a system, comprising:

[0006] When it is detected that the triggering condition for system usage assistance is met, the current system position and the current user question are obtained, and a current question vector corresponding to the current user question is obtained;

[0007] According to the current system position and the current question vector, if at least one matching document entry is found in the vector database, each matching document entry is input into a pre-trained target question-answering model, and first answer information output by the target question-answering model is obtained;

[0008] The first answer information is displayed to the user.

[0009] According to another aspect of the present invention, there is provided an auxiliary device for use with a system, comprising:

[0010] A question vector acquisition module, used to acquire the current system position and the current user question when it is detected that the trigger condition for system use assistance is met, and to acquire the current question vector corresponding to the current user question;

[0011] A first answer information acquisition module is used to input each matching document entry into a pre-trained target question-answering model according to the current system position and the current question vector, and obtain first answer information output by the target question-answering model if at least one matching document entry is found in the vector database;

[0012] The answer information display module is used to display the first answer information to the user.

[0013] According to another aspect of the present invention, an electronic device is provided, the electronic device comprising:

[0014] at least one processor; and

[0015] a memory communicatively connected to the at least one processor; wherein,

[0016] 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 to enable the at least one processor to execute the auxiliary method used by the system described in any embodiment of the present invention.

[0017] According to another aspect of the present invention, a computer-readable storage medium is provided, wherein the computer-readable storage medium stores a computer program, wherein the computer program is used to enable a processor to implement the auxiliary method used by the system described in any embodiment of the present invention when executed.

[0018] According to another aspect of the present invention, a computer program product is provided, comprising a computer program, wherein when the computer program is executed by a processor, the computer program implements the auxiliary method used by the system according to any embodiment of the present invention.

[0019] The technical solution of the embodiment of the present invention is as follows: when it is detected that the trigger conditions for system use assistance are met, the current system position and the current user question are obtained, and the current question vector corresponding to the current user question is obtained; based on the current system position and the current question vector, if at least one matching document entry is found in the vector database, each matching document entry is input into the pre-trained target question and answer model, and the first answer information output by the target question and answer model is obtained; the first answer information is displayed to the user; by first searching for precise knowledge based on the vector database, and then using the pre-trained large language model to answer questions, intelligent question and answer can be used to assist users in using the system, which can effectively assist users in using the business system and improve user experience.

[0020] 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

[0021] 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.

[0022] Figure 1 is a flow chart of an auxiliary method used by a system provided according to Embodiment 1 of the present invention;

[0023] Figure 2 is a flow chart of another auxiliary method used by a system provided according to Embodiment 1 of the present invention;

[0024] Figure 3 is a schematic diagram of the structure of an auxiliary device used in a system provided according to Embodiment 2 of the present invention;

[0025] Figure 4 It is a schematic diagram of the structure of an electronic device for implementing the auxiliary method used by the system of an embodiment of the present invention. DETAILED DESCRIPTION

[0026] 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.

[0027] 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 clearly listed, but may include other steps or units that are not clearly listed or inherent to these processes, methods, products or devices.

[0028] Embodiment 1

[0029] Figure 1A flowchart of a system use auxiliary method is provided for the first embodiment of the present invention. This embodiment is applicable to answering questions about the use of a user's system when the user uses a banking system. The method can be executed by an auxiliary device used by the system. The auxiliary device used by the system can be implemented in the form of hardware and / or software. Typically, the auxiliary device used by the system can be configured in an electronic device, such as a computer device or a server. Figure 1 As shown, the method includes:

[0030] S110. When it is detected that a trigger condition for system usage assistance is met, the current system position and the current user question are obtained, and a current question vector corresponding to the current user question is obtained.

[0031] Among them, the triggering condition for system usage assistance can be pre-set condition information that needs to be met to enable the system usage assistance function, for example, the user clicks the button corresponding to the system usage assistance function, the mouse pointer stays at a certain page position for a long time, etc.

[0032] In this embodiment, once the triggering condition for system use assistance is detected, the current system position can be obtained according to the position of the mouse pointer or the current stage of the system operation; at the same time, the current user question can be obtained according to the user's question selection operation or question input operation. The user question can be a question encountered by the user when using the bank system, such as what is the meaning of a noun, what should be done next, etc.

[0033] Optionally, detecting that a trigger condition for using the system assistance is met may include:

[0034] If it is detected that the user's stay time at the current input position reaches the set time, or the user has not made a selection in the first selection box on the current page and the stay time reaches the set time, or the mouse pointer stays at a content location for a set time, then it is determined that the trigger conditions for the system to use assistance are met.

[0035] In a specific example, the trigger condition for the system to use assistive functions can be set as the user staying at a certain input position for 8 seconds, or staying at the first selection box of the current page for 8 seconds without making a selection, or the mouse pointer staying at a position with content for 8 seconds. If any of the above conditions is detected, the system can be triggered to use assistive functions.

[0036] The advantage of the above settings is that the triggering conditions for the system to use auxiliary functions can be flexibly set, which can improve the flexibility of the system in using auxiliary functions.

[0037] Optionally, get the current user question, which can include:

[0038] Display candidate questions to the user. If a user selection operation on a candidate question is detected, obtain the target candidate question selected by the user as the current user question.

[0039] If no selection operation of the user on the candidate question is detected, then when a question input operation of the user is detected, the current user question input by the user is obtained.

[0040] In this embodiment, when it is detected that the system usage auxiliary function is successfully enabled, candidate questions can be displayed to the user in the form of pop-up windows or the like. Among them, the same candidate questions can be set for different system locations; different candidate questions can also be set for different system locations. If the user successfully finds the current system usage problem encountered in the candidate questions, the user can click to select the target candidate question. The processor can obtain the target candidate question as the current user question based on the user's click selection operation. If the user does not make the corresponding selection operation, the question can be input to feedback the natural language description of the current problem to the system. At this time, the processor can obtain the current user question entered by the user. Among them, the questions entered by the user can also be used by subsequent product managers to supplement system product documentation.

[0041] The advantage of the above setting is that it can expand the way of obtaining user questions, avoid the situation where users cannot ask questions, and improve the user experience.

[0042] S120. Based on the current system position and the current question vector, if at least one matching document entry is found in the vector database, each matching document entry is input into a pre-trained target question-answering model, and the first answer information output by the target question-answering model is obtained.

[0043] The vector database may be composed of a series of document entries, each of which may include a document identifier, a question or title vector, an answer or content vector, location information, etc. In this embodiment, the system product documents may be pre-processed with text extraction and vectorization to establish a vector database.

[0044] Specifically, among all the document entries in the vector database, document entries that match the current system position and the current question vector can be retrieved as matching document entries. Matching document entries are those that can accurately answer user questions. Among them, if it is detected that the question or title vector of a certain document entry matches the current question vector, and at the same time, the location information of the document entry is the same as the current system position, then the document entry can be determined as a matching document entry. In this embodiment, a similarity calculation method such as Euclidean distance and cosine similarity can be used to determine whether the question or title vector of the document entry matches the current question vector. For example, if it is detected that the similarity between two vectors is greater than or equal to a preset similarity threshold, then it can be determined that the two vectors match.

[0045] Secondly, if multiple matching document entries are successfully retrieved in the vector database, the recalled matching document entries can be input into the pre-trained target question-answering model, and the answers or contents in the matching document entries can be summarized through the target question-answering model to obtain answer information that conforms to the answer paradigm. Specifically, when processing the matching document entries through the target question-answering model, if the number of words in the answer or content in the matching document entry is less than 100, the answer or content can be directly output as the answer information; if the number of words is large and the number of matching document entries is large, all answers or contents can be summarized through the target question-answering model to obtain the main overview as the final answer information. It is worth noting that the answer information obtained by both methods requires the addition of the original text link corresponding to the answer or content, as well as the reference link in the text, etc.

[0046] The process of the target question-answering model outputting answer information involves prompt word engineering, and it is necessary to clarify the answer paradigm and instructions of the target question-answering model. For a conversational large language model (LLM), using prompts can trigger the model to generate specific types of outputs, such as constraining the answer format of the large language model.

[0047] Optionally, when establishing a target question-answering model, first, a basic large language model can be established, and documents and corresponding overviews can be obtained as training samples; then, the training samples can be used to train the basic large language model to obtain a trained target question-answering model.

[0048] S130: Display the first answer information to the user.

[0049] Specifically, the current answer information may be rendered and displayed in the answer information display area to complete the answer to the user's question. Alternatively, the current answer information may be displayed to the user in a pop-up window or the like.

[0050] The technical solution of the embodiment of the present invention is as follows: when it is detected that the trigger conditions for system use assistance are met, the current system position and the current user question are obtained, and the current question vector corresponding to the current user question is obtained; based on the current system position and the current question vector, if at least one matching document entry is found in the vector database, each matching document entry is input into the pre-trained target question and answer model, and the first answer information output by the target question and answer model is obtained; the first answer information is displayed to the user; by first searching for precise knowledge based on the vector database, and then using the pre-trained large language model to answer questions, intelligent question and answer can be used to assist users in using the system, which can effectively assist users in using the business system and improve user experience.

[0051] In another optional implementation of this embodiment, before searching for at least one matching document entry in the vector database according to the current system position and the current question vector, the following may also be included:

[0052] Obtaining system product documents, and performing text extraction on the system product documents to obtain text corpus and corresponding location information;

[0053] Obtaining a text vector corresponding to the text corpus, and generating a mapping relationship between the text vector and the position information as a document entry, and establishing a vector database based on each document entry;

[0054] The text corpus includes question-answer format and / or title-content format.

[0055] In this embodiment, a knowledge base can be established in advance, and the knowledge base can be vectorized to generate a vector database. Among them, the knowledge base needs to have the following characteristics: 1. Organize text corpus in the form of Q (question) and A (answer); 2. If the text corpus is not in the form of QA, there should be paragraph titles and paragraph content; 3. For corpus that is not in the first or second form, the title corresponding to the main content of the paragraph can be generated by a large language model to form the second form; 4. The current QA or title-content context position should be indicated, for example, it is a "global" QA, or a QA in "page 1->page 2->configuration item 3", which is used to help the system identify the location of the prompt at the time, narrow the scope of the question, and improve the quality of the question and the accuracy of the question.

[0056] Specifically, after obtaining the text corpus and the corresponding location information, the specified Embedding algorithm can be used to vectorize the question or title, and vectorize the answer or content, so as to generate the mapping relationship between the document identifier, the question or title vector, the answer or content vector, and the location information. Each mapping relationship can be used as a document entry, thereby establishing a vector database.

[0057] Among them, Embedding can be a vectorized representation of natural language, which can be at the word granularity, word granularity, sentence granularity, or paragraph granularity. The vector database is a special database for storing Embedding vectors, which can be retrieved by similarity instead of full-text retrieval by search engines or string matching by traditional databases.

[0058] The advantage of the above setting is that by establishing a vector database, accurate question-related knowledge can be acquired in advance, the "hallucination" problem of large language models can be avoided, and the accuracy of answer information can be improved.

[0059] Optionally, searching at least one matching document entry in a vector database according to the current system position and the current question vector may include:

[0060] Performing a match detection between the current system position and the position information of each document entry in the vector database, acquiring target position information matching the current system position, and determining the document entry corresponding to the target position information as a candidate document entry;

[0061] The current question vector is matched with the text vectors of each of the candidate document entries to obtain a target text vector that matches the current question vector, and the candidate document entry corresponding to the target text vector is determined as a matching document entry.

[0062] Specifically, when searching for matching document entries in a vector database, the position information can be matched first, that is, first determine whether the position information of each document entry is the same as the current system position; if it is detected that the position information of the current document entry is the same as the current system position, the current position information can be determined as the target position information, and the current document entry can be determined as a candidate document entry. Then, the similarity between the document vector of each candidate document entry and the current question vector can be calculated respectively. If it is detected that the similarity between the document vector of the current candidate document entry and the current question vector is greater than or equal to a preset similarity threshold, the current document vector can be determined as the target text vector, and the current candidate document entry can be determined as a matching document entry.

[0063] The advantage of the above setting is that it can improve the recall accuracy of matching document entries and improve the efficiency of answering questions.

[0064] In another optional implementation of this embodiment, after obtaining the current question vector corresponding to the current user question, the following may also be included:

[0065] According to the current system position and the current question vector, if no matching document entry is found in the vector database, the current system position and the current question vector are input into a pre-trained fine-tuned question-answering model, and second answer information output by the fine-tuned question-answering model is obtained;

[0066] Displaying the second answer information to the user;

[0067] The fine-tuning question-answering model is obtained by fine-tuning the basic large language model using a vector database.

[0068] In this embodiment, the vector database can be used to fine-tune the basic large language model on the basis of the basic large language model, so as to obtain a large language model suitable for the knowledge of the banking system domain, that is, a fine-tuned question-answering model. Among them, the fine-tuned question-answering model is mainly used for generalized answers, while the target question-answering model is mainly used for accurate answers. When the number of matching document entries recalled in the vector database is 0 according to the current system position and the current question vector, the current system position and the current question vector can be input into the pre-trained fine-tuned question-answering model, and the current user question is answered by the fine-tuned question-answering model to obtain the second answer information.

[0069] It should be noted that when the fine-tuned question-answering model cannot provide answer information, no prompts or answers may be given to the user, but the user may be asked to provide feedback on whether additional document content is needed to supplement the knowledge base.

[0070] The advantage of the above settings is that it can increase the probability of successfully answering user questions and improve the user's system usage experience.

[0071] In a specific implementation of this embodiment, the process of the auxiliary method used by the system can be as follows: Figure 2 As shown. First, the product manager uploads system product documents such as operation manuals, and performs format unification and text extraction on the system product documents to obtain document corpora in the form of questions-answers or titles-contents and corresponding location information. Then, the specified Embedding algorithm is used to obtain the text vector corresponding to the text corpus, and the mapping relationship between the document identifier, text vector and location information is generated, and then a vector database is established based on each mapping relationship.

[0072] Furthermore, when the trigger conditions for system usage assistance are detected, the current system position can be identified, and the current user question selected or input by the user can be obtained; then, according to the current system position and the current question vector, matching document entries can be retrieved from the vector database, for example, Q1 and Title 1 with high content relevance can be retrieved. Among them, the output format of the question-answering model can be set through the Prompt prompt, and the link information corresponding to the matching document entry can be extracted. Finally, the answer information corresponding to the current user question can be given through the target question-answering model LLM1 or the fine-tuned question-answering model LLM2.

[0073] In this embodiment, through the "assisted driving" system based on natural language processing, it is possible to actively identify user questions and provide answers to questions without the user asking them, thereby improving the system usage experience; moreover, the user experience can be improved without modifying the original system.

[0074] Embodiment 2

[0075] Figure 3 This is a schematic diagram of the structure of an auxiliary device used in a system provided in Embodiment 2 of the present invention. Figure 3 As shown, the device includes: a question vector acquisition module 210, a first answer information acquisition module 220 and an answer information display module 230; wherein,

[0076] The question vector acquisition module 210 is used to acquire the current system position and the current user question when it is detected that the trigger condition of the system use assistance is met, and acquire the current question vector corresponding to the current user question;

[0077] A first answer information acquisition module 220 is used to input each matching document entry into a pre-trained target question-answering model according to the current system position and the current question vector, if at least one matching document entry is found in the vector database, and obtain the first answer information output by the target question-answering model;

[0078] The answer information display module 230 is used to display the first answer information to the user.

[0079] The technical solution of the embodiment of the present invention is as follows: when it is detected that the trigger conditions for system use assistance are met, the current system position and the current user question are obtained, and the current question vector corresponding to the current user question is obtained; based on the current system position and the current question vector, if at least one matching document entry is found in the vector database, each matching document entry is input into the pre-trained target question and answer model, and the first answer information output by the target question and answer model is obtained; the first answer information is displayed to the user; by first searching for precise knowledge based on the vector database, and then using the pre-trained large language model to answer questions, intelligent question and answer can be used to assist users in using the system, which can effectively assist users in using the business system and improve user experience.

[0080] Optionally, the question vector acquisition module 210 is specifically used to determine that the trigger conditions for the system to use assistance are met if it is detected that the user's stay time at the current input position reaches a set time, or the user has not made a selection in the first selection box of the current page and the stay time reaches a set time, or the mouse pointer stays at a content position for a set time.

[0081] Optionally, the auxiliary devices used by the system also include:

[0082] A text extraction module is used to obtain system product documents and perform text extraction on the system product documents to obtain text corpus and corresponding location information;

[0083] A vector database establishment module is used to obtain the text vector corresponding to the text corpus, generate a mapping relationship between the text vector and the position information as a document entry, and establish a vector database according to each document entry;

[0084] The text corpus includes question-answer format and / or title-content format.

[0085] Optionally, the first answer information acquisition module 220 is specifically used to match the current system position with the position information of each document entry in the vector database, obtain target position information matching the current system position, and determine the document entry corresponding to the target position information as a candidate document entry;

[0086] The current question vector is matched with the text vectors of each of the candidate document entries to obtain a target text vector that matches the current question vector, and the candidate document entry corresponding to the target text vector is determined as a matching document entry.

[0087] Optionally, the auxiliary devices used by the system also include:

[0088] A second answer information acquisition module is used to input the current system position and the current question vector into a pre-trained fine-tuned question-answering model according to the current system position and the current question vector, and obtain second answer information output by the fine-tuned question-answering model if no matching document entry is found in the vector database;

[0089] An information display module, used for displaying the second answer information to the user;

[0090] The fine-tuning question-answering model is obtained by fine-tuning the basic large language model using a vector database.

[0091] Optionally, the question vector acquisition module 210 is specifically used to display candidate questions to the user, and if a selection operation of the user on the candidate question is detected, the target candidate question selected by the user is obtained as the current user question;

[0092] If no selection operation of the user on the candidate question is detected, then when a question input operation of the user is detected, the current user question input by the user is obtained.

[0093] The auxiliary device used by the system provided in the embodiment of the present invention can execute the auxiliary method used by the system provided in any embodiment of the present invention, and has the corresponding functional modules and beneficial effects of the execution method.

[0094] In the technical solution of the present disclosure, the collection, storage, use, processing, transmission, provision and disclosure of user personal information involved are in compliance with the provisions of relevant laws and regulations and do not violate public order and good morals.

[0095] Embodiment 3

[0096] Figure 4 A schematic diagram of the structure of an electronic device 30 that can be used to implement an embodiment of the present invention is shown. The electronic device 30 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 30 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.

[0097] like Figure 4As shown, the electronic device 30 includes at least one processor 31, and a memory connected to the at least one processor 31, such as a read-only memory (ROM) 32, a random access memory (RAM) 33, etc., wherein the memory stores a computer program that can be executed by at least one processor, and the processor 31 can perform various appropriate actions and processes according to the computer program stored in the read-only memory 32 or the computer program loaded from the storage unit 38 to the random access memory 33. In the RAM 33, various programs and data required for the operation of the electronic device 30 can also be stored. The processor 31, the ROM 32 and the RAM 33 are connected to each other through a bus 34. The input / output (I / O) interface 35 is also connected to the bus 34.

[0098] A number of components in the electronic device 30 are connected to the I / O interface 35, including: an input unit 36, such as a keyboard, a mouse, etc.; an output unit 37, such as various types of displays, speakers, etc.; a storage unit 38, such as a disk, an optical disk, etc.; and a communication unit 39, such as a network card, a modem, a wireless communication transceiver, etc. The communication unit 39 allows the electronic device 30 to exchange information / data with other devices through a computer network such as the Internet and / or various telecommunication networks.

[0099] The processor 31 may be a variety of general and / or dedicated processing components with processing and computing capabilities. Some examples of the processor 31 include, but are not limited to, a central processing unit, a graphics processing unit, various dedicated artificial intelligence computing chips, various processors running machine learning model algorithms, digital signal processors, and any appropriate processors, controllers, microcontrollers, etc. The processor 31 performs the various methods and processes described above, such as auxiliary methods used by the system.

[0100] In some embodiments, the auxiliary method used by the system may be implemented as a computer program, which is tangibly contained in a computer-readable storage medium, such as a storage unit 38. In some embodiments, part or all of the computer program may be loaded and / or installed on the electronic device 30 via the ROM 32 and / or the communication unit 39. When the computer program is loaded into the RAM 33 and executed by the processor 31, one or more steps of the auxiliary method used by the system described above may be performed. Alternatively, in other embodiments, the processor 31 may be configured to execute the auxiliary method used by the system in any other appropriate manner (e.g., by means of firmware).

[0101] Various embodiments of the systems and techniques described above herein can be implemented in digital electronic circuit systems, integrated circuit systems, field programmable gate arrays, application specific integrated circuits, application specific standard products, systems on a chip, load programmable logic devices, computer hardware, firmware, software, and / or combinations thereof. These various embodiments may 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.

[0102] 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.

[0103] 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 conjunction 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, a read-only memory, an erasable programmable read-only memory, an optical fiber, a portable compact disk read-only memory, an optical storage device, a magnetic storage device, or any suitable combination of the foregoing.

[0104] To provide interaction with a user, the systems and techniques described herein may be implemented on an electronic device 30 having: a display device (e.g., a cathode ray tube or a liquid crystal display) for displaying information to the user; and a keyboard and a pointing device (e.g., a mouse or a trackball) through which the user can provide input to the electronic device 30. 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).

[0105] 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: local area networks, wide area networks, blockchain networks, and the Internet.

[0106] A computing system may include clients and servers. Clients and servers are generally remote from each other and typically interact through a communication network. The relationship of client and server is generated by computer programs running on respective computers and having a client-server relationship to each other. The server may be a cloud server.

[0107] This embodiment may also include a computer program product, which includes a computer program. When the computer program is executed by a processor, it implements the auxiliary method used by the system provided by any embodiment of the present invention.

[0108] 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.

[0109] 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. An auxiliary method used by a system, characterized in that: include: When it is detected that the triggering condition for system usage assistance is met, the current system position and the current user question are obtained, and a current question vector corresponding to the current user question is obtained; According to the current system position and the current question vector, if at least one matching document entry is found in the vector database, each matching document entry is input into a pre-trained target question-answering model, and first answer information output by the target question-answering model is obtained; The first answer information is displayed to the user.

2. The method according to claim 1, characterized in that The trigger conditions for using the system assistance are detected, including: If it is detected that the user's stay time at the current input position reaches the set time, or the user has not made a selection in the first selection box on the current page and the stay time reaches the set time, or the mouse pointer stays at a content location for a set time, then it is determined that the trigger conditions for the system to use assistance are met.

3. The method according to claim 1, characterized in that Before searching at least one matching document entry in a vector database according to the current system position and the current question vector, the method further includes: Obtaining system product documents, and performing text extraction on the system product documents to obtain text corpus and corresponding location information; Obtaining a text vector corresponding to the text corpus, and generating a mapping relationship between the text vector and the position information as a document entry, and establishing a vector database based on each document entry; The text corpus includes question-answer format and / or title-content format.

4. The method according to claim 3, characterized in that: According to the current system position and the current question vector, at least one matching document entry is found in a vector database, including: Performing a match detection between the current system position and the position information of each document entry in the vector database, acquiring target position information matching the current system position, and determining the document entry corresponding to the target position information as a candidate document entry; The current question vector is matched with the text vectors of each of the candidate document entries to obtain a target text vector that matches the current question vector, and the candidate document entry corresponding to the target text vector is determined as a matching document entry.

5. The method according to claim 1, characterized in that After obtaining the current question vector corresponding to the current user question, the method further includes: According to the current system position and the current question vector, if no matching document entry is found in the vector database, the current system position and the current question vector are input into a pre-trained fine-tuned question-answering model, and second answer information output by the fine-tuned question-answering model is obtained; Displaying the second answer information to the user; The fine-tuning question-answering model is obtained by fine-tuning the basic large language model using a vector database.

6. The method according to claim 1, characterized in that Get the current user's question, including: Display candidate questions to the user. If a user selection operation on a candidate question is detected, obtain the target candidate question selected by the user as the current user question. If no selection operation of the user on the candidate question is detected, then when a question input operation of the user is detected, the current user question input by the user is obtained.

7. An auxiliary device used in a system, characterized in that: include: A question vector acquisition module, used to acquire the current system position and the current user question when it is detected that the trigger condition for system use assistance is met, and to acquire the current question vector corresponding to the current user question; A first answer information acquisition module is used to input each matching document entry into a pre-trained target question-answering model according to the current system position and the current question vector, if at least one matching document entry is found in the vector database, and obtain first answer information output by the target question-answering model; The answer information display module is used to display the first answer information to the user.

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 to enable the at least one processor to perform the auxiliary method used by the system according to any one of claims 1 to 6.

9. A computer-readable storage medium, characterized in that: The computer-readable storage medium stores a computer program, and the computer program is used to enable a processor to implement the auxiliary method used by the system according to any one of claims 1 to 6 when the computer program is executed.

10. A computer program product, characterized in that The invention comprises a computer program, which, when executed by a processor, implements the auxiliary method used by the system according to any one of claims 1 to 6.

Images