Business guidance generation method and device, equipment, storage medium and product

Abstract

The embodiment of the invention relates to the technical field of artificial intelligence, and provides a business guidance generation method and device, equipment, a storage medium and a product, and the method comprises the steps: carrying out the logic correction of a first demand text inputted by a user, extracting a professional noun and verb combination from the corrected first demand text, and obtaining a business logic text; obtaining a first business guidance text and a first business handling entry link according to the business logic text by utilizing a business guidance generation model; receiving feedback of the user to the first business guide text, and if the feedback is that the demand is not met, regenerating a second business guide text and a second business handling entry link according to a second demand text input by the user; and if the feedback is that the requirement is met, extracting an operation parameter and a service handling result parameter of the user for carrying out service handling according to the first service guidance text as training samples, and carrying out iterative training on the service guidance generation model. According to the embodiment of the invention, accurate business guidance can be provided for the user.

Description

Technical Field

[0001] The embodiments of this specification relate to the field of artificial intelligence technology, and in particular to a method, device, equipment, storage medium and product for generating business guidance. Background Art

[0002] With the development of financial technology, banks are accelerating their digital transformation. As an important part of online banking services, business intelligent guidance helps promote this process. Business intelligent guidance can automatically answer user questions and provide service suggestions to users. By providing intelligent guidance and services, it can reduce the workload of manual customer service, improve service response speed and efficiency, and reduce operating costs.

[0003] At present, the online banking system usually places the commonly used function entrances on the system homepage for users to handle business, and the uncommon function entrances are grouped through the classification tab page. Users can choose the appropriate function entrance according to their needs. During the business handling process, error prompts are used to guide users to complete the business handling. However, in this method, users are unfamiliar with and uncommon functions and usually cannot find the required function entrances. Especially for user groups who use the Internet less, it is difficult to understand the content scope of each tab page (classification). In addition, the error prompts for failed business handling completely rely on the prompt text set by the development or operation personnel for guidance. Users need to understand the meaning of the prompt first and then complete some business handling according to the requirements of the prompt, which is difficult and difficult to fully meet the diverse needs of users. Therefore, there is an urgent need for a business guide generation method that can provide users with accurate business guidance to meet the personalized needs of users. Summary of the invention

[0004] In view of the above-mentioned problems in the prior art, the purpose of the embodiments of this specification is to provide a business guidance generation method, device, equipment, storage medium and product to solve the problem in the prior art that it is difficult to provide accurate business guidance to users.

[0005] In order to solve the above technical problems, the specific technical solutions of the embodiments of this specification are as follows:

[0006] On the one hand, an embodiment of the present specification provides a method for generating a business guide, the method comprising:

[0007] receiving a first requirement text input by a user, and performing logic correction on the first requirement text;

[0008] Extracting professional nouns and verbs related to the business from the revised first requirement text, and combining the extracted professional nouns and verbs to obtain a business logic text;

[0009] Input the business logic text into a pre-trained business guidance generation model to obtain a first business guidance text and a first business handling entry link;

[0010] Receive user feedback on the first business guidance text. If the feedback is that the requirements are not met, re-receive a second requirement text input by the user, and repeat the above steps to generate a second business guidance text and a second business handling entry link according to the second requirement text using the real-time tuning mechanism in the business guidance generation model;

[0011] If the feedback is that the requirements are met, the operation parameters and business processing result parameters of the user's business processing according to the first business guidance text and the first business processing entry link are extracted, and the operation parameters and business processing result parameters are used as training samples to train the business guidance generation model, and the trained business guidance generation model is used as a new business guidance generation model to generate business guidance text.

[0012] Furthermore, the performing logical correction on the first requirement text includes:

[0013] Preprocessing the first requirement text to obtain a preprocessed first requirement text;

[0014] Performing semantic recognition on the preprocessed first requirement text to obtain semantic representation;

[0015] Inputting the semantic representation into a pre-trained logic detection model to obtain a logic error identifier in the pre-processed first requirement text;

[0016] A corresponding logic correction model is selected according to the logic error identifier to correct the preprocessed first requirement text to obtain a corrected first requirement text.

[0017] Furthermore, the extracting of professional terms and verbs related to the business from the revised first requirement text includes:

[0018] Performing word segmentation processing on the modified first requirement text to obtain a word segmentation list, wherein the word segmentation list includes position information of each word segmentation in the first requirement text;

[0019] Performing part-of-speech tagging on each word in the word list to obtain the part-of-speech of each word;

[0020] Filtering the word segmentation whose part of speech is noun from the word segmentation list, calculating the TF-IDF value of each noun, and obtaining an initial professional noun list according to the TF-IDF value;

[0021] Sorting the initial professional noun list according to the pre-built domain knowledge base, screening the initial professional nouns whose rankings are within the preset value, and obtaining a final professional noun list;

[0022] The word list is filtered according to the final professional noun list, and the word list is obtained by filtering the word list with the word part of verb.

[0023] Furthermore, the extracted professional nouns and verbs are combined to obtain a business logic text, including:

[0024] Performing semantic recognition on the extracted professional nouns and verbs to obtain semantic features of each professional noun and verb;

[0025] Obtain the position information of each professional noun and verb in the first requirement text;

[0026] The semantic features and corresponding position information of each professional noun and verb are fused to obtain a feature vector, which is input into a pre-trained business logic text generation model to obtain a business logic text.

[0027] Furthermore, the business guidance generation model includes a business guidance text generation sub-model and a business handling entry link generation sub-model;

[0028] The step of inputting the business logic text into a pre-trained business guidance generation model to obtain a first business guidance text and a first business handling entry link includes:

[0029] The business guidance text generation sub-model receives the business logic text, and generates a first business guidance text according to the business logic text;

[0030] The business processing entry link generation sub-model receives the first business guidance text, generates a business process according to the first business guidance text, obtains the business identifier corresponding to each business process, and queries the business processing system database according to the business identifier to obtain the corresponding first processing entry link.

[0031] Furthermore, the step of using the real-time optimization mechanism in the business guidance generation model to generate a second business guidance text and a second business handling entry link according to the second requirement text includes:

[0032] Adjusting the model parameters and / or model structure of the business guidance generation model according to the second requirement text to obtain a fine-tuned business guidance generation model;

[0033] Using the test data set to perform a performance evaluation on the fine-tuned business guidance generation model, and judging whether the fine-tuned business guidance generation model meets the preset standard according to the evaluation result;

[0034] If yes, input the second requirement text into the fine-tuned business guidance generation model to generate a second business guidance text and a second business handling entry link;

[0035] If not, the model parameters and / or model structure of the business guidance generation model are readjusted until the fine-tuned business guidance generation model meets the preset standards.

[0036] Furthermore, the using the operation parameters and the business handling result parameters as training samples to train the business guidance generation model includes:

[0037] Calculate the effective value of the operation parameter and business processing result parameter of each processing step in the business guidance text according to the preset effective value calculation rules;

[0038] Extracting business features and user features corresponding to each processing step in the business guidance text according to historical business processing data, wherein the business features include business type, complexity and processing success rate, and the user features include user historical behavior patterns;

[0039] Associating the valid values ​​of the operation parameters and business processing result parameters of each processing step in the business guidance text with the business characteristics and user characteristics corresponding to each processing step to obtain a label corresponding to each processing step in the business guidance text;

[0040] The business guidance generation model is trained according to each processing step and corresponding label in the business guidance text to obtain a trained business guidance generation model.

[0041] Furthermore, the operation parameters include click and return, the business processing result parameters include processing failure and processing success, and the effective value calculation rules include:

[0042] For the multiple processing steps included in the business guidance text, an initial effective value is set for the first processing step. If it is detected that the user clicks on the next processing step, the effective value of the processing step is obtained by increasing the initial effective value according to a preset ratio.

[0043] If it is detected that the user's operation for the next processing step is to return, then the effective value of the processing step is obtained by decreasing the initial effective value according to a preset ratio;

[0044] If it is detected that the business processing result parameter is successfully processed, then a preset value is added to the valid value of the last processing step to obtain the valid value of the business processing result parameter;

[0045] If it is detected that the service processing result parameter is a processing failure, the preset value is reduced based on the effective value of the last processing step to obtain the effective value of the service processing result parameter.

[0046] On the other hand, an embodiment of the present specification further provides a service guide generating device, the device comprising:

[0047] A receiving module, used for receiving a first requirement text input by a user, and performing logic correction on the first requirement text;

[0048] An extraction module, used to extract professional nouns and verbs related to the business from the revised first requirement text, and combine the extracted professional nouns and verbs to obtain a business logic text;

[0049] A first generating module, used for inputting the business logic text into a pre-trained business guidance generating model to obtain a first business guidance text and a first business handling entry link;

[0050] A second generation module is used to receive user feedback on the first business guidance text. If the feedback is that the first business guidance text does not meet the requirements, the second requirement text input by the user is received again, and the above steps are repeated to generate a second business guidance text and a second business handling entry link according to the second requirement text by using the real-time tuning mechanism in the business guidance generation model;

[0051] The model optimization module is used to extract the operation parameters and business processing result parameters of the user's business processing according to the first business guidance text and the first business processing entry link if the feedback is that the requirements are met, and use the operation parameters and business processing result parameters as training samples to train the business guidance generation model, and use the trained business guidance generation model as a new business guidance generation model to generate business guidance text.

[0052] On the other hand, an embodiment of the present specification further provides a computer device, including a memory, a processor, and a computer program stored in the memory, wherein when the computer program is executed by the processor, the instructions of any one of the above methods are executed.

[0053] On the other hand, an embodiment of the present specification further provides a computer-readable storage medium having a computer program stored thereon, wherein when the computer program is executed by a processor of a computer device, the computer program executes instructions of any one of the above-described methods.

[0054] On the other hand, the embodiments of this specification further provide a computer program product, which, when executed by a processor of a computer device, executes instructions of any one of the above methods.

[0055] By adopting the above technical solution, the business guidance generation method provided in the embodiment of this specification can perform logical correction on the demand text input by the user, ensure the accuracy and clarity of the input data, reduce the understanding deviation caused by the user's input error or unclear expression, and extract the professional nouns and verbs in the demand text and combine them into a business logic text, so as to more accurately understand the user's intention, thereby providing data support for the subsequent generation of targeted business guidance. By using the pre-trained business guidance generation model, the automatic generation from user demand to business guidance text and business handling entry link is realized, which greatly reduces manual intervention and improves processing efficiency and accuracy. The generated business guidance text and business handling entry link are adjusted in real time according to the user's feedback to ensure that the generated business guidance is more in line with the actual needs of the user, thereby significantly improving the user experience. In addition, by extracting the operation parameters and result parameters of the user's business handling according to the business guidance text and the handling entry link as training samples, the business guidance generation model is iteratively trained, so that the model can continuously learn and optimize to improve the ability to predict and meet future user needs, thereby improving the accuracy of the generated business guidance text.

[0056] The above description is only an overview of the technical solutions of some embodiments of this specification. In order to more clearly understand the technical means of some embodiments of this specification, they can be implemented in accordance with the contents of the specification. In order to make the above and other purposes, features and advantages of the embodiments of this specification more obvious and easy to understand, the following specifically cites the preferred embodiments and describes them in detail in conjunction with the attached drawings. BRIEF DESCRIPTION OF THE DRAWINGS

[0057] In order to more clearly illustrate the embodiments of this specification or the technical solutions in the prior art, the drawings required for use in the embodiments or the description of the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of this specification. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative work.

[0058] Figure 1 A schematic diagram showing an implementation environment of a business guide generating method in some embodiments of this specification;

[0059] Figure 2 A schematic diagram showing a process of a method for generating a business guide in some embodiments of this specification is shown;

[0060] Figure 3 A schematic diagram of a process of performing logical correction on the first requirement text in some embodiments of this specification is shown;

[0061] Figure 4A schematic diagram of a process of extracting professional terms and verbs related to a business from a revised first requirement text in some embodiments of this specification is shown;

[0062] Figure 5 A schematic diagram of a process of combining the extracted professional nouns and verbs to obtain a business logic text in some embodiments of this specification is shown;

[0063] Figure 6 A schematic diagram of a process for generating a first business guidance text and a first business handling entry link in some embodiments of this specification is shown;

[0064] Figure 7 A schematic diagram of the process of generating a second business guidance text and a second business handling portal link in some embodiments of this specification is shown;

[0065] Figure 8 A schematic diagram of a process of training the business guidance generation model by using the operation parameters and business handling result parameters as training samples in some embodiments of this specification is shown;

[0066] Fig. 9 A schematic diagram showing the module structure of a business guide generating device in some embodiments of this specification is shown;

[0067] Fig.10 A schematic diagram of the structure of a computer device in this specification is shown.

[0068] Description of the accompanying symbols:

[0069] 901, receiving module;

[0070] 902, extraction module;

[0071] 903, first generation module;

[0072] 904, second generation module;

[0073] 905. Model optimization module;

[0074] 1002. Computer equipment;

[0075] 1004. Processor;

[0076] 1006. Memory;

[0077] 1008. Driving mechanism;

[0078] 1010, input / output module;

[0079] 1012. Input device;

[0080] 1014. Output device;

[0081] 1016. Presentation equipment;

[0082] 1018. Graphical user interface;

[0083] 1020. Network interface;

[0084] 1022. Communication link;

[0085] 1024. Communication bus. DETAILED DESCRIPTION

[0086] The following will be combined with the drawings in the embodiments of this specification to clearly and completely describe the technical solutions in the embodiments of this specification. Obviously, the described embodiments are only part of the embodiments of this specification, not all of the embodiments. Based on the embodiments in this specification, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of this specification.

[0087] 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 acquisition, storage, use, and processing of data in the technical solutions described in the embodiments of this application are in compliance with the relevant provisions of national laws and regulations.

[0088] It should be noted that in the embodiments of this specification, certain software, components, models and other existing solutions in the industry may be mentioned, and they should be regarded as exemplary. Their purpose is only to illustrate the feasibility of implementing the technical solution of this application, but it does not mean that the applicant has or will necessarily use the solution.

[0089] It should be noted that the terms "first", "second", etc. in this specification and claims 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 this specification 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, device, product or equipment 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 equipment.

[0090] like Figure 1The figure shows a schematic diagram of the implementation environment of a business guidance generation method according to an embodiment of the present specification, which may include: a terminal 101 and a server 102, wherein the terminal 101 and the server 102 communicate with each other through a network, and the network may include a local area network (LAN), a wide area network (WAN), the Internet or a combination thereof, and is connected to a website, a user device (such as a computing device) and a back-end system.

[0091] The terminal 101 is used for the user to input the demand text and the feedback on the first business guidance text, and sends the demand text and the feedback on the first business guidance text to the server 102. In an optional embodiment, the terminal 101 may include but is not limited to electronic devices such as self-service terminal devices, desktop computers, tablet computers, laptop computers, smart wearable devices, etc. Optionally, the operating system running on the electronic device may include but is not limited to Android system, IOS system, Linux, Windows, etc. Of course, the terminal 101 is not limited to the above-mentioned electronic devices with a certain entity, and it can also be software running in the above-mentioned electronic devices.

[0092] The server end 102 can be used to receive a first demand text input by a user and perform logical correction on the first demand text; and to extract professional terms and verbs related to the business from the corrected first demand text, and combine the extracted professional terms and verbs to obtain a business logic text; and to input the business logic text into a pre-trained business guidance generation model to obtain a first business guidance text and a first business processing entry link; and to receive user feedback on the first business guidance text. If the feedback is that the demand is not met, the second demand text input by the user is received again, and the above steps are repeated to generate a second business guidance text and a second business processing entry link according to the second demand text using the real-time tuning mechanism in the business guidance generation model; and if the feedback is that the demand is met, the operation parameters and business processing result parameters of the user performing business processing according to the first business guidance text and the first business processing entry link are extracted, and the operation parameters and business processing result parameters are used as training samples to train the business guidance generation model, and the trained business guidance generation model is used as a new business guidance generation model to generate business guidance text. Optionally, the server end 102 can be an independent physical server, or a server cluster or distributed system composed of multiple physical servers, or a cloud server that provides basic cloud computing services such as cloud services, cloud databases, cloud computing, cloud functions, cloud storage, network services, cloud communications, middleware services, domain name services, security services, content distribution networks (CDN, Content Delivery Network), and big data and artificial intelligence platforms.

[0093] It should be noted that Figure 1 This is only one implementation environment of the business guide generation method provided in the embodiment of the present application. In actual application, other implementation environments may also be included.

[0094] Figure 2 This is a flowchart of a business guidance generation method provided in an embodiment of this specification. This specification provides method operation steps as described in the embodiment or flowchart, but may include more or fewer operation steps based on conventional or non-creative labor. The order of steps listed in the embodiment is only one way of executing the steps among many orders, and does not represent the only order of execution. When the actual system or device product is executed, it can be executed in the order of the method shown in the embodiment or the accompanying drawings or in parallel. Figure 2 As shown, an embodiment of this specification provides a method for generating a business guide, the method comprising:

[0095] S201: receiving a first requirement text input by a user, and performing logic correction on the first requirement text;

[0096] S202: extracting professional nouns and verbs related to the business from the revised first requirement text, and combining the extracted professional nouns and verbs to obtain a business logic text;

[0097] S203: Inputting the business logic text into a pre-trained business guidance generation model to obtain a first business guidance text and a first business handling entry link;

[0098] S204: receiving user feedback on the first business guidance text. If the feedback is that the requirements are not met, receiving the second requirement text input by the user again, and repeating the above steps, using the real-time tuning mechanism in the business guidance generation model to generate a second business guidance text and a second business handling entry link according to the second requirement text;

[0099] S205: If the feedback is that the demand is met, then extract the operation parameters and business processing result parameters of the user's business processing according to the first business guidance text and the first business processing entry link, use the operation parameters and business processing result parameters as training samples, train the business guidance generation model, and use the trained business guidance generation model as a new business guidance generation model to generate business guidance text.

[0100] By adopting the above technical solution, the business guidance generation method provided in the embodiment of this specification can perform logical correction on the demand text input by the user, ensure the accuracy and clarity of the input data, reduce the understanding deviation caused by the user's input error or unclear expression, and extract the professional nouns and verbs in the demand text and combine them into a business logic text, so as to more accurately understand the user's intention, thereby providing data support for the subsequent generation of targeted business guidance. By using the pre-trained business guidance generation model, the automatic generation from user demand to business guidance text and business handling entry link is realized, which greatly reduces manual intervention and improves processing efficiency and accuracy. The generated business guidance text and business handling entry link are adjusted in real time according to the user's feedback to ensure that the generated business guidance is more in line with the actual needs of the user, thereby significantly improving the user experience. In addition, by extracting the operation parameters and result parameters of the user's business handling according to the business guidance text and the handling entry link as training samples, the business guidance generation model is iteratively trained, so that the model can continuously learn and optimize to improve the ability to predict and meet future user needs, thereby improving the accuracy of the generated business guidance text.

[0101] In some embodiments, in order to improve the user experience, in step S201, the user's demand voice may be received first, and then the demand voice may be converted into demand text. In order to reduce the problem of misunderstanding caused by user input errors or unclear expressions, in the embodiments of this specification, in step S201, reference is made to Figure 3, the logical correction of the first requirement text includes:

[0102] S301: Preprocess the first requirement text to obtain the preprocessed first requirement text.

[0103] S302: Perform semantic recognition on the preprocessed first requirement text to obtain a semantic representation.

[0104] S303: Input the semantic representation into a pre-trained logical detection model to obtain the logical error identification in the preprocessed first requirement text.

[0105] S304: Select a corresponding logical correction model according to the logical error identification to correct the preprocessed first requirement text to obtain the corrected first requirement text.

[0106] Specifically, in step S301, preprocessing the first requirement text includes removing irrelevant characters in the first requirement text, such as HTML tags, special symbols, extra spaces, line breaks, etc., and removing words that frequently appear in the first requirement text but contribute little to the meaning of the requirement text, such as words like "de" (的) and "le" (了), and unifying words with the same meaning in the first requirement text to reduce lexical diversity. In step S302, performing semantic recognition on the preprocessed first requirement text to obtain a semantic representation can convert natural language into a semantic representation that can be understood by a computer, so that the computer can understand the meaning expressed by the first requirement text. Optionally, a pre-trained natural language processing model, such as BERT, Word2Vec, or Transformer, can be used to extract semantic features from the requirement text to generate a semantic representation. In step S303, the logical detection model is used to detect logical errors in the requirement text, such as semantic contradictions and incoherence. Inputting the semantic representation into the trained logical detection model to obtain the positions of logical errors in the first requirement text and the corresponding logical error identifications. In step S304, the logical correction model is used to automatically correct the logical errors in the first requirement text. Select the corresponding logical correction model according to the logical error identification, and perform some post-processing operations on the corrected first requirement text, such as grammar checking and fluency improvement, to further improve the quality of the first requirement text.

[0107] In the embodiment of this specification, in step S202, referring to Figure 4 , the extraction of professional nouns and verbs related to the business from the corrected first requirement text includes:

[0108] S401: Perform word segmentation on the corrected first requirement text to obtain a word segmentation list, and the word segmentation list contains the position information of each word segment in the first requirement text.

[0109] S402: Perform part-of-speech tagging on each word in the word list to obtain the part-of-speech of each word.

[0110] S403: Filter the word segmentation whose part of speech is noun from the word segmentation list, calculate the TF-IDF value of each noun, and obtain an initial professional noun list according to the TF-IDF value.

[0111] S404: Sort the initial professional noun list according to the pre-constructed domain knowledge base, filter the initial professional nouns whose rankings are within a preset value, and obtain a final professional noun list.

[0112] S405: Filter the participles of verbal nature in the participle list according to the final professional noun list to obtain a verb list.

[0113] Specifically, in step S401, the word segmentation tool can be used to segment the corrected first demand text, and the starting position and the ending position of each word segmentation in the first demand text can be marked to generate a word segmentation list, each element in the list contains the word segmentation itself and its position information, wherein the position information can be marked by adding a character index or a word index to the word segmentation. In step S402, the word segmentation tool in the word segmentation tool can be used to perform part-of-speech tagging on each word segmentation list, such as noun, verb, adjective, etc., to obtain the part-of-speech tag of each word segmentation, and the part-of-speech tag of each word segmentation is added to the word segmentation list. In step S403, professional nouns refer to words related to the business field, and the importance of nouns in the text is evaluated by screening out the word segmentation with a noun part of speech and calculating the TF-IDF value (term frequency-inverse document frequency value) of each noun. The higher the TF-IDF value, the more important the noun is in the current demand text and the more likely it is to be a professional noun. The word segmentation with a TF-IDF value greater than the preset threshold is extracted, thereby preliminarily screening out the word segmentation that is most likely to be a professional noun. To further ensure that the selected nouns are highly relevant to the business field, in step S404, the initial professional noun list is further screened through the domain knowledge base, which refers to a knowledge base containing professional nouns in the business field, which can be constructed according to the business field knowledge and stored in the form of a database. The initial professional noun list is compared with the domain knowledge base, and the nouns are sorted according to the matching degree, and the nouns ranked in the front preset value are screened as the final professional noun list. In step S405, verbs related to the final professional noun list are screened from the word list, and these verbs represent operations or behaviors related to professional nouns. Optionally, a verb list can be formed by calculating the semantic association or co-occurrence frequency between the verb and the final professional noun. For example, the first demand text entered by the user is: "Please help me check the current balance of my xx account (account number: xxxxxxx), and plan to transfer 3,000 yuan to Xiaox of xx institution at the beginning of next month", then according to the demand text entered by the user, the professional noun list can be extracted through the above steps as [xx account, balance, xx institution], and the verb list is [query, plan, transfer].

[0114] In the embodiment of this specification, by first extracting professional terms and verbs from the demand text and combining them into a business logic text, the user's intention can be understood more accurately, thereby providing data support for the subsequent generation of targeted business guidance. Figure 5 , the extracted professional nouns and verbs are combined to obtain business logic text, including:

[0115] S501: Perform semantic recognition on the extracted professional nouns and verbs to obtain semantic features of each professional noun and verb.

[0116] S502: Obtain the position information of each professional noun and verb in the first requirement text.

[0117] S503: The semantic features and corresponding position information of each professional noun and verb are integrated to obtain a feature vector, which is input into a pre-trained business logic text generation model to obtain a business logic text.

[0118] Specifically, in step S501, for the extracted professional nouns and verbs, semantic analysis techniques (such as word embedding, pre-trained language models, etc.) can be used to obtain their semantic features, and the semantic features can be used to represent the meaning of vocabulary in a specific context and the semantic relationship with other vocabulary. Optionally, the context information of professional nouns and verbs in sentences can also be extracted to more accurately capture their semantic features. In step S502, the position information of each professional noun and verb in the first demand text can be obtained from the word segmentation list, and then the position information is encoded into a vector form to be fused with the semantic features. The position encoding can include relative position (such as the distance from the beginning of the sentence) or absolute position (such as a specific character or word index). In step S503, the semantic features of each professional noun and verb are spliced ​​with the corresponding position information encoding to form a feature vector containing both information. The feature vector is input into a pre-trained business logic text generation model, and the business logic text generation model can be combined into a text that can highly summarize user needs and is logically normal according to the extracted professional nouns and verbs and their corresponding positional relationships. Continuing with the above example, if the extracted professional noun list is [xx account, balance, xx institution], and the verb list is [query, plan, transfer], the business logic text that can be obtained is: "Query xx account balance process, regular transfer process to xx institution".

[0119] In the embodiment of this specification, in step S203, the business guidance generation model includes a business guidance text generation sub-model and a business handling entry link generation sub-model. Figure 6 , the step of inputting the business logic text into a pre-trained business guidance generation model to obtain a first business guidance text and a first business handling entry link includes:

[0120] S601: The business guidance text generation sub-model receives the business logic text, and generates a first business guidance text according to the business logic text.

[0121] S602: The business processing entry link generation sub-model receives the first business guidance text, generates a business process according to the first business guidance text, obtains the business identifier corresponding to each business process, and queries the business processing system database according to the business identifier to obtain the corresponding first processing entry link.

[0122] Specifically, in step S601, the business guidance text refers to a detailed explanation of the business being handled, including business handling conditions, business handling steps, business handling operation points, risk warnings, precautions, etc. If the business to be handled contains preconditions, that is, other businesses must be handled before handling the business, the business guidance text should also contain the guidance text of the preconditioning business. The business guidance text generation submodel is used to generate the business guidance text according to the business logic text. The model can be a sequence to sequence (Seq2Seq) model, a knowledge-based reasoning model, etc., and it is trained through a large amount of historical business handling data to finally obtain a trained business guidance text generation submodel. In step S602, the business handling entry link generation submodel can be used to generate a business handling entry link according to the business guidance text. First, a business process is generated according to the business guidance text. For each business process, there is a business identifier that uniquely identifies the business process in the business handling system database. Through the business identifier, the business handling entry link generation submodel will initiate a query request to the business handling system database to obtain the handling entry link corresponding to the business process. This link may point to a web page, an API interface or a specific page in an application. Finally, the business processing entry link generation sub-model will integrate the queried processing entry link with the business guidance text to form a complete business guidance text containing specific operation guidance and direct processing entry, which will be sent to the user's corresponding terminal. In this way, users no longer need to manually search and enter complex business processing information, but can directly use the generated complete business guidance text to step by step complete the business processing process according to the guidance, and directly click the link to enter the processing page when necessary, which not only saves users' time and energy, but also reduces the risk of errors and disputes caused by improper operation.

[0123] In some embodiments, the first business guidance text and the first business processing entry link generated by the business guidance generation model may not meet user needs. In order to improve the user experience and the accuracy of the business guidance text generation, the generated business guidance text and the business processing entry link may be adjusted in real time according to user feedback to ensure that the generated business guidance better meets the actual needs of users. Specifically, a result feedback button may be set on the terminal page. If the user feedback is that the needs are not met, the second requirement text input by the user is received again, and steps S201-203 are repeated to utilize the real-time tuning mechanism in the business guidance generation model to generate the second business guidance text and the second business processing entry link according to the second requirement text. In the embodiments of this specification, refer to Figure 7 , the step of using the real-time tuning mechanism in the business guidance generation model to generate a second business guidance text and a second business handling entry link according to the second demand text includes:

[0124] S701: Adjust the model parameters and / or model structure of the business guidance generation model according to the second requirement text to obtain a fine-tuned business guidance generation model.

[0125] S702: Using a test data set to perform a performance evaluation on the fine-tuned business guidance generation model, and judging whether the fine-tuned business guidance generation model meets a preset standard according to the evaluation result.

[0126] S703: If yes, input the second requirement text into the fine-tuned business guidance generation model to generate a second business guidance text and a second business processing entry link.

[0127] S704: If not, readjust the model parameters and / or model structure of the business guidance generation model until the fine-tuned business guidance generation model meets the preset standard.

[0128] Specifically, in step S701, the model parameters and / or model structure of the trained business guidance generation model are fine-tuned according to the second requirement text re-entered by the user. The model parameters include neural network weights, biases, learning rates, etc., and the model parameters can be automatically adjusted using algorithms such as gradient descent method and Adam optimizer. The model structure can be adjusted by increasing or decreasing the number of network layers, adjusting the connection mode between network layers, or introducing new feature processing layers, such as attention mechanism layers. In the specific application process, the model parameters and model structure can be adjusted simultaneously according to the second requirement text, or only one of them can be adjusted. In step S702, after the business guidance generation model is fine-tuned, it is necessary to perform a performance evaluation on it in order to comprehensively evaluate the generalization ability and accuracy of the model. Specifically, the test data set used in the historical training process is obtained to perform a performance evaluation on it. The evaluation indicators can be accuracy, recall rate, F1 score, etc. After the test is completed, it is determined whether the fine-tuned business guidance generation model meets the preset standard according to the evaluation indicator value. If it meets the standard, it is used as a new business guidance generation model, and the second business guidance text and the second business handling entry link are generated according to the second requirement text. If not, readjust the model parameters and / or model structure according to the test results, and iterate the model until the business guidance generation model meets the preset standards. In this way, the generated business guidance text and business processing entry link are adjusted in real time according to user feedback to ensure that the generated business guidance is more in line with the actual needs of users, thereby significantly improving the user experience. Furthermore, if the second business guidance text and the second business processing entry link still do not meet user needs, continue to receive new demand texts entered by the user until the generated business guidance text and business processing entry link meet user needs. In this way, through multiple interactions between the user and the server, the business guidance text generation model can improve its ability to understand the user input demand text, thereby improving the accuracy of the generated business guidance text.

[0129] In some embodiments, if the user feedback is that the demand is met, the operation parameters and business processing result parameters of the user's business processing according to the first business guidance text and the first business processing entry link are extracted, and the operation parameters and business processing result parameters are used as training samples to train the business guidance generation model, and the trained business guidance generation model is used as a new business guidance generation model to generate business guidance text. Specifically, refer to Figure 8 As shown, the operation parameters and the business processing result parameters are used as training samples to train the business guidance generation model, including:

[0130] S801: Calculate the effective values ​​of the operation parameters and business processing result parameters of each processing step in the business guidance text according to the preset effective value calculation rules.

[0131] Specifically, the operation parameters include click and return, the business processing result parameters include processing failure and processing success, and the effective value is used to represent the user's recognition of the business processing flow generated by the business guidance generation model. The effective value calculation rules include:

[0132] (1) For the multiple processing steps included in the business guidance text, an initial effective value is set for the first processing step. If it is detected that the user clicks on the next processing step, the effective value of the processing step is obtained by increasing the initial effective value according to a preset ratio;

[0133] (2) If it is detected that the user's operation for the next processing step is to return, then the effective value of the processing step is obtained by decreasing the initial effective value according to a preset ratio;

[0134] (3) If it is detected that the service processing result parameter is successfully processed, a preset value is added to the effective value of the last processing step to obtain the effective value of the service processing result parameter;

[0135] (4) If it is detected that the service processing result parameter is a processing failure, the preset value is reduced based on the effective value of the last processing step to obtain the effective value of the service processing result parameter.

[0136] It can be understood that the user clicks to indicate that he agrees with the business handling process generated by the business guidance generation model, which is recorded as the generated business is valid. Each business handling involves multiple clicks, and the validity of each click increases with the steps. Successful and failed handling are also recorded as valid. The user's active exit and return are recorded as invalid, and the validity of each click decreases with the steps. Specifically, for the first handling step in the business guidance text, an initial valid value is set for it, such as 0.1. When the user successfully completes the current step and clicks to enter the next step, the system believes that the step is well executed. Therefore, based on the initial valid value of the step, according to the preset ratio, such as increasing by 10%, its valid value is increased. In this way, the business process that locks the steps that the user smoothly advances will receive a higher evaluation. If the user returns from the current step, it may be because the information needs to be rechecked or modified. The system believes that the step may need to be re-evaluated or there is some deficiency. Therefore, based on the initial valid value of the current step, according to the preset ratio, such as reducing by 5%, its valid value is decreased. When the business handling result parameter is detected as "handling successfully", it means that the user has successfully completed all steps and successfully achieved the goal. At this time, based on the effective value of the last processing step, a preset value is added, such as an increase of 20%, to reflect the successful completion of the entire business and the efficient operation of the user. If the business processing result parameter is "processing failure", it means that although the user may have completed some steps, the business was ultimately not successfully processed. At this time, based on the effective value of the last processing step, a preset value is reduced, such as a reduction of 15%, to reflect the failure of business processing and possible problems. Therefore, through the above-mentioned effective value calculation rules, the user's recognition of the business processing steps in the business guidance text generated by the business guidance generation model can be reflected, so that the business guidance generation model can be adjusted according to the effective value to improve the effectiveness of the business guidance text generated by the model.

[0137] S802: Extracting business features and user features corresponding to each processing step in the business guidance text based on historical business processing data, wherein the business features include business type, complexity and processing success rate, and the user features include user historical behavior patterns.

[0138] Specifically, from the historical business processing data, relevant business features are extracted for each processing step, including but not limited to business type, business complexity, and processing success rate, among which business complexity can be measured by the number of processing steps or the complexity of required materials, etc. User features are also extracted from the historical business processing data, including but not limited to user historical behavior patterns, such as common operation paths, return rates, modification frequencies, etc., as well as user types, such as novices, old users, and high-frequency users.

[0139] S803: Associate the valid values ​​of the operation parameters and business processing result parameters of each processing step in the business guidance text with the business characteristics and user characteristics corresponding to each processing step to obtain a label corresponding to each processing step in the business guidance text.

[0140] Specifically, the valid values ​​of the operation parameters and business processing result parameters of each processing step are associated with the corresponding business characteristics and user characteristics extracted from the historical business processing data to establish a multi-dimensional array. Each array contains information on the valid values, business characteristics and user characteristics of the processing step, and the array is used as the label of the processing step for the optimization training of the subsequent business guidance generation model.

[0141] S804: The business guidance generation model is trained according to each processing step and corresponding label in the business guidance text to obtain a trained business guidance generation model.

[0142] Specifically, a data sample containing business processing steps and corresponding labels is used to train the business guidance generation model. During the training process, the business guidance generation model will learn how to predict or optimize the order, content or prompts of the processing steps according to the business processing steps and their corresponding valid values, user characteristics, and business characteristics, so as to improve the success rate of business processing and user satisfaction. Through iterative training and verification, the model parameters and structure are continuously adjusted until the model shows good generalization ability and accuracy on the test set. The trained business guidance generation model is used as a new business guidance generation model to generate business guidance text. In this way, by extracting the operation parameters and result parameters of the user's business processing according to the business guidance text and the processing entry link as training samples, the business guidance generation model is iteratively trained, so that the model can continuously learn and optimize to improve the ability to predict and meet future user needs, thereby improving the accuracy of the generated business guidance text.

[0143] Based on the above-mentioned method for generating a business guide, an embodiment of this specification also provides a corresponding device for generating a business guide. The device may include a system (including a distributed system), software (application), module, component, server, client, etc. using the method described in the embodiment of this specification and a device combined with necessary implementation hardware. Based on the same innovative concept, the device in one or more embodiments provided in the embodiment of this specification is as described in the following embodiments. Since the implementation scheme and method for solving the problem of the device are similar, the implementation of the specific device in the embodiment of this specification can refer to the implementation of the aforementioned method, and the repetitions will not be repeated. As used below, the term "unit" or "module" can be a combination of software and / or hardware that implements a predetermined function. Although the device described in the following embodiments is preferably implemented in software, the implementation of hardware, or a combination of software and hardware, is also possible and conceived.

[0144] Specifically, Fig. 9 This is a schematic diagram of the module structure of an embodiment of a business guide generating device provided in the embodiment of this specification, referring to Fig. 9 As shown, a business guidance generating device provided in an embodiment of this specification includes:

[0145] A receiving module 901 is used to receive a first requirement text input by a user and perform logic correction on the first requirement text;

[0146] An extraction module 902 is used to extract professional nouns and verbs related to the business from the corrected first requirement text, and combine the extracted professional nouns and verbs to obtain a business logic text;

[0147] The first generating module 903 is used to input the business logic text into a pre-trained business guidance generating model to obtain a first business guidance text and a first business handling entry link;

[0148] The second generating module 904 is used to receive user feedback on the first business guidance text. If the feedback is that the requirements are not met, the second requirement text input by the user is received again, and the above steps are repeated to generate a second business guidance text and a second business handling entry link according to the second requirement text by using the real-time tuning mechanism in the business guidance generating model;

[0149] The model optimization module 905 is used to extract the operation parameters and business processing result parameters of the user's business processing according to the first business guidance text and the first business processing entry link if the feedback is that the requirements are met, and use the operation parameters and business processing result parameters as training samples to train the business guidance generation model, and use the trained business guidance generation model as a new business guidance generation model to generate business guidance text.

[0150] The beneficial effects obtained by the device provided in the embodiments of this specification are consistent with the beneficial effects obtained by the above method and will not be repeated here.

[0151] Reference Fig.10 As shown, based on the above-mentioned method for generating a business guide, a computer device 1002 is also provided in an embodiment of this specification, wherein the above-mentioned method runs on the computer device 1002. The computer device 1002 may include one or more processors 1004, such as one or more central processing units (CPUs), each of which may implement one or more hardware threads. The computer device 1002 may also include any memory 1006, which is used to store any kind of information such as code, settings, data, etc. Non-limitingly, for example, the memory 1006 may include any one or more combinations of the following: any type of RAM, any type of ROM, flash memory device, hard disk, optical disk, etc. More generally, any memory may use any technology to store information. Further, any memory may provide volatile or non-volatile retention of information. Further, any memory may represent a fixed or removable component of the computer device 1002. In one case, when the processor 1004 executes an associated instruction stored in any memory or combination of memories, the computer device 1002 may perform any operation of the associated instruction. The computer device 1002 also includes one or more drive mechanisms 1008 for interacting with any storage, such as a hard disk drive mechanism, an optical disk drive mechanism, etc.

[0152] The computer device 1002 may also include an input / output module 1010 (I / O) for receiving various inputs (via input devices 1012) and for providing various outputs (via output devices 1014). A specific output mechanism may include a presentation device 1016 and an associated graphical user interface (GUI) 1018. In other embodiments, the input / output module 1010 (I / O), the input device 1012, and the output device 1014 may not be included, and the computer device 1002 may be used as a computer device in a network. The computer device 1002 may also include one or more network interfaces 1020 for exchanging data with other devices via one or more communication links 1022. One or more communication buses 1024 couple the components described above together.

[0153] The communication link 1022 may be implemented in any manner, for example, through a local area network, a wide area network (e.g., the Internet), a point-to-point connection, etc., or any combination thereof. The communication link 1022 may include any combination of hardwired links, wireless links, routers, gateway functions, name servers, etc. governed by any protocol or combination of protocols.

[0154] Corresponding to Figures 1 to 8 In addition to the method shown, an embodiment of the specification also provides a computer-readable storage medium, on which a computer program is stored, and when the computer program is executed by a processor, the steps of the above method are executed.

[0155] The embodiment of the present specification also provides a computer-readable instruction, wherein when the processor executes the instruction, the program therein causes the processor to execute the following Figures 1 to 8 The method shown.

[0156] The embodiments of the present specification also provide a computer program product, including at least one instruction or at least one program, wherein the at least one instruction or the at least one program is loaded and executed by a processor to implement the following Figures 1 to 8 The method shown.

[0157] It should be understood that in the various embodiments of this specification, the size of the serial numbers of the above-mentioned processes does not mean the order of execution. The execution order of each process should be determined by its function and internal logic, and should not constitute any limitation on the implementation process of the embodiments of this specification.

[0158] It should also be understood that in the embodiments of this specification, the term "and / or" is only a description of the association relationship of the associated objects, indicating that three relationships can exist. For example, A and / or B can represent: A exists alone, A and B exist at the same time, and B exists alone. In addition, the character " / " in this specification generally indicates that the associated objects before and after are in an "or" relationship.

[0159] Those of ordinary skill in the art will appreciate that the units and algorithm steps of each example described in conjunction with the embodiments disclosed in this specification can be implemented in electronic hardware, computer software, or a combination of the two. In order to clearly illustrate the interchangeability of hardware and software, the composition and steps of each example have been generally described in the above description according to function. Whether these functions are performed in hardware or software depends on the specific application and design constraints of the technical solution. Professional and technical personnel can use different methods to implement the described functions for each specific application, but such implementation should not be considered to be beyond the scope of this specification.

[0160] Those skilled in the art can clearly understand that, for the convenience and brevity of description, the specific working processes of the systems, devices and units described above can refer to the corresponding processes in the aforementioned method embodiments and will not be repeated here.

[0161] In the several embodiments provided in this specification, it should be understood that the disclosed systems, devices and methods can be implemented in other ways. For example, the device embodiments described above are only schematic. For example, the division of the units is only a logical function division. There may be other division methods in actual implementation, such as multiple units or components can be combined or integrated into another system, or some features can be ignored or not executed. In addition, the mutual coupling or direct coupling or communication connection shown or discussed can be an indirect coupling or communication connection through some interfaces, devices or units, or it can be an electrical, mechanical or other form of connection.

[0162] The units described as separate components may or may not be physically separated, and the components shown as units may or may not be physical units, that is, they may be located in one place or distributed on multiple network units. Some or all of the units may be selected according to actual needs to achieve the purpose of the embodiments of this specification.

[0163] In addition, each functional unit in each embodiment of this specification may be integrated into one processing unit, or each unit may exist physically separately, or two or more units may be integrated into one unit. The above integrated unit may be implemented in the form of hardware or in the form of software functional units.

[0164] If the integrated unit is implemented in the form of a software functional unit and sold or used as an independent product, it can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of this specification is essentially or the part that contributes to the prior art, or all or part of the technical solution can be embodied in the form of a software product, and the computer software product is stored in a storage medium, including a number of instructions to enable a computer device (which can be a personal computer, a server, or a network device, etc.) to perform all or part of the steps of the method described in each embodiment of this specification. The aforementioned storage medium includes: U disk, mobile hard disk, read-only memory (ROM, Read-Only Memory), random access memory (RAM, Random Access Memory), disk or optical disk and other media that can store program codes.

[0165] Specific embodiments are used in this specification to illustrate the principles and implementation methods of this specification. The description of the above embodiments is only used to help understand the methods and core ideas of this specification. At the same time, for those skilled in the art, according to the ideas of this specification, there will be changes in the specific implementation methods and application scope. In summary, the content of this specification should not be understood as a limitation on this specification.

Claims

1. A method for generating a business guide, characterized in that: The method comprises: receiving a first requirement text input by a user, and performing logic correction on the first requirement text; Extracting professional nouns and verbs related to the business from the revised first requirement text, and combining the extracted professional nouns and verbs to obtain a business logic text; Input the business logic text into a pre-trained business guidance generation model to obtain a first business guidance text and a first business handling entry link; Receive user feedback on the first business guidance text. If the feedback is that the requirements are not met, re-receive a second requirement text input by the user, and repeat the above steps to generate a second business guidance text and a second business handling entry link according to the second requirement text using the real-time tuning mechanism in the business guidance generation model; If the feedback is that the requirements are met, the operation parameters and business processing result parameters of the user's business processing according to the first business guidance text and the first business processing entry link are extracted, and the operation parameters and business processing result parameters are used as training samples to train the business guidance generation model, and the trained business guidance generation model is used as a new business guidance generation model to generate business guidance text.

2. The method according to claim 1, characterized in that The logical correction of the first requirement text includes: Preprocessing the first requirement text to obtain a preprocessed first requirement text; Performing semantic recognition on the preprocessed first requirement text to obtain semantic representation; Inputting the semantic representation into a pre-trained logic detection model to obtain a logic error identifier in the pre-processed first requirement text; A corresponding logic correction model is selected according to the logic error identifier to correct the preprocessed first requirement text to obtain a corrected first requirement text.

3. The method according to claim 1, characterized in that The extracting of professional terms and verbs related to the business from the revised first requirement text includes: Performing word segmentation processing on the modified first requirement text to obtain a word segmentation list, wherein the word segmentation list includes position information of each word segmentation in the first requirement text; Performing part-of-speech tagging on each word in the word list to obtain the part-of-speech of each word; Filtering the word segmentation whose part of speech is noun from the word segmentation list, calculating the TF-IDF value of each noun, and obtaining an initial professional noun list according to the TF-IDF value; Sorting the initial professional noun list according to the pre-built domain knowledge base, screening the initial professional nouns whose rankings are within the preset value, and obtaining a final professional noun list; The word list is filtered according to the final professional noun list, and the word list is obtained by filtering the word list with the word part of verb.

4. The method according to claim 3, characterized in that The extracted professional nouns and verbs are combined to obtain business logic text, including: Performing semantic recognition on the extracted professional nouns and verbs to obtain semantic features of each professional noun and verb; Obtain the position information of each professional noun and verb in the first requirement text; The semantic features and corresponding position information of each professional noun and verb are fused to obtain a feature vector, which is input into a pre-trained business logic text generation model to obtain a business logic text.

5. The method according to claim 1, characterized in that The business guidance generation model includes a business guidance text generation sub-model and a business handling entry link generation sub-model; The step of inputting the business logic text into a pre-trained business guidance generation model to obtain a first business guidance text and a first business handling entry link includes: The business guidance text generation sub-model receives the business logic text, and generates a first business guidance text according to the business logic text; The business processing entry link generation sub-model receives the first business guidance text, generates a business process according to the first business guidance text, obtains the business identifier corresponding to each business process, and queries the business processing system database according to the business identifier to obtain the corresponding first processing entry link.

6. The method according to claim 1, characterized in that The generating a second business guidance text and a second business handling entry link according to the second requirement text by using the real-time tuning mechanism in the business guidance generation model includes: Adjusting the model parameters and / or model structure of the business guidance generation model according to the second requirement text to obtain a fine-tuned business guidance generation model; Using the test data set to perform a performance evaluation on the fine-tuned business guidance generation model, and judging whether the fine-tuned business guidance generation model meets the preset standard according to the evaluation result; If yes, input the second requirement text into the fine-tuned business guidance generation model to generate a second business guidance text and a second business handling entry link; If not, the model parameters and / or model structure of the business guidance generation model are readjusted until the fine-tuned business guidance generation model meets the preset standards.

7. The method according to claim 1, characterized in that The using the operation parameters and the business processing result parameters as training samples to train the business guidance generation model includes: Calculate the effective value of the operation parameter and business processing result parameter of each processing step in the business guidance text according to the preset effective value calculation rules; Extracting business features and user features corresponding to each processing step in the business guidance text according to historical business processing data, wherein the business features include business type, complexity and processing success rate, and the user features include user historical behavior patterns; Associating the valid values ​​of the operation parameters and business processing result parameters of each processing step in the business guidance text with the business characteristics and user characteristics corresponding to each processing step to obtain a label corresponding to each processing step in the business guidance text; The business guidance generation model is trained according to each processing step and corresponding label in the business guidance text to obtain a trained business guidance generation model.

8. The method according to claim 7, characterized in that The operation parameters include click and return, the business processing result parameters include processing failure and processing success, and the effective value calculation rules include: For the multiple processing steps included in the business guidance text, an initial effective value is set for the first processing step. If it is detected that the user clicks on the next processing step, the effective value of the processing step is obtained by increasing the initial effective value according to a preset ratio. If it is detected that the user's operation for the next processing step is to return, then the effective value of the processing step is obtained by decreasing the initial effective value according to a preset ratio; If it is detected that the business processing result parameter is successfully processed, then a preset value is added to the valid value of the last processing step to obtain the valid value of the business processing result parameter; If it is detected that the service processing result parameter is a processing failure, the preset value is reduced based on the effective value of the last processing step to obtain the effective value of the service processing result parameter.

9. A business guidance generating device, characterized in that: The device comprises: A receiving module, used for receiving a first requirement text input by a user, and performing logic correction on the first requirement text; An extraction module, used to extract professional nouns and verbs related to the business from the revised first requirement text, and combine the extracted professional nouns and verbs to obtain a business logic text; A first generating module, used for inputting the business logic text into a pre-trained business guidance generating model to obtain a first business guidance text and a first business handling entry link; A second generation module is used to receive user feedback on the first business guidance text. If the feedback is that the first business guidance text does not meet the requirements, the second requirement text input by the user is received again, and the above steps are repeated to generate a second business guidance text and a second business handling entry link according to the second requirement text by using the real-time tuning mechanism in the business guidance generation model; The model optimization module is used to extract the operation parameters and business processing result parameters of the user's business processing according to the first business guidance text and the first business processing entry link if the feedback is that the requirements are met, and use the operation parameters and business processing result parameters as training samples to train the business guidance generation model, and use the trained business guidance generation model as a new business guidance generation model to generate business guidance text.

10. A computer device comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, characterized in that: When the processor executes the computer program, the method according to any one of claims 1 to 8 is implemented.

11. A computer-readable storage medium, characterized in that: The computer-readable storage medium stores a computer program, and when the computer program is executed by a processor, the method according to any one of claims 1 to 8 is implemented.

12. A computer program product, characterized in that The method comprises at least one instruction or at least one program, wherein the at least one instruction or the at least one program is loaded and executed by a processor to implement the method according to any one of claims 1 to 8.

Images