Service providing method and apparatus, service request processing method
By constructing field association information and determining the filling order in the domain model, and using a large language model to process the initial text, the accuracy and completeness issues of field filling in the domain model are solved, achieving efficient and accurate business provision and request processing.
Patent Information
- Application Number
- CN202511134273.X
- Authority / Receiving Office
- CN · China
- Patent Type
- Patents(China)
- Current Assignee / Owner
- Filing Date
- 2025-08-14
- Publication Date
- 2025-11-21
- Estimated Expiration
- 2045-08-14
AI Technical Summary
In existing technologies, the accuracy and completeness of the field population process in domain models are difficult to guarantee, especially when users input incomplete initial text, which can lead to errors in the execution of business processes.
By pre-building the association information between the fields to be filled in the domain model, the filling order is determined, and the field values are extracted or deduced from the initial text using a large language model, ensuring the accuracy and completeness of field filling.
It improves the accuracy of field population, lowers the knowledge threshold for users, and allows users to complete business requests without having to enter the complete text of all field values, thus enhancing the user experience.
Smart Images

Figure CN120633609B_ABST
Abstract
Description
TECHNICAL FIELD
[0001] Embodiments of the present specification relate to the technical field of artificial intelligence, and in particular to a business providing method and apparatus, a business request processing method. BACKGROUND
[0002] A domain model is an abstract representation of business rules and logic in a specific business domain, and the domain model can describe entities and relationships between entities in the business domain through objects. A field is one of the basic components of an object, and the field is used to describe the specific characteristics of the object in the domain model. For example, in a transaction business, the objects of the domain model can include entities such as orders and accounts, the fields corresponding to the order object can include order numbers, order creation times, etc., and the fields corresponding to the account object can include account identifiers, account credit levels, account types, etc.
[0003] Generally, when a user has a business requirement, the fields in the domain model corresponding to the business should be filled in first, that is, the field values of the fields in the domain model need to be determined, and after the field filling is completed, the domain model can be called and processed by subsequent business logic, thereby providing the user with the business. It can be seen that accurate field filling of the domain model is an important prerequisite for correct execution of the business process. SUMMARY
[0004] Therefore, an embodiment of the present specification provides a business providing method. One or more embodiments of the present specification also relate to a business request processing method, a business providing apparatus, a computing device, a computer-readable storage medium, and a computer program product to solve the technical defects in the prior art.
[0005] According to a first aspect of an embodiment of the present specification, a business providing method is provided, including:
[0006] obtaining an initial text, wherein the initial text is a description text of a target business;
[0007] determining field association information corresponding to the initial text and an initial domain model, wherein the initial domain model includes a plurality of fields to be filled, and the field association information includes an association relationship between the fields to be filled;
[0008] determining a filling order according to the field association information, filling the fields to be filled in the initial domain model according to the initial text in the filling order, and obtaining a target domain model, wherein the filling order includes at least one field to be filled;
[0009] providing the target business based on the target domain model.
[0010] According to a second aspect of the embodiments of the present specification, a service request processing method is provided, the method is applied to a cloud-side device, and includes:
[0011] receiving a service request sent by an end-side device, wherein the service request carries an initial text;
[0012] determining field association information corresponding to the service request and an initial domain model, wherein the initial domain model includes a plurality of fields to be filled, and the field association information includes an association relationship between the fields to be filled;
[0013] determining a filling order according to the field association information, filling the fields to be filled in the initial domain model according to the initial text in the filling order, and obtaining a target domain model, wherein the filling order includes at least one field to be filled;
[0014] processing the service request based on the target domain model, obtaining a request processing result corresponding to the service request, and sending the request processing result to the end-side device.
[0015] According to a third aspect of the embodiments of the present specification, a service providing apparatus is provided, including:
[0016] an obtaining module configured to obtain an initial text, wherein the initial text is a description text of a target service;
[0017] a first determining module configured to determine field association information corresponding to the initial text and an initial domain model, wherein the initial domain model includes a plurality of fields to be filled, and the field association information includes an association relationship between the fields to be filled;
[0018] a first filling module configured to determine a filling order according to the field association information, fill the fields to be filled in the initial domain model according to the initial text in the filling order, and obtain a target domain model, wherein the filling order includes at least one field to be filled;
[0019] a providing module configured to provide the target service based on the target domain model.
[0020] According to a fourth aspect of the embodiments of the present specification, a computing device is provided, including:
[0021] a memory and a processor;
[0022] The memory is used to store computer programs / instructions, and the processor is used to execute the computer programs / instructions, which realize the steps of the above method when executed by the processor.
[0023] According to a fifth aspect of the embodiments of the present specification, a computer readable storage medium is provided, which stores computer programs / instructions, which, when executed by a processor, implement the steps of the above method.
[0024] According to a sixth aspect of the embodiments of the present specification, a computer program product is provided, which includes computer programs / instructions, which, when executed by a processor, implement the steps of the above method.
[0025] One embodiment of the present specification can obtain an initial text, the initial text being a description text of a requested target service; determine field association information corresponding to the initial text and an initial domain model, the initial domain model including a plurality of to-be-filled fields, and the field association information including an association relationship between the to-be-filled fields; determine a filling order according to the field association information, and fill the to-be-filled fields in the initial domain model according to the initial text in the filling order to obtain a target domain model, wherein the filling order includes at least one to-be-filled field; and provide the target service based on the target domain model.
[0026] One embodiment of the present specification fills the fields in the domain model corresponding to the initial text in the filling order determined by the field association information corresponding to the initial text, which can avoid missing the to-be-filled fields and improve the accuracy of field filling. Moreover, since the field association information is the association information between the to-be-filled fields, in the case where the initial text does not include the field values of all the to-be-filled fields, the field values of the to-be-filled fields not included in the initial text can be determined based on the field association information, and the completeness of field filling can still be ensured. That is, the foregoing method supports receiving an initial text that does not contain the field values of all the to-be-filled fields, i.e., an incomplete initial text. In actual application, a user does not need to input a complete text containing the field values of all the to-be-filled fields, but can input a simple demand text, which reduces the user knowledge threshold and improves user experience. BRIEF DESCRIPTION OF DRAWINGS
[0027] Figure 1 is a flowchart of a service providing method provided by one embodiment of the present specification;
[0028] Figure 2 is a schematic diagram of an association relationship provided by one embodiment of the present specification;
[0029] Figure 3 is a flowchart of a financial resource borrowing service providing method provided by one embodiment of the present specification;
[0030] Figure 4 is a flowchart of a service request processing method provided by one embodiment of the present specification;
[0031] Figure 5 is a structural schematic diagram of a service providing apparatus provided by an embodiment of the present specification;
[0032] Figure 6 is a structural schematic diagram of a service request processing apparatus provided by an embodiment of the present specification;
[0033] Figure 7 is an architecture diagram of a service request processing system provided by an embodiment of the present specification;
[0034] Figure 8 is a structural block diagram of a computing device provided by an embodiment of the present specification. DETAILED DESCRIPTION
[0035] In the following description, numerous specific details are set forth in order to provide a thorough understanding of the present specification. However, the present specification can be practiced without the specific details, other than in the examples described herein. In other instances, well-known methods, procedures, and components have not been described in detail so as not to unnecessarily obscure aspects of the present specification.
[0036] The terminology used in one or more embodiments of the present specification is for the purpose of describing particular embodiments only and is not intended to be limiting of one or more embodiments of the present specification. As used in one or more embodiments of the present specification and the accompanying claims, the singular forms "a," "an," and "the" are intended to include the plural forms as well, unless the context clearly indicates otherwise. It will be further understood that the terms "comprises" and / or "comprising," when used in one or more embodiments of the present specification, specify the presence of stated features, integers, steps, operations, elements, and / or components, but do not preclude the presence or addition of one or more other features, integers, steps, operations, elements, components, and / or groups thereof.
[0037] It will be understood that, although the terms first, second, etc. can be used herein to describe various information, these terms are not intended to denote a temporal or chronological order. Rather, these terms are used solely to distinguish one from another only. For example, without departing from the scope of one or more embodiments of the present specification, first can be termed second; likewise, second can be termed first. The term "if' as used herein, can be interpreted as meaning "when" or "in response to determining" depending on the context.
[0038] 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 for analysis, stored data, displayed data, etc.) involved in the present specification are all information and data authorized by the user or authorized by all parties, and the collection, use and processing of related data need to comply with relevant laws, regulations and standards in relevant regions, and provide corresponding operation portal for user to choose authorization or refusal.
[0039] In one or more embodiments of the present specification, a large model refers to a deep learning model with a large number of model parameters, usually containing hundreds of millions, billions, tens of billions, hundreds of billions, or even tens of billions of model parameters. The large model can also be referred to as a foundation model. Through large-scale unlabeled corpus pre-training of the large model, a pre-trained model with hundreds of millions of parameters is output. Such a model can adapt to a wide range of downstream tasks, and the model has good generalization ability. For example, a large language model (LLM) and a multi-modal pre-training model.
[0040] In actual application, the large model only needs a small amount of samples to fine-tune the pre-trained model and can be applied to different tasks. The large model can be widely applied to natural language processing (NLP) and computer vision fields. Specifically, it can be applied to computer vision field tasks such as visual question answering (VQA), image captioning (IC), image generation, and natural language processing field tasks such as text-based sentiment classification, text summary generation, and machine translation. The main application scenarios of the large model include digital assistants, intelligent robots, search, online education, office software, e-commerce, intelligent design, etc.
[0041] As described in the background, a domain model is an abstract representation of core concepts, rules, entities, and their relationships in a specific business domain. The domain model describes the key objects involved in the business domain and also expresses the interaction mode and business rules between these objects. The domain model is the core of business logic and is the bridge connecting business requirements and technical implementation. In actual application, in order to ensure that the business process can be executed correctly, effective data filling needs to be provided for the fields in the domain model.
[0042] In the present specification, a business providing method is provided. The present specification also relates to a business request processing method, a business providing device, a computing device, a computer readable storage medium, and a computer program product. The embodiments are described in detail one by one below.
[0043] Referring to Figure 1 , Figure 1 A flowchart of a business providing method according to an embodiment of the present specification is shown, which specifically includes the following steps.
[0044] Step 102: obtaining an initial text, wherein the initial text is a description text of a target service requested by a user.
[0045] The execution subject of the technical solution of the present application specification can be any computing device with computing capability, such as a terminal, a server, etc., and the present specification does not make specific limitations.
[0046] In actual application, the computing device can provide at least one service to the user, and the present specification does not limit the specific type and quantity of the service provided by the computing device. For example, the service provided includes but is not limited to transaction service, resource borrowing service, learning guidance service, and industry service, etc.
[0047] In one or more embodiments of the present specification, the initial text can be input by the user to the computing device, and the computing device can receive the initial text input by the user. Specifically, when the user has a service demand, the user can input the description text of the target service, i.e., the initial text, to the computing device, and the computing device can perform subsequent processing based on the initial text input by the user to provide the target service for the user.
[0048] It should be noted that the present specification does not limit the way in which the computing device obtains the initial text, i.e., the way in which the user requests the target service. In order to improve the convenience and speed of the user requesting the service, in the present specification, the user can request the target service through text input, voice input, etc.
[0049] In one or more embodiments of the present specification, the computing device can display a service page, and the service interface can include an input control for inputting text, and the user can input the description text of the target service, i.e., the initial text, in the input control, so that the computing device can obtain the initial text.
[0050] In one or more embodiments of the present specification, the computing device can further include a voice recognition module, and the user can issue a voice instruction through a semantic input method, and the voice instruction is used to instruct the computing device to provide the target service for the user, and the computing device can analyze the voice instruction through the voice recognition module to generate corresponding text content, thereby obtaining the initial text corresponding to the voice instruction.
[0051] Step 104: determining field association information corresponding to the initial text and an initial domain model, wherein the initial domain model includes a plurality of fields to be filled, and the field association information is the association information between the fields to be filled.
[0052] As described above, the computing device can provide at least one service for the user, such as a transaction service, a resource borrowing service, a learning assistance service, an outing service, and the like. In actual applications, a person skilled in the art can pre-construct a domain model corresponding to each service, such as a domain model corresponding to a transaction service, a domain model corresponding to a resource borrowing service, a domain model corresponding to a learning assistance service, and a domain model corresponding to an outing service.
[0053] Further, in order to meet the diversified needs of users, for each service, the service can be divided into a plurality of sub-services based on different division rules, and a corresponding domain model can be constructed for each sub-service. For example, for a transaction service, the transaction service can be divided into a cross-border transaction sub-service and an in-country transaction sub-service based on regions, and corresponding domain models can be constructed respectively. The transaction service can also be divided into a risky transaction sub-service and a non-risky transaction sub-service based on transaction amounts, and corresponding domain models can be constructed respectively. Similarly, for a resource borrowing service, the resource borrowing service can be divided into a financial resource borrowing service and a storage resource borrowing service based on resource types, and corresponding domain models can be constructed respectively. In actual applications, each service can be divided and each domain model can be constructed based on actual needs, and the present specification does not make specific limitations.
[0054] It should be noted that how to construct a domain model corresponding to a service is a relatively mature technology at present, and the present specification does not make specific limitations.
[0055] In actual applications, the domain model includes a plurality of fields to be filled. When providing a service for a user, the computing device needs to determine a domain model corresponding to a service requested by the user, fill the fields in the domain model corresponding to the service requested by the user, and then provide the service requested by the user based on the filled domain model.
[0056] It should be noted that the fields to be filled in the domain model are pre-determined, and the fields to be filled can be at least part of the fields in the domain model or all of the fields in the domain model. The fields to be filled in the domain model can be pre-determined based on actual needs.
[0057] At present, with the rapid development of large language models (LLM), it has become a key way to assign values to fields in a domain model through a large language model. However, with the continuous optimization and upgrading of business systems and the continuous updating of user needs, the number of entities in the business is increasing, and therefore the number of fields in the domain model is increasing, making it difficult for the LLM to accurately assign values to each field.
[0058] In the present specification, in order to improve the accuracy of field filling, the association information between the fields to be filled in corresponding to each domain model can be constructed in advance. For example, for a transaction business, the fields can include an order number, an order creation time, etc. Generally, an order number can be assigned only after an order is created, so the association information between the order number and the order creation time can be that the order number is generated after the order creation time. For another example, for a financial resource borrowing business, the fields can include a financial resource borrowing target and a financial resource borrowing limit, etc. Generally, the financial resource borrowing target is different, and the financial resource borrowing target is the service provider corresponding to the financial resource borrowing business, and the financial resource borrowing limit is different, so the association information between the financial resource borrowing target and the financial resource borrowing limit can be that the financial resource borrowing limit is determined according to the financial resource borrowing target.
[0059] In one or more embodiments of the present specification, the fields to be filled in each domain model have field association information, which is the association information between the fields to be filled in the domain model. For example, there are two domain models, domain model A and domain model B, then the association information between the fields to be filled in domain model A can be constructed in advance, and the association information between the fields to be filled in domain model B can be constructed in advance.
[0060] It should be noted that the field association information can be the association information between part of the fields to be filled in the domain model, or the association information between all the fields to be filled in the domain model, which can be determined according to actual needs.
[0061] In addition, the present specification does not limit the specific representation form of the association information. In actual application, the representation form of the association information includes but is not limited to a graph structure, a linked list structure, and a text description, etc. For example, for the above-mentioned transaction business and financial resource borrowing business, the present specification takes a directed graph in a graph structure as an example, and provides a schematic diagram of an association relationship, which can be specifically referred to as Figure 2 . Referring to Figure 2 , Figure 2 a schematic diagram of an association relationship provided by one embodiment of the present specification.
[0062] In the present specification, after the domain model and the field association information corresponding to the domain model are constructed in advance, the computing device can determine the field association information corresponding to the initial text and the initial domain model after the initial text is obtained.
[0063] In actual applications, when determining the field association information corresponding to the initial text and the initial domain model, the target business corresponding to the initial text can be determined first, and the initial domain model and the field association information can be determined according to the target business. For example, the computing device can input the initial text into a machine learning model such as a keyword extraction model or an intent recognition model, so as to obtain the target business corresponding to the initial text output by the machine learning model. The computing device can match the corresponding initial domain model according to the target business, and obtain the field association information corresponding to the initial domain model.
[0064] Step 106: determining a filling order according to the field association information, and filling the fields to be filled in the initial domain model according to the initial text in the filling order to obtain a target domain model, wherein the filling order includes at least one field to be filled.
[0065] In actual applications, the computing device can fill the fields to be filled in the initial domain model according to the initial text based on the association information between the fields to be filled provided by the field association information through a pre-trained large language model, so as to obtain a target domain model. Specifically, the computing device can determine a current field to be filled according to the association information between the fields to be filled provided by the field association information, and extract the field value of the current field to be filled from the initial text through the large language model.
[0066] Since there is a dependency relationship between different fields, the dependency relationship between the fields, i.e., the field association information, can provide guidance information for the large language model, thereby improving the accuracy of the filled field values.
[0067] In one or more embodiments of the present specification, the filling order can be determined according to the field association information. The filling order refers to the filling order of the fields to be filled, and the filling order includes at least one field to be filled. For example, the field association information is A including B, and B and C have an association relationship, so the filling order can be A, B, and C. How to determine the filling order according to the field association information is not specifically limited in the present specification, for example, the order of the fields with less association relationship can be arranged in front of the fields with more association relationship. The filling order can also be determined according to the field association information based on actual business needs.
[0068] Filling the fields to be filled in the initial domain model according to the initial text in the filling order includes:
[0069] Determining a current field to be filled according to the filling order;
[0070] Determining the field value of the current field to be filled according to the initial text, and filling the current field to be filled in the initial domain model according to the field value.
[0071] In one or more embodiments of the present specification, in order to improve the accuracy of the field value determined by the LLM to improve the accuracy of field filling, the current field to be filled can be one, that is, the filling order is the filling order of each field to be filled.
[0072] In one or more embodiments of the present specification, the filling order can be constructed as a linked list, and each element in the linked list is a field to be filled. In turn, each field to be filled can be filled in order according to the linked list. Of course, the filling order can also be constructed as a graph structure or a text description, and the present specification does not make specific limitations.
[0073] In the present specification, determining the field value of the current field to be filled according to the initial text comprises:
[0074] determining whether the initial text includes the field value corresponding to the current field to be filled;
[0075] if yes, extracting the field value corresponding to the current field to be filled from the initial text;
[0076] if no, obtaining the field value corresponding to the current field to be filled according to a preset condition corresponding to the current field to be filled.
[0077] As can be seen, when the field value corresponding to the current field to be filled can be determined from the initial text, that is, the initial text includes the field value corresponding to the current field to be filled, the field value corresponding to the current field to be filled can be extracted from the initial text. When the field value corresponding to the current field to be filled cannot be determined from the initial text, that is, the initial text does not include the field value corresponding to the current field to be filled, the field value corresponding to the current field to be filled can be obtained according to a preset condition corresponding to the current field to be filled.
[0078] It should be noted that extracting the field value corresponding to the current field to be filled from the initial text can use a key field extraction model, a large language model, etc., and the present specification does not make specific limitations.
[0079] Obtaining the field value corresponding to the current field to be filled according to the preset condition corresponding to the current field to be filled comprises:
[0080] in the case where the initial text satisfies the preset condition, obtaining the field value corresponding to the current field to be filled from the preset condition;
[0081] in the case where the initial text does not satisfy the preset condition, generating question information according to the current field to be filled, and obtaining the field value corresponding to the current field to be filled based on the reply information corresponding to the question information.
[0082] In one or more embodiments of the present specification, the preset condition can be whether there is a mapping relationship between the field value corresponding to the current to-be-filled field and the field value corresponding to the to-be-filled field contained in the initial text, if there is a mapping relationship, the preset condition is met, if there is no mapping relationship, the preset condition is not met.
[0083] The mapping relationship is pre-set, specifically, a pre-set mapping relationship between the field value corresponding to the current to-be-filled field and the field value corresponding to the to-be-filled field contained in the initial text. It should be understood that the mapping relationship can also reflect the mapping relationship between the current to-be-filled field and the to-be-filled field contained in the initial text, the mapping relationship between the field value corresponding to the current to-be-filled field and the to-be-filled field contained in the initial text, and the mapping relationship between the current to-be-filled field and the field value corresponding to the to-be-filled field contained in the initial text.
[0084] Therefore, when the initial text does not include the field value corresponding to the current to-be-filled field, the field value corresponding to the to-be-filled field contained in the initial text can be determined first, and then the target mapping relationship between the field value corresponding to the to-be-filled field contained in the initial text and the current to-be-filled field is determined from the pre-set mapping relationship, so that the field value corresponding to the current to-be-filled field in the target mapping relationship can be obtained. Of course, the target mapping relationship can be determined, that is, there is a target mapping relationship, which meets the preset condition, and the target mapping relationship cannot be determined, that is, there is no target mapping relationship, which does not meet the preset condition. The target mapping relationship is a mapping relationship between the current to-be-filled field and the field value corresponding to the to-be-filled field contained in the initial text, and the target mapping relationship includes the field value corresponding to the current to-be-filled field.
[0085] For example, in the field of storage resource borrowing, the fields to be filled include a storage resource borrowing service provider and a storage resource borrowing time limit. Assuming that the current field to be filled is the storage resource borrowing time limit, the initial text input by the user is: borrowing storage resources of service provider A, according to the initial text, it can be determined that the field to be filled contained in the initial text is: the storage resource borrowing service provider, and it is determined that the field value corresponding to the field to be filled contained in the initial text is: the storage resource service provider A. In practice, because different storage resource borrowing service providers set different storage resource borrowing time limits, a mapping relationship between different service providers and storage resource borrowing time limits can be pre-set, and it is further assumed that the mapping relationship set includes: the storage resource borrowing time limit corresponding to the storage resource service provider A is one week, and the storage resource borrowing time limit corresponding to the storage resource service provider B is one month. Although the initial text input by the user does not include the field value corresponding to the storage resource borrowing time limit, the field value corresponding to the field to be filled contained in the initial text and the target mapping relationship of the current field to be filled can be determined, that is, the target mapping relationship is: the storage resource borrowing time limit corresponding to the storage resource service provider A is one week, and then it can be obtained that the current field to be filled, the storage resource borrowing time limit, is one week.
[0086] Based on the foregoing method, without the user providing the field value corresponding to each field to be filled, the field values of part of the fields to be filled can be derived according to the field association information and the preset condition, thereby reducing the user knowledge threshold and improving the user experience.
[0087] In actual application, when the initial text does not satisfy the preset condition, the question information corresponding to the current field to be filled needs to be generated, the user can reply according to the question information, so that the computing device can obtain the reply information corresponding to the question information in response to the reply to the question information, and thus the field value corresponding to the current field to be filled can be extracted from the reply information.
[0088] It should be noted that, in the case that the field value corresponding to the current field to be filled still cannot be obtained based on the reply information corresponding to the question information, the field association information and the initial domain model can be re-determined according to the initial text and the reply information. That is, the user has an updated business intent, so it is necessary to re-determine the target business requested by the user based on the initial text and the reply text, and to re-determine the field association information and the initial domain model, and to provide the user with the re-determined target business according to the above business providing method.
[0089] Further, the field value corresponding to the current field to be filled is obtained according to the preset condition corresponding to the current field to be filled, including:
[0090] The field value corresponding to the current field to be filled is obtained according to the preset condition corresponding to the current field to be filled;
[0091] update the initial text according to the obtained field value corresponding to the current to-be-filled field.
[0092] In one or more embodiments of the present specification, since the field value of the to-be-filled field not included in the initial text needs to be determined based on the field value of the to-be-filled field included in the initial text, and since the to-be-filled fields in the present specification are filled in sequence according to the filling order determined based on the field association information, the initial text can be updated after obtaining the field value of the to-be-filled field not included in the initial text. The updating of the initial text can specifically be adding the determined field value of the to-be-filled field not included in the initial text to the initial text.
[0093] When filling each field value in sequence according to the filling order, the initial text can be dynamically updated, so that more context information, i.e., more field value information, can be provided for the determination of the next to-be-filled field value, thereby improving the accuracy of the determined next to-be-filled field value.
[0094] Step 108: providing the target service based on the target domain model.
[0095] In actual applications, after obtaining the field-filled target domain model, the target service can be provided based on the target domain model.
[0096] It should be noted that providing a service corresponding to a domain model for a user based on the domain model is a relatively mature technology at present, and the present specification does not make specific limitations. For example, after obtaining the target domain model, the computing device can call a business execution engine corresponding to the target service, and then the target domain model can be transmitted as a parameter into the business execution engine, and the business logic is executed to provide the target service for the user.
[0097] Based on the above service providing method, the fields in the domain model corresponding to the initial text are filled in the filling order determined based on the field association information corresponding to the initial text, which can avoid missing to-be-filled fields and improve the accuracy of field filling. Moreover, since the field association information is the association information between to-be-filled fields, in the case where the initial text does not include the field values of all to-be-filled fields, the field values of the to-be-filled fields not included in the initial text can be determined based on the field association information, and the completeness of field filling can still be ensured. That is, the foregoing method supports receiving an initial text that does not include the field values of all to-be-filled fields, i.e., an incomplete initial text. In actual applications, the user does not need to input a complete text including the field values of all to-be-filled fields, but only needs to input a simple demand text, which reduces the user knowledge threshold and improves the user experience.
[0098] Further, in the present specification, extracting the field value corresponding to the current to-be-filled field from the initial text comprises:
[0099] obtaining prompt text corresponding to the current to-be-filled field;
[0100] extracting the field value corresponding to the current to-be-filled field from the initial text according to the prompt text.
[0101] For each current to-be-filled field, the prompt text corresponding to the current to-be-filled field is dynamically set to prompt the large language model to extract the field value corresponding to the current to-be-filled field, thereby improving the accuracy of the extracted field value. When dynamically setting the prompt text corresponding to the current to-be-filled field, the prompt text corresponding to the current to-be-filled field can be generated based on a preset prompt framework.
[0102] In one or more embodiments of the present specification, the large language model for extracting the field value of the to-be-filled field from the initial text can be a model based on the Transformer architecture. In actual applications, the model based on the Transformer architecture has limited modeling capability for long-distance context dependency when processing long input sequences. Therefore, in the present specification, the extraction of the field value is performed field by field according to the filling order, which can reduce the interference of irrelevant information in the initial text, improve the ability of the large language model to capture and extract key context information in the initial text, thereby overcoming the limitations of the Transformer architecture in long-distance dependency modeling and alleviating the inaccuracy of obtaining long-distance knowledge under the transformer architecture.
[0103] In one or more embodiments of the present specification, the domain model can be a domain model based on JAVA engineering. In actual applications, the set operation of the java domain can be enhanced, so that the natural language requirements of the user can be received as parameter input.
[0104] The following describes the business providing method provided in the present specification with reference to the accompanying drawings. Figure 3 The business providing method provided in the present specification is further described by taking the application of the business providing method in the field of financial resource borrowing business as an example. Specifically, Figure 3 FIG. 1 shows a flowchart of a financial resource borrowing business providing method according to an embodiment of the present specification, which specifically comprises the following steps.
[0105] Step 302: obtaining an initial text, wherein the initial text is a description text of a request for a financial resource borrowing business.
[0106] Step 304: determining field association information corresponding to the initial text and an initial domain model, wherein the initial domain model comprises a plurality of to-be-filled fields, and the field association information is association information between the to-be-filled fields.
[0107] Step 306: determining a filling order according to the field association information, filling the fields to be filled in the initial domain model according to the initial text in the filling order, and obtaining a target domain model, wherein the filling order includes at least one field to be filled.
[0108] Step 308: providing the financial resource borrowing service based on the target domain model.
[0109] The financial resource borrowing service refers to a service activity in which a user borrows financial resources from a service provider to meet his / her needs in the financial field. In the financial resource borrowing service, the service provider can provide the user with a financial resource borrowing service, and the user can request the service provider for the financial resource borrowing service. The service provider can be a financial service platform, a bank, a financial institution, etc., and the financial resources include but are not limited to currency, bonds, stocks, etc.
[0110] In actual applications, when requesting the financial resource borrowing, the user can input an initial text in a device or a terminal application provided by the service provider and used to handle the financial resource borrowing service, where the initial text is a description text of the user's request for the financial resource borrowing service. Then, the server corresponding to the service provider or the device provided by the service provider can obtain the initial text. Further, the field association information corresponding to the initial text and the initial domain model can be determined, where the initial domain model includes a plurality of fields to be filled, and the field association information is the association information between the fields to be filled. Finally, the fields to be filled in the initial domain model can be filled according to the initial text and the field association information, and a target domain model can be obtained. The financial resource borrowing service can be provided based on the target domain model.
[0111] Of course, the initial domain model is a domain model corresponding to the financial resource borrowing service, which can be pre-set as described above, and the fields to be filled in the domain model corresponding to the financial resource borrowing service can also be pre-set, which will not be described here.
[0112] In addition, as described above, the financial resource borrowing service can also be divided into a plurality of sub-services based on more specific service requirements, such as a currency borrowing sub-service, a stock pledge sub-service, a bond purchase sub-service, etc. Then, a domain model corresponding to each sub-service can be set, and the fields to be filled in the domain model corresponding to each sub-service can be determined, which can be pre-set according to actual requirements, and will not be described here.
[0113] It should be noted that in this embodiment, the specific method for providing the user with the financial resource borrowing service based on the initial text is consistent with the above-mentioned service providing method, which can be referred to the above-mentioned service providing method, and will not be described here.
[0114] Based on the above financial resource borrowing business providing method, the fields in the domain model corresponding to the initial text are filled according to the filling order determined by the field association information corresponding to the initial text, so that the missing fields to be filled can be avoided, and the accuracy of field filling is improved. Moreover, since the field association information is the association information between the fields to be filled, in the case that the initial text does not include the field values of all the fields to be filled, the field values of the fields to be filled which are not included in the initial text can be determined based on the field association information, and the integrity of field filling can still be guaranteed. That is, the above method supports receiving an initial text which does not include the field values of all the fields to be filled, that is, an incomplete initial text. In actual application, the user does not need to input a complete text which includes the field values of all the fields to be filled, but can input a simple demand text, which reduces the user knowledge threshold and improves the user experience.
[0115] The present specification also provides an embodiment of a processing method of a service request, Figure 4 A flow chart of a processing method of a service request is shown in an embodiment of the present specification. As shown in the figure, Figure 4 The processing method of the service request includes:
[0116] Step 402: receiving a service request sent by an end-side device, wherein the service request carries an initial text.
[0117] Step 404: determining field association information corresponding to the service request and an initial domain model, wherein the initial domain model includes a plurality of fields to be filled, and the field association information is the association information between the fields to be filled.
[0118] Step 406: determining a filling order according to the field association information, and filling the fields to be filled in the initial domain model according to the initial text according to the filling order to obtain a target domain model, wherein the filling order includes at least one field to be filled.
[0119] Step 408: processing the service request based on the target domain model to obtain a request processing result corresponding to the service request, and sending the request processing result to the end-side device.
[0120] In actual application, when a user has a demand, the user can input an initial text on an end-side device. After the end-side device receives the initial text input by the user, the end-side device can send a service request carrying the initial text to a cloud-side device.
[0121] The cloud-side device can receive a service request sent by the end-side device, the service request carrying an initial text. The cloud-side device determines field association information corresponding to the service request and an initial domain model, wherein the initial domain model includes a plurality of fields to be filled, and the field association information is association information between the fields to be filled. According to the field association information, a filling order is determined, and the fields to be filled in the initial domain model are filled according to the initial text in the filling order, to obtain a target domain model, wherein the filling order includes at least one field to be filled. The service request is processed based on the target domain model, to obtain a request processing result corresponding to the service request, and the request processing result is sent to the end-side device. Thus, the end-side device can provide a target service to the user based on the request processing result.
[0122] It should be noted that the process of obtaining the target domain model according to the initial text in the service request is consistent with the above-mentioned service providing method, and will not be described in detail here.
[0123] Based on the above-mentioned service request processing method, the fields in the domain model corresponding to the initial text can be filled according to the filling order determined by the field association information corresponding to the initial text, so that the fields to be filled can be avoided to be missed, and the accuracy of field filling is improved. Moreover, since the field association information is association information between the fields to be filled, in the case that the initial text does not include field values of all the fields to be filled, the field values of the fields to be filled that are not included in the initial text can be determined based on the field association information, so that the completeness of field filling can still be ensured. That is, the above-mentioned method supports receiving an initial text that does not contain field values of all the fields to be filled, i.e., an incomplete initial text. In actual application, the user does not need to input a complete text containing field values of all the fields to be filled, but can input a simple demand text, which reduces the user knowledge threshold and improves the user experience.
[0124] Corresponding to the above-mentioned service providing method embodiment, the present specification also provides a service providing device embodiment, Figure 5 A structure schematic diagram of a service providing device provided by an embodiment of the present specification is shown. As shown in the figure, Figure 5 The device includes:
[0125] The obtaining module 502 is configured to obtain an initial text, wherein the initial text is a description text of a target service;
[0126] The first determining module 504 is configured to determine field association information corresponding to the initial text and an initial domain model, wherein the initial domain model includes a plurality of fields to be filled, and the field association information is association information between the fields to be filled;
[0127] The first filling module 506 is configured to determine a filling sequence according to the field association information, fill the initial field model according to the initial text in the filling sequence, and obtain a target field model, wherein the filling sequence includes at least one field to be filled.
[0128] The providing module 508 is configured to provide the target service based on the target field model.
[0129] Optionally, the first determining module 504 is further configured to determine a target service corresponding to the initial text, and determine the field association information and the initial field model according to the target service.
[0130] Optionally, the first filling module 506 is further configured to determine a current field to be filled according to the filling sequence, determine a field value of the current field to be filled according to the initial text, and fill the current field to be filled in the initial field model according to the field value.
[0131] Optionally, the first filling module 506 is further configured to determine whether the initial text includes a field value corresponding to the current field to be filled, extract the field value corresponding to the current field to be filled from the initial text if yes, and obtain the field value corresponding to the current field to be filled according to a preset condition corresponding to the current field to be filled if no.
[0132] Optionally, the first filling module 506 is further configured to obtain the field value corresponding to the current field to be filled from the preset condition in a case where the initial text satisfies the preset condition, and obtain the field value corresponding to the current field to be filled according to reply information corresponding to question information of the current field to be filled in a case where the initial text does not satisfy the preset condition.
[0133] Optionally, the first filling module 506 is further configured to obtain the field value corresponding to the current field to be filled according to a preset condition corresponding to the current field to be filled, and update the initial text according to the obtained field value corresponding to the current field to be filled.
[0134] Optionally, the first filling module 506 is further configured to obtain prompt text corresponding to the current field to be filled, and extract the field value corresponding to the current field to be filled from the initial text according to the prompt text.
[0135] Optionally, the first filling module 506 is further configured to, in a case where the reply information corresponding to the question information cannot obtain the field value corresponding to the current field to be filled, determine the field association information and the initial domain model again according to the initial text and the reply information.
[0136] The above is a schematic scheme of the service providing apparatus in this embodiment. It should be noted that the technical scheme of the service providing apparatus and the technical scheme of the service providing method described above belong to the same concept, and the details of the technical scheme of the service providing apparatus that are not described in detail can be referred to the description of the technical scheme of the service providing method.
[0137] Based on the above service providing apparatus, the fields in the domain model corresponding to the initial text can be filled in the filling order determined by the field association information corresponding to the initial text, the missing fields to be filled can be avoided, and the accuracy of field filling is improved. Moreover, since the field association information is the association information between the fields to be filled, in a case where the initial text does not include the field values of all the fields to be filled, the field values of the fields to be filled that are not included in the initial text can be determined based on the field association information, and the completeness of field filling can still be ensured. That is, the above method supports receiving an initial text that does not contain the field values of all the fields to be filled, i.e., an incomplete initial text. In actual application, the user does not need to input a complete text containing the field values of all the fields to be filled, but can input a simple demand text, which reduces the user knowledge threshold and improves the user experience.
[0138] Corresponding to the above service request processing method embodiment, the present specification also provides a service request processing apparatus embodiment, Figure 6 A structural schematic diagram of a service request processing apparatus provided by one embodiment of the present specification is shown. As shown in the figure, Figure 6 The apparatus includes:
[0139] The receiving module 602 is configured to receive a service request sent by an end-side device, wherein the service request carries an initial text;
[0140] The second determining module 604 is configured to determine field association information corresponding to the service request and an initial domain model, wherein the initial domain model includes a plurality of fields to be filled, and the field association information is the association information between the fields to be filled;
[0141] The second filling module 606 is configured to determine a filling order according to the field association information, fill the fields to be filled in the initial domain model according to the initial text in the filling order, and obtain a target domain model, wherein the filling order includes at least one field to be filled;
[0142] The processing module 608 is configured to process the service request based on the target domain model, obtain a request processing result corresponding to the service request, and send the request processing result to the terminal device.
[0143] The above is a schematic scheme of the service request processing apparatus of the embodiment. It should be noted that the technical scheme of the service request processing apparatus and the technical scheme of the service request processing method described above belong to the same concept, and the details of the technical scheme of the service request processing apparatus that are not described in detail can be referred to the description of the technical scheme of the service request processing method.
[0144] Based on the service request processing apparatus, the fields in the domain model corresponding to the initial text can be filled in the filling order determined based on the field association information corresponding to the initial text, the missing fields to be filled in can be avoided, and the accuracy of field filling is improved. Moreover, since the field association information is the association information between the fields to be filled in, in the case that the initial text does not include the field values of all the fields to be filled in, the field values of the fields to be filled in that are not included in the initial text can be determined based on the field association information, and the completeness of field filling can still be ensured. That is, the foregoing method supports receiving an initial text that does not include the field values of all the fields to be filled in, i.e., an incomplete initial text. In actual application, a user does not need to input a complete text including the field values of all the fields to be filled in, but can input a simple demand text, which reduces the user knowledge threshold and improves user experience.
[0145] Referring to Figure 7 , Figure 7 An architecture diagram of a service request processing system provided by one embodiment of the present specification is shown, and the service request system can include a client 100 and a server 200.
[0146] The client 100 is configured to send a service request to the server 200, wherein the service request carries an initial text.
[0147] The server 200 is configured to receive the service request sent by the client, determine field association information corresponding to the service request and an initial domain model, wherein the initial domain model includes a plurality of fields to be filled in, and the field association information is the association information between the fields to be filled in; determine a filling order according to the field association information, fill the fields to be filled in in the initial domain model according to the initial text in the filling order, and obtain a target domain model, wherein the filling order includes at least one field to be filled in; process the service request based on the target domain model, obtain a request processing result corresponding to the service request, and send the request processing result to the client 100.
[0148] The client 100 is further configured to receive the request processing result sent by the server 200.
[0149] By applying the scheme of the embodiments of the present specification, the fields in the domain model corresponding to the initial text can be filled based on the field association information corresponding to the initial text, the missing fields to be filled can be avoided, and the accuracy of field filling can be improved. Moreover, since the field association information is the association information between the fields to be filled, in the case that the initial text does not include the field values of all the fields to be filled, the field values of the fields to be filled that are not included in the initial text can be determined based on the field association information, and the integrity of field filling can still be guaranteed. That is, the foregoing method supports receiving an initial text that does not include the field values of all the fields to be filled, i.e., an incomplete initial text. In actual application, the user does not need to input a complete text including the field values of all the fields to be filled, but can input a simple demand text, which reduces the user knowledge threshold and improves the user experience.
[0150] The service request processing system can include a plurality of clients 100 and a server 200, wherein the client 100 can be referred to as an end-side device, and the server 200 can be referred to as a cloud-side device. The plurality of clients 100 can establish a communication connection through the server 200. In a service request processing scenario, the server 200 is used to provide a service request processing service between the plurality of clients 100. The plurality of clients 100 can respectively act as a sending end or a receiving end, and implement communication through the server 200.
[0151] The user can interact with the server 200 through the client 100 to receive data sent by other clients 100 or send data to other clients 100, and the like. In a service request processing scenario, the user can publish a data stream to the server 200 through the client 100. The server 200 generates a request processing result according to the data stream, and pushes the request processing result to other clients that establish a communication connection.
[0152] The client 100 and the server 200 establish a connection through a network. The network provides a medium for a communication link between the client 100 and the server 200. The network can include various connection types, such as a wired, wireless communication link, or a fiber optic cable, and the like. The data transmitted by the client 100 can need to be processed through encoding, transcoding, compression, and the like before being published to the server 200.
[0153] The client 100 can be a browser, an APP (Application), or a web application such as an H5 (HyperText Markup Language 5) application, or a light application (also known as a mini-program, a lightweight application), or a cloud application, etc. The client 100 can be developed based on a software development kit (SDK) of a corresponding service provided by the server 200, such as an RTC (Real Time Communication) SDK, etc. The client 100 can be deployed in a computing device, and needs to rely on the device or some APP in the device, etc. The computing device can have a display screen and support information browsing, etc., such as a personal mobile terminal such as a mobile phone, a tablet computer, a personal computer, etc. Various other types of applications can also be configured in the computing device, such as human-computer dialogue applications, model training applications, text processing applications, web browser applications, shopping applications, search applications, instant communication tools, mailbox clients, social platform software, etc.
[0154] The server 200 can include servers providing various services, such as servers providing communication services for multiple clients, servers for background training to support models used on clients, servers processing data sent by clients, etc. It should be noted that the server 200 can be implemented as a distributed server cluster composed of multiple servers, or as a single server. The server can also be a server of a distributed system, or a server combined with a blockchain. The server can also be a cloud server of cloud services, cloud databases, cloud computing, cloud functions, cloud storage, network services, cloud communication, middleware services, domain name services, security services, content distribution networks (CDNs), and big data and artificial intelligence platforms, etc. basic cloud computing services, or an intelligent cloud computing server or intelligent cloud host with artificial intelligence technology.
[0155] It should be noted that the business providing method and the business request processing method provided in the embodiments of the present specification are generally executed by the server, but in other embodiments of the present specification, the client can also have similar functions as the server, so as to execute the business providing method and the business request processing method provided in the embodiments of the present specification. In other embodiments, the business providing method and the business request processing method provided in the embodiments of the present specification can also be executed by the client and the server together.
[0156] Figure 8A structural block diagram of a computing device 800 is shown, according to an embodiment of the present application. The components of the computing device 800 include, but are not limited to, a memory 810 and a processor 820. The processor 820 is connected to the memory 810 through a bus 830, and a database 850 is used to store data.
[0157] The computing device 800 also includes an access device 840 that enables the computing device 800 to communicate via one or more networks 860. Examples of these networks include a public switched telephone network (PSTN), a local area network (LAN), a wide area network (WAN), a personal area network (PAN), or combinations of such networks, such as the Internet. The access device 840 can include one or more of any type of network interface (for example, a network interface card (NIC)), wired or wireless, such as an IEEE 802.11 wireless local area network (WLAN) wireless interface, a Worldwide Interoperability for Microwave Access (Wi-MAX) interface, an Ethernet interface, a Universal Serial Bus (USB) interface, a cellular network interface, a Bluetooth interface, a Near Field Communication (NFC) interface, and so on.
[0158] In an embodiment of the present application, the above-mentioned components of the computing device 800, as well as other components not shown in the figure, can be connected to each other, for example, through a bus. It should be understood that, Figure 8 the computing device 800 can be connected to each other through other means. For example, the components of the computing device 800 can be connected to each other through a point-to-point connection, a shared bus, a cross-bar switch, or other means. Figure 8 The structural block diagram of the computing device shown is only for the purpose of example, and is not a limitation on the scope of the present application. Other components can be added or replaced as needed by those skilled in the art.
[0159] The computing device 800 can be any type of stationary or mobile computing device, including a mobile computer or mobile computing device (e.g., a tablet computer, a personal digital assistant, a laptop computer, a notebook computer, a netbook, etc.), a mobile phone (e.g., a smartphone), a wearable computing device (e.g., a smart watch, smart glasses, etc.), or other type of mobile device, or a stationary computing device such as a desktop computer or a personal computer (PC). The computing device 800 can also be a mobile or stationary server.
[0160] The processor 820 is configured to execute the computer program / instructions, which, when executed by the processor, implement the steps of the service providing method and the service request processing method.
[0161] The above is a schematic solution of the computing device of the embodiment. It should be noted that the technical solution of the computing device belongs to the same concept as the technical solution of the service providing method and the service request processing method. For details of the technical solution of the computing device that are not described in detail, reference can be made to the description of the technical solution of the service providing method and the service request processing method.
[0162] An embodiment of the present specification also provides a computer readable storage medium storing computer program / instructions, which, when executed by the processor, implement the steps of the service providing method and the service request processing method.
[0163] Each of the embodiments in the present specification is described in a progressive manner, and the same or similar parts between the embodiments can be referred to each other. Each of the embodiments mainly describes the difference from other embodiments. In particular, for the computer readable storage medium embodiment, since it is basically similar to the service providing method and the service request processing method embodiment, the description is relatively simple, and the relevant parts can be referred to the description of the service providing method and the service request processing method embodiment.
[0164] An embodiment of the present specification also provides a computer program product comprising computer program / instructions, which, when executed by the processor, implement the steps of the service providing method and the service request processing method.
[0165] The above is a schematic solution of the computer program product of the embodiment. It should be noted that the technical solution of the computer program product belongs to the same concept as the technical solution of the service providing method and the service request processing method. For details of the technical solution of the computer program product that are not described in detail, reference can be made to the description of the technical solution of the service providing method and the service request processing method.
[0166] The above detailed description has shown, described, and pointed out novel features of the application. The foregoing can be implemented in other embodiments or carried out in additional or different ways without departing from the spirit and essential characteristics of the application. The present embodiments are therefore to be considered in all respects as illustrative and not restrictive, the scope of the application being indicated by the appended claims rather than by the foregoing description, and all changes which come within the meaning and range of equivalency of the claims are therefore intended to be embraced therein.
[0167] The computer readable medium can include any entity or apparatus that can carry the computer program code, recording medium, U disk, mobile hard disk, magnetic disk, optical disk, computer memory, Read-Only Memory (ROM), Random Access Memory (RAM), electrical carrier signal, telecommunication signal, software distribution medium, etc. It should be noted that the computer readable medium can include appropriate contents according to the requirements of patent practice, for example, according to the patent practice in some regions, the computer readable medium does not include electrical carrier signals and telecommunication signals.
[0168] It should be noted that the above describes specific embodiments of the present specification. Other embodiments are within the scope of the appended claims. In some cases, the actions or steps recited in the claims can be performed in an order different than the order in which they are recited in the embodiments, and still achieve desirable results. In addition, the processes depicted in the figures do not necessarily require the particular order shown, or sequential order, to achieve the desired results. In certain implementations, multitasking and parallel processing can also be advantageous.
[0169] In the above embodiments, the description of each embodiment has its own focus, and the parts not described in detail in a certain embodiment can be referred to the relevant description of other embodiments.
[0170] The preferred embodiments of the present specification disclosed above are only used to help explain the present specification. Alternative embodiments do not describe all the details and limit the present application to the specific embodiments described. Obviously, many modifications and changes can be made according to the content of the embodiments of the present specification. The present specification selects and specifically describes these embodiments in order to better explain the principles and practical applications of the embodiments of the present specification, so that those skilled in the art can well understand and utilize the present specification. The present specification is limited only by the claims and their full scope and equivalents.
Claims
1. A service providing method characterized by comprising: The method comprises: obtaining initial text, wherein the initial text is a description text of a target service; determining field association information corresponding to the initial text and an initial domain model, wherein the initial domain model comprises a plurality of to-be-filled fields, and the field association information comprises an association relationship between the to-be-filled fields; determining a filling order according to the field association information, filling the to-be-filled fields in the initial domain model according to the initial text in the filling order through a large language model, and obtaining a target domain model, wherein the filling order is a filling order of the to-be-filled fields, and the filling order comprises at least one to-be-filled field; providing the target service based on the target domain model.
2. The method of claim 1, wherein, The method comprises: determining a target service corresponding to the initial text; determining the field association information and the initial domain model according to the target service.
3. The method of claim 2, wherein, The method comprises: determining a current to-be-filled field according to the filling order; determining a field value of the current to-be-filled field according to the initial text, and filling the current to-be-filled field in the initial domain model according to the field value.
4. The method of claim 3, wherein, The method comprises: determining a field value of the current to-be-filled field according to the initial text, comprising: determining whether the initial text comprises a field value corresponding to the current to-be-filled field; if yes, extracting the field value corresponding to the current to-be-filled field from the initial text; 5. The method of claim 4, wherein, if no, obtaining the field value corresponding to the current to-be-filled field according to a preset condition corresponding to the current to-be-filled field. The method comprises: in a case where the initial text satisfies the preset condition, obtaining the field value corresponding to the current to-be-filled field from the preset condition; 6. The method of claim 4, wherein, in a case where the initial text does not satisfy the preset condition, generating question information according to the current to-be-filled field, and obtaining the field value corresponding to the current to-be-filled field based on reply information corresponding to the question information. The method comprises: obtaining the field value corresponding to the current to-be-filled field according to the preset condition corresponding to the current to-be-filled field; 7. The method of claim 4, wherein, updating the initial text according to the obtained field value corresponding to the current to-be-filled field. The method comprises: obtaining the field value corresponding to the current to-be-filled field from the initial text according to the prompt text.
8. The method of claim 4, wherein, The method further comprises: in a case where the field value corresponding to the current to-be-filled field cannot be obtained based on reply information corresponding to question information, re-determining the field association information and the initial domain model according to the initial text and the reply information.
9. A service request processing method characterized by comprising: The method is applied to a cloud-side device, comprising: A service request sent by a terminal-side device, wherein the service request carries an initial text; Determine the field association information corresponding to the service request and the initial domain model, wherein the initial domain model includes a plurality of fields to be filled, and the field association information includes the association relationship between the fields to be filled; According to the field association information, determine the filling order, and fill the fields to be filled in the initial domain model according to the initial text through a large language model according to the filling order to obtain a target domain model, wherein the filling order is the filling order of the fields to be filled, and the filling order includes at least one field to be filled. Based on the target domain model, process the service request to obtain a request processing result corresponding to the service request, and send the request processing result to the terminal-side device.
10. A service providing apparatus characterized by comprising: Comprise: An acquisition module configured to acquire an initial text, wherein the initial text is a description text of a target service; A first determination module configured to determine field association information corresponding to the initial text and an initial domain model, wherein the initial domain model includes a plurality of fields to be filled, and the field association information includes the association relationship between the fields to be filled; A first filling module configured to determine a filling order according to the field association information, and fill the fields to be filled in the initial domain model according to the initial text through a large language model according to the filling order to obtain a target domain model, wherein the filling order is the filling order of the fields to be filled, and the filling order includes at least one field to be filled. A providing module configured to provide the target service based on the target domain model.
11. A computing device, comprising: Comprise: A memory and a processor; The memory is used to store computer programs / instructions, and the processor is used to execute the computer programs / instructions, which realize the steps of the method of any one of claims 1 to 9 when executed by the processor.
12. A computer-readable storage medium, characterized in that, It stores computer programs / instructions, which realize the steps of the method of any one of claims 1 to 9 when executed by the processor.
13. A computer program product, characterised in that, It includes computer programs / instructions, which realize the steps of the method of any one of claims 1 to 9 when executed by the processor.
Citation Information
Patent Citations
Interactive service processing method and system
CN111930288A
Data providing method and device, server and storage medium
CN116204560A
Business document generation method and device, computer equipment and storage medium
CN116226117A