Service request processing method and device and electronic equipment

Abstract

The embodiment of the invention provides a service request processing method and device and electronic equipment, and relates to the field of big data. The method comprises the following steps: in response to an obtained service request, preprocessing the service request to obtain a service vector corresponding to the service request, analyzing and processing the service vector to obtain a task table corresponding to the service vector, determining a job module corresponding to each task, and controlling the operation module to execute corresponding task processing to obtain at least one processing result. According to the method, the operation module of the service request is intelligently determined, and all the service requests are processed at the same time, so that the processing efficiency is improved.

Description

Technical Field

[0001] The present application relates to the field of big data, and in particular to a method, device and electronic device for processing business requests. Background Art

[0002] With the widespread use of mobile applications in the financial system, financial institutions are facing a large number of business requests from a vast number of users. These business requests are not only numerous but also diverse in type. At the same time, users have extremely high expectations for timely and accurate service. Promptly responding to business requests is a key factor in improving customer satisfaction.

[0003] Currently, financial systems primarily rely on rule-based engine solutions to process mobile application requests, using manually pre-defined business rules to achieve business diversion. When a user initiates a business request, the system matches the request content with pre-defined rules. If the request meets the conditions of a rule, the corresponding action is executed.

[0004] However, as the business continues to change and expand, the number of rules increases rapidly. Every time a new financial product or service is launched, new rules need to be manually added. Whenever a business changes, the relevant rules need to be manually modified one by one. The lack of intelligent processing methods for business requests leads to low processing efficiency. Summary of the Invention

[0005] The present application provides a method, device and electronic device for processing business requests to solve the problems caused by the manual preset rule engine processing method in the existing business request processing process, especially when the number of business request jobs is large, resulting in a technical problem that the matching rule engine process is time-consuming and inefficient.

[0006] In a first aspect, the present application provides a method for processing a service request, comprising:

[0007] In response to the acquired service request, pre-process the service request to obtain a service vector corresponding to the service request;

[0008] Analyzing and processing the business vector to obtain a task table corresponding to the business vector, wherein the task table includes at least one task;

[0009] Determine the job module corresponding to each task, and control the job module to perform corresponding task processing to obtain at least one processing result.

[0010] Optionally, analyzing and processing the business vector to obtain a task table corresponding to the business vector includes:

[0011] constructing a multi-dimensional dialogue vector based on the service vector, wherein the multi-dimensional dialogue vector includes: the service request, the service vector, and historical service requests;

[0012] Analyzing and processing the multi-dimensional conversation vector to obtain a user intention list;

[0013] According to the user intention list, the multi-dimensional dialogue vector is split to obtain a subtask list, wherein the subtasks are related to the user intention. Figure 1 One to one correspondence;

[0014] Assigning a weight value to any subtask in the subtask list, and using the weight value as a label for the subtask;

[0015] Combine the multiple subtasks and the labels corresponding to the subtasks to obtain a task table.

[0016] Optionally, determining a job module corresponding to each task and controlling the job module to perform corresponding task processing to obtain at least one processing result includes:

[0017] Determine the key word vector of any subtask in the task table;

[0018] Matching a business system interface identifier for the subtask according to the key word segmentation vector;

[0019] Determining the operation module of the subtask according to the identifier;

[0020] The operation module is controlled to process the subtask and obtain a processing result.

[0021] Optionally, the method further includes:

[0022] When multiple subtasks match the same system interface identifier, determining a weight value of any one of the multiple subtasks according to the label of the subtask;

[0023] Sort multiple weight values ​​in descending order;

[0024] The job module processes the multiple subtasks in sequence based on the weight value sorting.

[0025] Optionally, preprocessing the service request to obtain a service vector corresponding to the service request includes:

[0026] Splitting the business request to obtain multiple word segments corresponding to the business request;

[0027] Performing conversion processing on any one of the multiple segmentations to obtain a segmentation vector;

[0028] Analyzing and processing the plurality of word segmentation vectors to determine at least one key word segmentation vector among the plurality of word segmentation vectors;

[0029] The at least one key word segmentation vector is encapsulated to obtain a business vector.

[0030] Optionally, the method further includes:

[0031] Obtain resource consumption data of job modules in real time;

[0032] Determining whether there is a job module in the resource consumption data whose processing response time exceeds a first threshold, wherein the processing response time exceeding the first threshold indicates that the job module is in a high-load state;

[0033] When there is a job module whose processing response time exceeds the first threshold, resource allocation is optimized through a dynamic load balancing algorithm, and idle resources of the low-load job module are allocated to the job module whose processing response time exceeds the first threshold, so that the job module whose processing response time exceeds the first threshold can run stably.

[0034] Optionally, the method further includes:

[0035] Encapsulating the processing result corresponding to the processed business request, and storing the encapsulated business request and the processing result in a cache database;

[0036] Matching a business system interface identifier for the subtask according to the key word segmentation vector includes:

[0037] Determining whether a corresponding processing result exists in the cache database based on the subtask and the key word segmentation vector corresponding to the subtask;

[0038] If it does not exist, the business system interface identifier is matched for the subtask according to the key word segmentation vector.

[0039] In a second aspect, the present application provides a device for processing a service request, comprising:

[0040] a processing module, configured to pre-process the service request in response to the acquired service request to obtain a service vector corresponding to the service request;

[0041] The processing module is further configured to analyze and process the business vector to obtain a task table corresponding to the business vector, wherein the task table includes at least one task;

[0042] A determination module is used to determine the operation module corresponding to each task;

[0043] The processing module is further used to control the operation module to perform corresponding task processing and obtain at least one processing result.

[0044] Optionally, the processing module is further configured to construct a multi-dimensional conversation vector based on the business vector, wherein the multi-dimensional conversation vector includes: the business request, the business vector, and historical business requests;

[0045] The processing module is further configured to analyze and process the multi-dimensional conversation vector to obtain a user intention list;

[0046] The processing module is further configured to split the multi-dimensional dialogue vector according to the user intention list to obtain a subtask list, wherein the subtasks are related to the user intention list. Figure 1 One to one correspondence;

[0047] The processing module is further configured to assign a weight value to any subtask in the subtask list and use the weight value as a label for the subtask;

[0048] The processing module is further configured to combine the plurality of subtasks and the labels corresponding to the subtasks to obtain a task table.

[0049] Optionally, the determination module is further configured to determine a key word segmentation vector in any subtask in the task table;

[0050] The processing module is further configured to match a business system interface identifier for the subtask based on the key word segmentation vector;

[0051] The determining module is further configured to determine the operation module of the subtask according to the identifier;

[0052] The processing module is further configured to control the operation module to process the subtasks and obtain processing results.

[0053] Optionally, the determination module is further configured to determine a weight value of any one of the multiple subtasks according to the labels of the subtasks when multiple subtasks match the same system interface identifier;

[0054] The processing module is further configured to sort the multiple weight values ​​in descending order;

[0055] The processing module is also used for the operation module to process multiple subtasks in sequence based on the weight value sorting.

[0056] Optionally, the processing module is further configured to split the business request to obtain multiple word segments corresponding to the business request;

[0057] The processing module is further configured to perform conversion processing on any one of the multiple segmentations to obtain a segmentation vector;

[0058] The processing module is further configured to analyze and process the plurality of word segmentation vectors to determine at least one key word segmentation vector among the plurality of word segmentation vectors;

[0059] The processing module is further configured to encapsulate the at least one key word segmentation vector to obtain a business vector.

[0060] Optionally, the device further includes: an acquisition module and a judgment module;

[0061] The acquisition module is used to acquire resource consumption data of the operation module in real time;

[0062] The determination module is configured to determine whether there is a job module in the resource consumption data whose processing response time exceeds a first threshold, wherein the processing response time exceeding the first threshold indicates that the job module is in a high-load state;

[0063] The processing module is also used to optimize resource allocation through a dynamic load balancing algorithm when there is a job module whose processing response time exceeds a first threshold, and allocate idle resources of the low-load job module to the job module whose processing response time exceeds the first threshold, so as to enable the job module whose processing response time exceeds the first threshold to operate stably.

[0064] Optionally, the processing module is further configured to encapsulate a processing result corresponding to the processed business request, and store the encapsulated business request and the processing result in a cache database;

[0065] Matching a business system interface identifier for the subtask according to the key word segmentation vector includes:

[0066] The determining module is further configured to determine whether a corresponding processing result exists in the cache database based on the subtask and the key word segmentation vector corresponding to the subtask;

[0067] The processing module is further configured to match a business system interface identifier for the subtask according to the key word segmentation vector when there is no corresponding processing result in the cache database.

[0068] In a third aspect, the present application provides an electronic device, comprising: a processor, and a memory communicatively connected to the processor;

[0069] The memory stores computer-executable instructions;

[0070] The processor executes the computer-executable instructions stored in the memory to implement the method for processing the service request as described in the first aspect and various possible implementations of the first aspect.

[0071] In a fourth aspect, the present application provides a computer-readable storage medium having computer execution instructions stored thereon, which, when executed by a processor, are used to implement the method for processing business requests as described in the first aspect and various possible implementations of the first aspect.

[0072] In a fifth aspect, the present application provides a program product, including a computer program, which implements the above-mentioned method for processing business requests when executed by a processor.

[0073] The service request processing method, device, and electronic device provided in this application preprocess the service request in response to an acquired service request to obtain a service vector corresponding to the service request. The service vector is then analyzed and processed to obtain a task table corresponding to the service vector, which includes at least one task. The task table then determines the job module corresponding to each task, and controls the job module to execute the corresponding task processing to obtain at least one processing result. This method improves processing efficiency by intelligently determining the job module for a service request and completing the simultaneous processing of all service requests. BRIEF DESCRIPTION OF THE DRAWINGS

[0074] The accompanying drawings, which are incorporated in and constitute a part of this specification, illustrate embodiments consistent with the present application and, together with the description, serve to explain the principles of the present application.

[0075] Figure 1 A process diagram of a method for processing a business request provided in this application Figure 1 ;

[0076] Figure 2 A process diagram of a method for processing a business request provided in this application Figure 2 ;

[0077] Figure 3 A process diagram of a method for processing a business request provided in this application Figure 3 ;

[0078] Figure 4 A schematic diagram of the structure of a service request processing device provided in this application;

[0079] Figure 5 A schematic diagram of the structure of a business request processing device provided in this application.

[0080] The above drawings illustrate specific embodiments of the present application, which will be described in more detail below. These drawings and the textual description are not intended to limit the scope of the present application in any way, but rather to illustrate the concepts of the present application to those skilled in the art by reference to specific embodiments. DETAILED DESCRIPTION

[0081] Exemplary embodiments will be described in detail herein, with examples illustrated in the accompanying drawings. In the following description, when referring to the drawings, identical numerals in different figures represent identical or similar elements, unless otherwise indicated. The embodiments described in the following exemplary embodiments are not intended to represent all embodiments consistent with the present application. Rather, they are merely examples of apparatus and methods consistent with certain aspects of the present application, as detailed in the appended claims.

[0082] It should be noted that the user information (including but not limited to user device information, user personal information, etc.) and data (including but not limited to data used for analysis, stored data, displayed data, etc.) involved in this application are all information and data authorized by the user or fully authorized by all parties, and the collection, storage, use, processing, transmission, provision, disclosure and application of relevant data comply with relevant laws, regulations and standards, take necessary confidentiality measures, do not violate public order and good morals, and provide corresponding operation entrances for users to choose to authorize or refuse.

[0083] In addition, this application involves conducting big data analysis of user information (including but not limited to personal biometrics, identity data, consumption data, asset data, electronic terminal operation data, etc.), and using artificial intelligence technology to make automated decisions, and making technical solutions that have a significant impact on personal rights and interests based on the results of automated decisions. The application provides users with corresponding operation entrances for them to choose to agree or reject the results of automated decisions; if the user chooses to reject, the expert decision-making process will be entered.

[0084] It should be noted that the business request processing method, device and electronic device provided in this application can be used in the big data field, and can also be used in any field other than big data. The application field of the business request processing method, device and electronic device in this application is not limited.

[0085] With the widespread adoption of mobile applications in the financial system, financial institutions are facing a massive influx of business requests from a vast number of users. These requests are not only numerous and diverse, but also place extremely high demands on the timeliness and accuracy of services. Users expect to receive efficient, stable, and secure financial services anytime, anywhere. Therefore, rapidly responding to business requests has become a key factor in improving customer satisfaction and enhancing market competitiveness.

[0086] Currently, the processing of mobile business requests in financial systems primarily relies on a combination of a rules engine and manually preset business rules. The system matches user requests against pre-defined rule logic. If a request meets the conditions of a specific rule, the corresponding processing flow is automatically executed. This mechanism achieves a certain degree of automated business diversion and processing.

[0087] However, with the continued development and expansion of the business, the number of rules has rapidly increased. Whenever a new financial product or service is launched, the corresponding processing rules must be manually added. When the business undergoes changes, existing rules must be adjusted and maintained one by one. Due to the lack of an intelligent processing mechanism, the current request processing method is inefficient and unable to flexibly respond to complex and changing business needs, seriously affecting the system's responsiveness and operational efficiency.

[0088] To address the above issues, this application proposes a method for processing business requests. This method preprocesses the business request in response to an acquired business request to obtain a business vector corresponding to the business request. The business vector is then analyzed and processed to obtain a task table corresponding to the business vector. The task table includes at least one task. The job module corresponding to each task is determined, and the job module is controlled to execute the corresponding task processing to obtain at least one processing result. This method improves processing efficiency by intelligently determining the job module for a business request and completing the simultaneous processing of all business requests.

[0089] The following specific embodiments describe in detail the technical solution of the present application and how the technical solution of the present application solves the above-mentioned technical problems. The following specific embodiments can be combined with each other, and the same or similar concepts or processes may not be repeated in some embodiments. The embodiments of the present application will be described below in conjunction with the accompanying drawings.

[0090] Figure 1 A schematic diagram of a process for processing a service request provided in an embodiment of the present application Figure 1 .like Figure 1 As shown, the method for processing a service request provided in this embodiment includes:

[0091] S101 : Pre-process the acquired service request to obtain a service vector corresponding to the service request.

[0092] In financial systems, when users initiate service requests through mobile apps or other channels, preprocessing is required to respond to the received service requests. The purpose of preprocessing is to convert the original service requests into a format that can be further analyzed and processed by computers, namely a service vector. A service vector is a structured data representation that can more efficiently support subsequent operations such as intent recognition, task allocation, and service processing.

[0093] Preprocessing refers to the process of cleaning, standardizing, and structuring a business request before it enters the core analysis module. Common preprocessing steps include:

[0094] Text cleaning: Remove meaningless characters, duplicate content, emojis, etc.

[0095] Word segmentation: Split continuous statements into meaningful words or phrases. For example, the sentence "Transfer 5000 yuan" is processed to obtain: "Transfer", "5000", "yuan".

[0096] Stop word removal: Filter out common words such as "de", "le", "ma", etc. that do not affect semantic understanding.

[0097] Convert the features of the business request into the form of a numerical vector to form a business vector. The business vector can be further processed by a machine learning model or a data analysis tool. In this way, a large number of user requests can be efficiently processed, and the service efficiency can be improved.

[0098] S102. Analyze and process the business vector to obtain a task table corresponding to the business vector.

[0099] Among them, the business vector is a numerical representation that can be recognized by a computer and contains the core features and semantic information of the business request. Analyzing and processing the business vector maps the business vector to a specific task table, and each task in the task table corresponds to a specific business operation or service process. Analyzing and processing the business vector to obtain the task table corresponding to the business vector can efficiently decompose the user's business request into executable tasks, which is convenient for subsequent allocation to the corresponding business modules for processing.

[0100] Analyze the business vector through a machine learning model or a rule engine to identify the user's true intention. The intention refers to the specific goal that the user hopes to achieve through the business request, such as: querying the balance, transferring money, applying for a loan, etc. According to the identified intention, map the business vector to specific tasks. Each task corresponds to a clear business operation or service process. Combine all the mapped tasks into a task table.

[0101] By analyzing and processing the business vector, a task table corresponding to the business vector is obtained. It can efficiently process the user's complex requests and improve the user's satisfaction.

[0102] S103. Determine the job module corresponding to each task, and control the job module to execute the corresponding task processing to obtain at least one processing result.

[0103] The job module is the component in the system responsible for specific business processing. Each task in the task table must be executed by a specific job module. The job module executes the corresponding business logic based on the type and content of the task to obtain the processing result.

[0104] By determining the job modules corresponding to each task and controlling these modules to perform task processing, the system can efficiently complete user business requests and improve business processing efficiency.

[0105] This embodiment provides a method for processing service requests. In response to an acquired service request, the method preprocesses the service request to obtain a service vector corresponding to the service request. The method then analyzes and processes the service vector to obtain a task table corresponding to the service vector, which includes at least one task. The method then determines the job module corresponding to each task, controls the job module to execute the corresponding task, and obtains at least one processing result. This method improves processing efficiency by intelligently determining the job module for a service request and simultaneously processing all service requests.

[0106] Figure 2 A schematic diagram of a process for processing a service request provided in an embodiment of the present application Figure 2 .like Figure 2 As shown, in Figure 1 Based on the embodiment, a method for processing a service request is described in detail, including:

[0107] S201: Split the acquired business request into parts to obtain multiple word segments corresponding to the business request.

[0108] In financial systems, when users submit business requests through mobile apps or other channels, these requests are typically in the form of natural language text. To further process these business requests, computers must first break them down, breaking down complex text into smaller, meaningful tokens.

[0109] Use word segmentation algorithms or tools to split text into multiple independent words. The purpose of word segmentation is to convert a continuous text sequence into a discrete set of words to facilitate subsequent semantic analysis.

[0110] Word segmentation is an important step in natural language processing. It can help computers better understand and analyze user intentions, and provide a basis for subsequent task allocation, intent recognition, and business processing.

[0111] S202: Perform conversion processing on any one of the multiple word segments to obtain a word segmentation vector.

[0112] Word segmentation vectors are a method of converting text data into numerical representations, making them easier to process using machine learning models or data analysis tools. Word segmentation vectors are generated by converting any one of multiple word segments. This is typically achieved using embedding technology. This conversion process captures the semantic relationships between words, providing richer semantic information for subsequent intent recognition, task assignment, and business processing.

[0113] S203: Analyze and process the multiple word segmentation vectors to determine at least one key word segmentation vector among the multiple word segmentation vectors.

[0114] A word segmentation vector is a numerical representation of text data. However, not all word segmentation vectors are equally important for understanding user intent or executing business logic. By analyzing and processing multiple word segmentation vectors, we can identify key word segmentation vectors. These key word segmentation vectors can more accurately reflect the core needs of users.

[0115] S204: Encapsulate at least one key word segmentation vector to obtain a business vector.

[0116] The encapsulation process usually involves vector merging, weighted summation, averaging, or other more complex operations to ensure that the generated business vector can accurately reflect the user's real needs.

[0117] Encapsulating key word vectors to generate business vectors is a crucial step in transforming the semantic information of user requests into structured data that the system can process. A business vector is a comprehensive vector that represents the core semantics of the entire business request, facilitating subsequent intent recognition, task assignment, and business processing.

[0118] S205: Construct a multi-dimensional dialogue vector based on the business vector.

[0119] The multidimensional conversation vector includes the service request, service vector, and historical service requests. The purpose of constructing the multidimensional conversation vector is to more comprehensively understand and process user service requests. The multidimensional conversation vector not only contains the semantic information of the current service request but also integrates the user's historical service requests. This multidimensional information integration helps the system better understand the user's context, behavior patterns, and preferences.

[0120] Specifically, in one possible implementation, a user submits a service request: "I want to check my account balance and view my credit card bill." This service request is pre-processed to obtain a computer-processable service vector corresponding to the request. Based on this service vector, a multi-dimensional conversation vector is constructed by integrating historical service requests, such as "check credit card limit." This multi-dimensional conversation vector includes: {service request; service vector; historical service requests}. Specifically, the multi-dimensional conversation vector is: {"check my account balance and view credit card bill" text; "check my account balance and view credit card bill" service vector; "check credit card limit"}.

[0121] By constructing multi-dimensional conversation vectors, the system can more comprehensively understand user requests and contextual information, thereby providing more accurate and personalized services and improving user experience.

[0122] S206: Analyze and process the multi-dimensional dialogue vector to obtain a user intention list.

[0123] The multidimensional conversation vector integrates multiple dimensions of information, including the semantics of the current business request and the user's past business requests. Machine learning or deep learning models are used to analyze and process the multidimensional conversation vector, identifying the user's true intent and generating a user intent list. Each intent corresponds to a specific business operation or service requirement. The user intent list represents the system's semantic analysis of the user request and provides a basis for subsequent task allocation and business processing.

[0124] Specifically, one possible implementation uses a pre-trained intent recognition model to analyze and process the multi-dimensional vector: {"Query my account balance and view credit card bill" text; "Query my account balance and view credit card bill" business vector; "Query credit card limit"}. Based on the current business vector and historical context, the model identifies the user's true intent and generates a user intent list:

[0125] meaning Figure 1 : Check account balance.

[0126] meaning Figure 2 : Check your credit card bill.

[0127] Based on the historical request of "checking credit card limit", the system can further confirm that the user may have a high degree of concern about credit cards, and can further provide users with value-added services such as credit card application limit increase and credit card installment payment.

[0128] By analyzing user intent and historical context, the system can provide more personalized services. Accurate intent recognition can reduce user operation steps, increase service request response speed, and thus improve user experience.

[0129] S207: Split the multi-dimensional dialogue vector according to the user intent list to obtain a subtask list.

[0130] Among them, by analyzing the multi-dimensional dialogue vector, the multi-dimensional dialogue vector is split according to the user intention list to obtain a sub-task list. Figure 1 One to one, each intent represents a specific need or operation goal of the user.

[0131] Specifically, a possible implementation method is based on the following user intent list:

[0132] meaning Figure 1 : Check account balance.

[0133] meaning Figure 2 : Check your credit card bill.

[0134] Split the multi-dimensional dialogue vector to obtain a subtask list:

[0135] Subtask 1: “Query the account balance” vector representation.

[0136] Subtask 2: “Query the vector representation of the credit card bill”.

[0137] By splitting the multi-dimensional conversation vector according to the user intent list, the user's complex business requests can be broken down into specific subtasks, ensuring that each subtask is processed accurately, improving system efficiency and user experience.

[0138] S208: Assign a weight value to any subtask in the subtask list, and use the weight value as a label of the subtask.

[0139] S209: Combine multiple subtasks and labels corresponding to the subtasks to obtain a task table.

[0140] The criteria for weight assignment are determined based on business requirements and system rules. Weights can be assigned based on task urgency, importance, user priority, or other business logic. Each subtask in the subtask list is assigned a weight value. The assigned weight value is used as the subtask's label and stored in the subtask data structure. All subtasks and their corresponding weight labels are combined into a task table. The task table is a structured data set that records detailed information about each subtask, including task content, weight label, and more.

[0141] S210: Determine the operation module corresponding to each task, and control the operation module to execute the corresponding task processing to obtain at least one processing result.

[0142] Step S210 is similar to step S103 and will not be described again here.

[0143] Optionally, the resource consumption data of the job module is obtained in real time. That is, the system obtains the resource consumption data of the job module in real time, including CPU usage, memory usage, disk I / O, network bandwidth, etc.

[0144] Determine whether any job module in the resource consumption data has a response processing time exceeding a first threshold, where the response processing time exceeding the first threshold indicates that the job module is in a high-load state. Specifically, based on a preset first threshold, such as the response processing time, determine whether any job module has a response processing time exceeding the threshold. A response processing time exceeding the threshold typically indicates that the job module is in a high-load state.

[0145] In the case where there is a job module whose processing response time exceeds the first threshold, resource allocation is optimized through a dynamic load balancing algorithm, and the idle resources of the low-load job module are allocated to the job module whose processing response time exceeds the first threshold, so that the job module whose processing response time exceeds the first threshold can run stably. That is, when a high-load job module is detected, resource allocation is optimized through a dynamic load balancing algorithm. The algorithm can allocate the idle resources of the low-load job module to the high-load job module based on the current usage of system resources. The idle resources of the low-load job module, such as CPU, memory, etc., are dynamically allocated to the high-load job module to ensure that the high-load job module can run stably. The effect of resource allocation is continuously monitored to ensure that the response time of the high-load job module returns to the normal range. If the problem is still not resolved, the resource allocation strategy can be further adjusted or an alarm can be triggered.

[0146] By monitoring resource consumption data in real time and optimizing resource allocation through dynamic load balancing algorithms, the financial system can efficiently process user requests, ensure that each job module can run stably, and improve the overall system performance and user experience.

[0147] This embodiment provides a method for processing business requests, which responds to the acquired business requests, performs splitting processing to obtain multiple word segments, and converts the word segments into word segmentation vectors. Key word segmentation vectors are determined by analysis and encapsulated as business vectors. Based on the business vectors, a multi-dimensional dialogue vector is constructed in combination with business requests and historical business requests. The multi-dimensional dialogue vectors are analyzed to obtain a list of user intents, and tasks are split according to user intents to generate a list of subtasks, where each subtask corresponds to a user intent. The job module for each task is determined, and the job module is controlled to process the tasks to obtain the processing results. The processing efficiency of business requests is improved, and the user experience and the timeliness of service responses are improved.

[0148] Figure 3 A schematic diagram of a process for processing a service request provided in an embodiment of the present application Figure 3 .like Figure 3 As shown, in Figure 1 Based on the embodiment, a possible implementation method for determining the job module corresponding to each task and controlling the job module to perform the corresponding task processing to obtain at least one processing result is described in detail, including:

[0149] S301. Determine the key word vector of any subtask in the task table.

[0150] Each subtask in the task table corresponds to a specific user need or operational intent. To more accurately process these subtasks, the system needs to identify key information within the subtask, specifically key word vectors. Key word vectors are vectors extracted from the subtask's textual description that represent the subtask's core semantics. By determining key word vectors, the system can better understand the specific content of the subtask, thereby more efficiently matching the corresponding business logic and job modules.

[0151] S302: Match the business system interface identifier for the subtask according to the key word segmentation vector.

[0152] Among them, matching the key word vectors of the subtask with the business system interface identifier is an important step in realizing automated task processing. The business system interface identifier is a unique identifier for each functional module in the system, which is used to call specific business logic or services.

[0153] By matching the keyword segmentation vector with the interface identifier, the system can automatically identify the specific business module that needs to be called by the subtask, thereby achieving efficient task processing and resource allocation.

[0154] Optionally, when multiple subtasks match the same system interface identifier, the weight value of any of the subtasks is determined based on the subtask's tag. Specifically, when multiple subtasks match the same system interface identifier, these subtasks and their corresponding weight values ​​are recorded. The weight value is typically a numerical value that indicates the priority of the subtask.

[0155] Sort multiple weight values ​​in descending order. That is, sort the weight values ​​of multiple subtasks in descending order. Subtasks with larger weight values ​​have higher priority.

[0156] The job module processes multiple subtasks sequentially based on their weights. That is, the job module processes subtasks sequentially based on their weights. High-weighted subtasks are prioritized to ensure a quick response to critical tasks.

[0157] By assigning weight values ​​to subtasks and sorting them according to the weight values, the job module can efficiently process multiple subtasks that match the same interface identifier, ensuring efficient system operation and user experience.

[0158] S303: Determine the operation module of the subtask according to the identifier.

[0159] S304: The control operation module processes the subtask and obtains a processing result.

[0160] Based on the business system interface identifier matched to the subtask, the corresponding job module in the system is searched. Each interface identifier is typically associated with a specific job module. The subtask is assigned to the specified job module. Upon receiving the subtask, the job module initiates the corresponding business logic processing flow based on the subtask's content and requirements. The job module processes the subtask and generates a processing result after completing the subtask processing.

[0161] By determining the job module based on the interface identifier and controlling the job module to process the subtasks, the system can efficiently complete the user's request, ensure that each subtask is accurately processed, and improve the system's efficiency and user experience.

[0162] S305 , encapsulating the processing result corresponding to the processed business request, and storing the encapsulated business request and the processing result in a cache database.

[0163] The processed business request and its corresponding processing results are encapsulated and stored in a cache database. The use of a cache database can significantly increase the system's read speed and reduce the access pressure on the main database, thereby improving the overall system performance and user experience.

[0164] S306: Based on the subtask and the key word segmentation vector corresponding to the subtask, determine whether there is a corresponding processing result in the cache database.

[0165] S307: If there is no corresponding processing result in the cache database, match the business system interface identifier for the subtask according to the key word vector.

[0166] Among them, the key word vector can represent the core semantics of the subtask. The key word vector of each subtask is extracted. The key word vector of the subtask is used as the query condition to check whether there is a corresponding processing result in the cache database.

[0167] If the corresponding processing result exists in the cache database, the result is used directly.

[0168] If the corresponding processing result does not exist in the cache database, the subtask is matched to the corresponding business system interface identifier based on the key word vector of the subtask through semantic analysis or predefined mapping rules. Based on the matched interface identifier, the corresponding business module is called to process the subtask and the processing result is stored in the cache database for subsequent query.

[0169] This embodiment provides a method for processing business requests, which performs semantic analysis on the subtasks in the task table, identifies key word vectors, and matches the corresponding business system interface and execution module based on the vector to achieve automatic scheduling and processing of subtasks. In the case where multiple tasks match the same interface, priority sorting is performed based on the weight labels of the subtasks to ensure that high-priority tasks are executed first, thereby improving processing efficiency and resource utilization. After completing the processing, the system encapsulates the results and stores them in the cache database. At the same time, it supports cache retrieval based on semantic features, and can directly reuse existing results when they exist to avoid repeated processing and further improve processing speed and efficiency.

[0170] Figure 4 This is a schematic diagram of the structure of a service request processing device provided by this application. Figure 4 As shown, the present application provides a service request processing device, and the service request processing device 400 includes:

[0171] The processing module 401 is configured to pre-process the acquired service request in response to the acquired service request to obtain a service vector corresponding to the service request;

[0172] The processing module 401 is further configured to analyze and process the business vector to obtain a task table corresponding to the business vector, wherein the task table includes at least one task;

[0173] Determination module 402, used to determine the operation module corresponding to each task;

[0174] The processing module 401 is further configured to control the operation module to execute corresponding task processing and obtain at least one processing result.

[0175] Optionally, the processing module 401 is further configured to construct a multi-dimensional conversation vector based on the business vector, where the multi-dimensional conversation vector includes: the business request, the business vector, and historical business requests;

[0176] The processing module 401 is further configured to analyze and process the multi-dimensional conversation vector to obtain a user intention list;

[0177] The processing module 401 is further configured to split the multi-dimensional dialogue vector according to the user intention list to obtain a subtask list, wherein the subtasks are related to the user intention list. Figure 1 One to one correspondence;

[0178] The processing module 401 is further configured to assign a weight value to any subtask in the subtask list and use the weight value as a label of the subtask;

[0179] The processing module 401 is further configured to combine the plurality of subtasks and the labels corresponding to the subtasks to obtain a task table.

[0180] Optionally, the determining module 402 is further configured to determine a key word segmentation vector in any subtask in the task table;

[0181] The processing module 401 is further configured to match a business system interface identifier for the subtask based on the key word segmentation vector;

[0182] The determining module 402 is further configured to determine the operation module of the subtask according to the identifier;

[0183] The processing module 401 is further configured to control the operation module to process the subtask and obtain a processing result.

[0184] Optionally, the determining module 402 is further configured to determine a weight value of any one of the multiple subtasks according to the labels of the subtasks when multiple subtasks match the same system interface identifier;

[0185] The processing module 401 is further configured to sort the multiple weight values ​​in descending order;

[0186] The processing module 401 is also used by the operation module to process multiple subtasks in sequence based on the weight value sorting.

[0187] Optionally, the processing module 401 is further configured to split the business request to obtain multiple word segments corresponding to the business request;

[0188] The processing module 401 is further configured to perform conversion processing on any one of the multiple segmentations to obtain a segmentation vector;

[0189] The processing module 401 is further configured to analyze and process the multiple word segmentation vectors to determine at least one key word segmentation vector among the multiple word segmentation vectors;

[0190] The processing module 401 is further configured to encapsulate the at least one key word segmentation vector to obtain a business vector.

[0191] Optionally, the device further includes: an acquisition module 403 and a judgment module 404;

[0192] The acquisition module 403 is used to acquire resource consumption data of the operation module in real time;

[0193] The determination module 404 is configured to determine whether there is a job module in the resource consumption data whose processing response time exceeds a first threshold, wherein the processing response time exceeding the first threshold indicates that the job module is in a high load state;

[0194] The processing module 401 is also used to optimize resource allocation through a dynamic load balancing algorithm when there is a job module whose processing response time exceeds the first threshold, and allocate idle resources of the low-load job module to the job module whose processing response time exceeds the first threshold, so as to enable the job module whose processing response time exceeds the first threshold to operate stably.

[0195] Optionally, the processing module 401 is further configured to encapsulate a processing result corresponding to the processed business request, and store the encapsulated business request and the processing result in a cache database;

[0196] Matching a business system interface identifier for the subtask according to the key word segmentation vector includes:

[0197] The determining module 402 is further configured to determine whether a corresponding processing result exists in the cache database based on the subtask and the key word segmentation vector corresponding to the subtask;

[0198] The processing module 401 is further configured to match a business system interface identifier for the subtask according to the key word segmentation vector when there is no corresponding processing result in the cache database.

[0199] The implementation principle and technical effects of the service request processing device provided in the embodiment of the present application are similar to the implementation methods of the various parts of the aforementioned service request processing method, and will not be repeated here.

[0200] Figure 5 This is a schematic diagram of the structure of a service request processing device provided by this application. Figure 5 As shown, the present application provides a service request processing device, and the service request processing device 500 includes: a receiver 501, a transmitter 502, a processor 503 and a memory 504.

[0201] Receiver 501, for receiving instructions and data;

[0202] Transmitter 502, used to send instructions and data;

[0203] Memory 504, for storing computer-executable instructions;

[0204] The processor 503 is configured to execute the computer-executable instructions stored in the memory 504 to implement the various steps of the method for processing a service request in the above embodiment. For details, please refer to the relevant description in the embodiment of the method for processing a service request.

[0205] Optionally, the memory 504 may be independent or integrated with the processor 503 .

[0206] When the memory 504 is independently provided, the electronic device further includes a bus for connecting the memory 504 and the processor 503 .

[0207] The implementation principle and technical effects of the electronic device provided in this embodiment can be found in the aforementioned embodiments and will not be described in detail here.

[0208] An embodiment of the present application further provides a computer-readable storage medium, wherein the computer-readable storage medium stores computer-executable instructions. When a processor executes the computer-executable instructions, the method described in any of the above embodiments is implemented.

[0209] An embodiment of the present application further provides a computer program product, including a computer program, which implements the method described in any of the aforementioned embodiments when executed by a processor.

[0210] It should be noted that for the aforementioned method embodiments, for the sake of simplicity, they are all expressed as a series of action combinations, but those skilled in the art should be aware that this application is not limited by the order of the actions described, because according to this application, certain steps can be performed in other orders or simultaneously. Secondly, those skilled in the art should also be aware that the embodiments described in this specification are all optional embodiments, and the actions and modules involved are not necessarily required by this application.

[0211] It should be further noted that, although the various steps in the flowchart are shown in sequence as indicated by the arrows, these steps are not necessarily performed in the order indicated by the arrows. Unless otherwise specified herein, there is no strict order restriction on the execution of these steps, and these steps may be performed in other orders. Moreover, at least a portion of the steps in the flowchart may include multiple sub-steps or multiple stages, and these sub-steps or stages are not necessarily performed at the same time, but may be performed at different times. The execution order of these sub-steps or stages is not necessarily to be performed in sequence, but may be performed in turn or alternately with other steps or at least a portion of the sub-steps or stages of other steps.

[0212] It should be understood that the above-described device embodiments are merely illustrative, and the device of the present application may also be implemented in other ways. For example, the division of units / modules in the above-described embodiments is merely a logical functional division, and actual implementations may employ other division methods. For example, multiple units, modules, or components may be combined or integrated into another system, or some features may be omitted or not implemented.

[0213] In addition, unless otherwise specified, the functional units / modules in the various embodiments of the present application may be integrated into a single unit / module, each unit / module may exist physically separately, or two or more units / modules may be integrated together. The aforementioned integrated units / modules may be implemented in the form of hardware or software program modules.

[0214] In the above embodiments, the description of each embodiment has its own emphasis. For parts not described in detail in a particular embodiment, please refer to the relevant description of other embodiments. The technical features of the above embodiments can be combined in any way. To keep the description concise, not all possible combinations of the technical features in the above embodiments are described. However, as long as there is no contradiction in the combination of these technical features, they should be considered to be within the scope of this specification.

[0215] Those skilled in the art will readily appreciate other embodiments of the present application after considering the specification and practicing the invention disclosed herein. This application is intended to cover any variations, uses, or adaptations of the present application that follow the general principles of the present application and include common knowledge or customary techniques in the art not disclosed herein. The description and examples are to be considered as exemplary only, and the true scope and spirit of the present application are indicated by the following claims.

[0216] It should be understood that the present application is not limited to the exact structure described above and shown in the drawings, and that various modifications and changes may be made without departing from the scope thereof. The scope of the present application is limited only by the appended claims.

Claims

1. A method for processing a service request, characterized in that: The method comprises: In response to the acquired service request, pre-process the service request to obtain a service vector corresponding to the service request; Analyzing and processing the business vector to obtain a task table corresponding to the business vector, wherein the task table includes at least one task; Determine the job module corresponding to each task, and control the job module to perform corresponding task processing to obtain at least one processing result.

2. The method according to claim 1, characterized in that The analyzing and processing the business vector to obtain a task table corresponding to the business vector includes: constructing a multi-dimensional dialogue vector based on the service vector, wherein the multi-dimensional dialogue vector includes: the service request, the service vector, and historical service requests; Analyzing and processing the multi-dimensional conversation vector to obtain a user intention list; Splitting the multidimensional conversation vector according to the user intent list to obtain a subtask list, wherein the subtasks correspond one-to-one to the user intents; Assigning a weight value to any subtask in the subtask list, and using the weight value as a label for the subtask; Combine the multiple subtasks and the labels corresponding to the subtasks to obtain a task table.

3. The method according to claim 2, characterized in that The determining of the operation module corresponding to each task and controlling the operation module to execute the corresponding task processing to obtain at least one processing result includes: Determine the key word vector of any subtask in the task table; Matching a business system interface identifier for the subtask according to the key word segmentation vector; Determining the operation module of the subtask according to the identifier; The operation module is controlled to process the subtask and obtain a processing result.

4. The method according to claim 3, characterized in that The method further comprises: When multiple subtasks match the same system interface identifier, determining a weight value of any one of the multiple subtasks according to the label of the subtask; Sort multiple weight values ​​in descending order; The job module processes the multiple subtasks in sequence based on the weight value sorting.

5. The method according to claim 1, wherein The pre-processing of the service request to obtain a service vector corresponding to the service request includes: Splitting the business request to obtain multiple word segments corresponding to the business request; Performing conversion processing on any one of the multiple segmentations to obtain a segmentation vector; Analyzing and processing the plurality of word segmentation vectors to determine at least one key word segmentation vector among the plurality of word segmentation vectors; The at least one key word segmentation vector is encapsulated to obtain a business vector.

6. The method according to claim 3, characterized in that The method further comprises: Obtain resource consumption data of job modules in real time; Determining whether there is a job module in the resource consumption data whose processing response time exceeds a first threshold, wherein the processing response time exceeding the first threshold indicates that the job module is in a high-load state; When there is a job module whose processing response time exceeds the first threshold, resource allocation is optimized through a dynamic load balancing algorithm, and idle resources of the low-load job module are allocated to the job module whose processing response time exceeds the first threshold, so that the job module whose processing response time exceeds the first threshold can run stably.

7. The method according to claim 3, characterized in that The method further comprises: Encapsulating the processing result corresponding to the processed business request, and storing the encapsulated business request and the processing result in a cache database; Matching a business system interface identifier for the subtask according to the key word segmentation vector includes: Determining whether a corresponding processing result exists in the cache database based on the subtask and the key word segmentation vector corresponding to the subtask; If it does not exist, the business system interface identifier is matched for the subtask according to the key word segmentation vector.

8. A service request processing device, characterized in that: The device comprises: a processing module, configured to pre-process the service request in response to the acquired service request to obtain a service vector corresponding to the service request; The processing module is further configured to analyze and process the business vector to obtain a task table corresponding to the business vector, wherein the task table includes at least one task; A determination module is used to determine the operation module corresponding to each task; The processing module is further used to control the operation module to perform corresponding task processing and obtain at least one processing result.

9. An electronic device, characterized in that: include: a processor, and a memory communicatively connected to the processor; The memory stores computer-executable instructions; The processor executes the computer-executable instructions stored in the memory to implement the method according to any one of claims 1 to 7.

10. A computer-readable storage medium, characterized in that The computer-readable storage medium stores computer-executable instructions, which are used to implement the method according to any one of claims 1 to 7 when executed by a processor.

11. A computer program product, characterized in that The invention comprises a computer program, which implements the method according to any one of claims 1 to 7 when being executed by a processor.

Images