Service duration determination method and device, equipment, medium and program product

Abstract

The embodiment of the invention provides a service duration determination method and device, equipment, a medium and a program product, and relates to the field of financial science and technology or other related fields. The method comprises the following steps: receiving a service type input by a client in a reservation system, dynamically loading a corresponding structured field, and performing natural language processing on a fuzzy service description input by the client so as to analyze and generate a standardized service code; and integrating the structured field and the standardized business code to generate a standardized demand tree representing the complete business demand of the customer, and finally predicting the service duration of the customer at the bank outlet based on the standardized demand tree. According to the method provided by the invention, the business demand identification precision, the service duration prediction accuracy and the bank outlet service resource configuration efficiency are improved.

Description

Technical Field

[0001] This application relates to the field of financial technology or other related fields, and in particular to a method, apparatus, device, medium and program product for determining service duration. Background Technology

[0002] Bank branches are a crucial offline channel for financial services, and their service efficiency and customer experience significantly impact the overall operational level of banks. With the increasing complexity of financial services and the growing personalization of customer needs, the traditional bank branch service model faces severe challenges in areas such as information collection, resource allocation, and time-based forecasting.

[0003] Currently, bank branches typically use a basic appointment system to supplement offline services. Users need to pre-select appointment information (such as service type and time slot) through online channels. After receiving the appointment information, the basic appointment system allocates an approximate service time to the customer based on preset information. For example, if a customer selects "transfer" service when making an appointment, the system determines the corresponding time for this service to be a fixed value (such as 15 minutes) based on a preset mapping relationship, and allocates service resources according to this fixed value.

[0004] However, existing appointment systems determine the fixed time for each service based on a preset mapping relationship. This can easily lead to a large discrepancy between the predicted results and the actual processing time, affecting the rational allocation of service resources and reducing the efficiency of service processing. Summary of the Invention

[0005] This application provides a method, apparatus, equipment, medium, and program product for determining service duration, in order to improve the accuracy and reliability of predicting business service duration and improve business processing efficiency.

[0006] Firstly, this application provides a method for determining service duration, including:

[0007] Receive the service type entered by the customer in the appointment system; dynamically load the corresponding structured fields based on the service type;

[0008] Natural language processing is used to parse and generate standardized business codes from the fuzzy business descriptions input by customers.

[0009] By integrating structured fields with standardized business codes, a standardized requirement tree representing the complete business needs of customers is generated.

[0010] Based on the standardized demand tree, predict the service time of customers at bank branches.

[0011] Secondly, this application provides a device for determining service duration, comprising:

[0012] The receiving module is used to receive the service type entered by the customer in the appointment system; and dynamically load the corresponding structured fields according to the service type.

[0013] The processing module is used to perform natural language processing on the fuzzy business descriptions input by customers, parse them, and generate standardized business codes.

[0014] The processing module is also used to generate a standardized requirement tree that represents the complete business requirements of customers by integrating structured fields with standardized business codes;

[0015] The processing module is also used to predict the duration of customer service at bank branches based on a standardized demand tree.

[0016] Thirdly, embodiments of this application provide an electronic device, including: a memory and a processor;

[0017] The memory stores the instructions that the computer executes;

[0018] The processor executes computer execution instructions stored in memory, causing the processor to perform the first aspect and / or various possible implementations of the first aspect as described above.

[0019] Fourthly, embodiments of this application provide a computer-readable storage medium storing computer-executable instructions, which, when executed by a processor, are used to implement the first aspect and / or various possible implementations of the first aspect.

[0020] Fifthly, embodiments of this application provide a computer program product, including a computer program that, when executed by a processor, implements the first aspect and / or various possible implementations of the first aspect.

[0021] The service duration determination method, apparatus, equipment, medium, and program products provided in this application receive the service type input by the customer in the appointment system and dynamically load the corresponding structured fields. Simultaneously, natural language processing is used to parse the fuzzy service description input by the customer to generate standardized service codes. Then, the structured fields and standardized service codes are integrated to generate a standardized demand tree representing the customer's complete service needs. Finally, the service duration at the bank branch is predicted based on the standardized demand tree. By dual-verifying and complementing the customer's actively selected service type with the standardized codes obtained through natural language parsing, a complete, accurate, and structured description of customer needs is constructed. This provides a high-quality, unambiguous input foundation for subsequent predictions, effectively solving the problems of inaccurate service demand parsing and large service duration prediction deviations caused by the single dimension of information collection and fuzzy customer descriptions in traditional appointment systems. This achieves the effects of improving the accuracy of service demand identification, the accuracy of service duration prediction, and the efficiency of bank branch service resource allocation. Attached Figure Description

[0022] The accompanying drawings, which are incorporated in and form part of this specification, illustrate embodiments consistent with this application and, together with the description, serve to explain the principles of this application.

[0023] Figure 1 A schematic diagram illustrating the method for determining the service duration provided in this application;

[0024] Figure 2 A flowchart illustrating the method for determining the service duration provided in this application;

[0025] Figure 3 A schematic diagram of the device for determining the service duration provided in this application;

[0026] Figure 4 A schematic diagram of the structure of the electronic device provided in this application.

[0027] The accompanying drawings illustrate specific embodiments of this application, which will be described in more detail below. These drawings and descriptions are not intended to limit the scope of the concept in any way, but rather to illustrate the concept of this application to those skilled in the art through reference to particular embodiments. Detailed Implementation

[0028] Exemplary embodiments will now be described in detail, examples of which are illustrated in the accompanying drawings. When the following description relates to the drawings, unless otherwise indicated, the same numbers in different drawings denote the same or similar elements. The embodiments described in the following exemplary embodiments do not represent all embodiments consistent with this application. Rather, they are merely examples of apparatuses and methods consistent with some aspects of this application as detailed in the appended claims.

[0029] 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, data stored, data displayed, etc.) involved in this application are all information and data authorized by the user or fully authorized by all parties. Furthermore, the collection, storage, use, processing, transmission, provision, disclosure, and application of the relevant data all comply with the relevant laws, regulations, and standards of the relevant countries and regions, have taken necessary confidentiality measures, do not violate public order and good morals, and provide corresponding operation access points for users to choose to authorize or refuse.

[0030] Furthermore, the technical solution involved in this application, which involves 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 the use of artificial intelligence technology for automated decision-making, and makes decisions that have a significant impact on personal rights based on the results of automated decision-making, provides users with corresponding operation entry points for users to choose to agree to or reject the results of automated decision-making; if the user chooses to reject, the process will proceed to the expert decision-making process.

[0031] It should be noted that the methods, apparatus, equipment, media, and program products for determining service duration provided in this application can be used in the field of fintech or other related fields, or in any field other than fintech or other related fields. The application fields of the methods, apparatus, equipment, media, and program products for determining service duration in this application are not limited.

[0032] As crucial offline touchpoints for financial services, bank branches are required to provide customers with services such as transaction processing, consultation, and product recommendations. However, with the increasing complexity of financial transactions and the growing personalized needs of customers, the traditional bank branch service model faces significant challenges. For example, customers may misjudge the processing time before arriving at the bank, leading to severe queues during peak hours (e.g., 9:00-10:00 AM and 2:00-3:00 PM on weekdays), with average waiting times exceeding 40 minutes, further resulting in a significant decline in customer satisfaction.

[0033] To address the aforementioned issues, existing technologies employ a queuing system where customers arrive at bank branches and are called in a number. After queuing, the system obtains a service number and schedules customers based on the availability of available service windows. However, this existing mechanism struggles to predict customer demand, hindering dynamic adjustments to service resource allocation. This can lead to congestion during peak hours, requiring customers to queue repeatedly and negatively impacting their experience. Furthermore, some banks have established basic appointment systems to provide online booking, allowing customers to pre-select service types and arrival times. However, these systems only support basic service registration (e.g., deposits, transfers) and lack in-depth analysis of customer needs. Consequently, some customers may need to return to the bank in person due to incomplete documentation. Additionally, existing technologies predict service durations based on static empirical values ​​(e.g., "average transfer time is 15 minutes") and then allocate service resources (e.g., number of tellers, number of open service windows) at bank branches accordingly. This approach struggles to adapt to the combined effects of dynamic variables such as business complexity, customer profiles (e.g., VIP customer priority), and differences in teller skills, leading to significant prediction errors, reduced business processing efficiency, increased time costs for customers, and exacerbated operational pressures on bank branches.

[0034] In summary, existing technologies are prone to misidentifying business types or overlooking key customer needs, reducing the accuracy of service duration predictions and the precision of business demand analysis; existing predictive models rely solely on static experience values, which can easily increase prediction errors; and existing technologies fail to verify the compliance of materials submitted by customers during the appointment process, leading to frequent issues with missing materials and the possibility of customers making repeated visits to the store.

[0035] The method for determining service duration in this application involves receiving the business type input by the customer in the appointment system; dynamically loading the corresponding structured fields based on the business type; parsing and generating standardized business codes through natural language processing of the fuzzy business description input by the customer; integrating the structured fields and standardized business codes to generate a standardized requirement tree representing the customer's complete business needs; and predicting the customer's service duration at the bank branch based on the standardized requirement tree. This application improves the accuracy and reliability of service duration prediction, further helping to reduce the rate of repeat visits to bank branches and reduce fluctuations in the bank's service duration.

[0036] The method for determining service duration provided in this application aims to solve the above-mentioned technical problems in the prior art.

[0037] The technical solution of this application and how the technical solution of this application solves the above-mentioned technical problems are described in detail below with specific embodiments. These specific embodiments can be combined with each other, and the same or similar concepts or processes may not be described again in some embodiments. The embodiments of this application will be described below with reference to the accompanying drawings.

[0038] This application applies to customer appointment services at bank branches. Figure 1 A schematic diagram illustrating the scenario of the method for determining the service duration provided in this application, as shown below. Figure 1 As shown, the appointment system includes a front-end interaction module, a back-end prediction module, and a service resource configuration module. The front-end interaction module refers to the online appointment platform (e.g., a mobile banking app or official website). Users needing to conduct business submit appointment requests through the front-end interaction module. Upon receiving the appointment request, the back-end prediction module predicts the service duration based on the appointment information indicated in the request. After obtaining the predicted service duration, the service resource configuration module adjusts resources accordingly. For example, after a customer submits an appointment request through the front-end interaction module, the back-end prediction module predicts the service duration, allowing the service resource configuration module to dynamically adjust service resources based on the prediction results and provide reminders for missing documents, thereby optimizing the customer's time spent at the bank branch and reducing queue congestion.

[0039] Figure 2 A flowchart illustrating the method for determining the service duration provided in this application is shown below. Figure 2 As shown, the method includes:

[0040] S201. Receive the business type entered by the customer in the appointment system; dynamically load the corresponding structured fields according to the business type.

[0041] More specifically, when customers select or enter a business category (e.g., "cross-border remittance") in the appointment system, the appointment system automatically loads the corresponding field configurations (e.g., "receiving country" drop-down menu) based on the business type, ensuring that the business requirements entered by the customer are field-based.

[0042] Optionally, based on the business type, the corresponding structured fields are dynamically loaded, specifically including: mapping the business type to a predefined business type code; querying and obtaining the field configuration information corresponding to the business type code based on the field mapping table; and generating user interface elements based on the interactive form component by rendering the field configuration information into an interactive form component for customer input.

[0043] In one possible embodiment, the specific implementation of dynamically loading the corresponding structured fields according to the business type in step S201 is as follows: The front-end system obtains the original data of the business type selected or entered by the customer in the appointment interface; for example, the customer selects the "Cross-border Remittance" option. The business type data is received by the business parsing engine and mapped to a predefined standardized business type code; for example, "Cross-border Remittance" is mapped to code "XW001". Then, based on the business type code, the field mapping table stored in the system database is queried to obtain the corresponding field configuration information. The field configuration information is received by the front-end rendering engine, parsed, and rendered into the corresponding interactive form component; for example, a "Receiving Country" drop-down menu is generated. Finally, a user interface containing the interactive form component is generated and presented to the customer.

[0044] This embodiment utilizes a field mapping table query mechanism to ensure that field configuration information corresponding to different business types can be quickly retrieved, avoiding redundancy or omissions in field loading. The generation of interactive form components improves the convenience and accuracy of customer input, allowing customers to complete the field-based input of business requirements through drop-down menus and other methods without manually entering complex parameters. It achieves accurate and dynamic matching between front-end collected fields and back-end business logic, improving the completeness of information collection while avoiding irrelevant fields from interfering with users, thus enhancing the system's intelligence and user experience.

[0045] Optionally, before receiving the service type entered by the customer in the appointment system, the voice consultation information of the customer is received through the voice recognition module and converted into text information; a vague service description is determined based on the text information.

[0046] Optionally, the speech recognition module refers to the algorithm module that converts speech signals into text, and the text transcription result refers to the textualized output of the speech consultation, such as "I need to process an overseas remittance".

[0047] For example, a speech recognition module transcribes the voice content of a customer's telephone inquiry (e.g., "overseas remittance") into text, generating a transcription result, such as "I need to process an overseas remittance." This transcription result is then used as a fuzzy business description and input into a natural language processing step to generate standardized business codes.

[0048] For example, a speech recognition module transcribes the voice content (e.g., "overseas remittance") from a customer's telephone inquiry into text, generating a transcription result (e.g., "I need to process an overseas remittance"). This transcription result is then used as a vague business description, and natural language processing is used to generate standardized business codes. In this embodiment, the combination of a speech recognition module and natural language processing allows the voice inquiry content to be incorporated into the business requirements parsing process.

[0049] In one possible embodiment, a speech recognition module acquires raw data of a customer's voice inquiry input via telephone, such as the customer's spoken message "overseas remittance." The speech recognition module then processes this voice signal, converting it into a corresponding text transcription result, for example, obtaining the text information "I need to process an overseas remittance." A natural language processing engine receives this text transcription result and uses it as input data for a vague business description. The natural language processing engine performs semantic parsing and intent recognition on the text information, outputting a corresponding standardized business code, for example, parsing "I need to process an overseas remittance" into the business code "cross-border remittance."

[0050] This embodiment utilizes the collaborative work of speech recognition and natural language processing to transform customer voice inquiries into standardized business codes. This solution effectively expands the input channels for business requirement parsing, enabling the system to support voice interaction, adapting to diverse customer usage habits, and enhancing the convenience and applicability of the appointment system.

[0051] S202. Perform natural language processing on the fuzzy business description input by the customer, parse it, and generate standardized business codes.

[0052] More specifically, fuzzy business descriptions refer to unstructured business descriptions provided by customers (e.g., "processing overseas funds"), while standardized business codes refer to mapping fuzzy descriptions to standard business type codes (e.g., "cross-border remittance"). For example, by performing natural language processing on the fuzzy business description (e.g., "processing overseas funds") input by the customer, a natural language processing model is used to parse the semantic intent of the fuzzy business description and generate a standardized business code (e.g., "cross-border remittance").

[0053] Optionally, natural language processing is performed on the fuzzy business description input by the customer to parse and generate standardized business codes. Specifically, this includes: using a natural language processing model to perform semantic understanding on the fuzzy business description and extracting the semantic feature vector of the fuzzy business description; matching the semantic feature vector with the feature vector corresponding to each standard business in the standard business code library based on similarity; and selecting the standard business with the highest similarity as the standardized business code based on the matching result.

[0054] Optionally, the natural language processing model refers to a semantic parsing model based on the Transformer architecture; the semantic vector refers to the text semantic feature vector extracted by the natural language processing model (e.g., the vector representation of "processing overseas funds"); and the standard business coding library refers to a database used to store all standard business types and their semantic vectors (e.g., {"cross-border remittance": [0.3, 0.7, 0.1]}).

[0055] For example, a semantic vector is extracted from a vague description of customer input (e.g., "processing overseas funds") using a natural language processing model to generate its vector representation, for example, [0.3, 0.7, 0.1]. Then, the similarity (e.g., cosine similarity) of this vector with semantic vectors in a standard business code library is calculated to determine the business code with the highest matching degree (e.g., "cross-border remittance") as the standardized business code.

[0056] In one possible embodiment, fuzzy business description text data (e.g., "processing overseas funds") input by the customer is obtained. A pre-trained natural language processing model is used to perform semantic understanding on the fuzzy business description to extract semantic feature vectors, for example, obtaining a vector representation [0.3, 0.7, 0.1]. Then, the semantic feature vectors are compared with the feature vectors corresponding to each standard business in the standard business coding library (e.g., calculating cosine similarity). Based on the similarity matching results, the standard business with the highest similarity is selected as the standardized business code (e.g., determining "cross-border remittance"). Finally, the standardized business code obtained in this embodiment is integrated as core identification information into the subsequently generated standardized demand tree, obtaining the business type basis of the demand tree, and providing an accurate business type basis for subsequent dynamic field loading and service duration prediction.

[0057] This embodiment achieves accurate conversion from vague descriptions to standard business codes through the collaborative processing of semantic feature extraction and similarity matching. Specifically, by using semantic feature extraction, a deeper understanding of the semantic connotation of customer expressions is achieved, effectively avoiding recognition errors caused by non-standard expressions. Furthermore, by employing a similarity matching mechanism, the corresponding business type is accurately selected from the standard business database, further improving the accuracy of the parsing results. Therefore, the solution in this embodiment enhances the appointment system's adaptability to diverse customer expressions and strengthens the robustness of business requirement parsing.

[0058] S203. By integrating structured fields with standardized business codes, a standardized requirement tree representing the complete business requirements of the customer is generated.

[0059] More specifically, a standardized requirement tree refers to a structured data model that includes business types, field names, and field values. For example, a standardized requirement tree containing business types, field names, and field values ​​is generated by integrating dynamically loaded structured fields with parsed standardized business codes. Throughout this process, the business type input serves as the trigger for dynamically loading fields, and the output of natural language processing serves as the core code of the requirement tree. The integration of structured fields and standardized business codes enhances the completeness and standardization of the requirement tree.

[0060] Optionally, after generating a standardized requirement tree representing the customer's complete business needs, text extraction is performed on the images of business materials uploaded by the customer using optical character recognition technology; the extracted text is compared with a preset compliance rule base to verify the completeness and compliance of the business materials; and the integrity status marker of the business materials in the standardized requirement tree is updated based on the verification results.

[0061] Optionally, the preset compliance rule base refers to a database of pre-set rules for storing compliance verification of materials (e.g., "the authorization letter must be stamped with the official seal"); the integrity status mark of business materials refers to a status field that identifies whether the business materials are complete (e.g., "complete" or "missing").

[0062] Optionally, an algorithm module for extracting text content from an image can be used, employing an optical character recognition module.

[0063] For example, an optical character recognition (OCR) module extracts text from business materials uploaded by the customer (e.g., authorization letters) to generate parsable text content. Then, the extracted text is compared with rules in a pre-defined compliance rule base (e.g., "Authorization letters must bear an official seal") to generate a material compliance report (e.g., "Authorization letter lacks an official seal"). Finally, the integrity status flags of the business materials in the standardized requirements tree are adjusted based on the material compliance report.

[0064] In one possible embodiment, the data acquisition and data processing steps specifically include: acquiring image data of business materials uploaded by the customer (e.g., a photo of an authorization letter) through an optical character recognition (OCR) module; then processing the image data using OCR technology to extract the text content; comparing the extracted text content with rules in a preset compliance rule base (e.g., "the authorization letter must be stamped with an official seal") to obtain the comparison result; and generating a material compliance report based on the comparison result, and updating the material integrity status marker in the standardized requirements tree, for example, marking "authorization letter" as "missing".

[0065] In one possible embodiment, the application steps after obtaining the data specifically include: using the updated integrity status marker of the business materials obtained in the above embodiment as an important environmental feature parameter, and incorporating it into the feature engineering of subsequent service duration prediction. When the integrity status marker of the business materials is "missing," the service duration prediction model will adjust the prediction duration accordingly to improve the accuracy of reflecting the time required for on-site material replenishment.

[0066] This embodiment achieves automated compliance verification of business materials by combining optical character recognition (OCR) technology with a compliance rule base. Specifically, OCR technology improves the efficiency of extracting the text content of the materials, the compliance rule base provides accurate verification standards, and dynamically updating the integrity status markers of the business materials allows the standardized requirement tree to reflect the compliance status of the materials in real time. Therefore, the above technical solution provides more comprehensive input features for subsequent service duration prediction, which helps to further improve the accuracy and practicality of the prediction results.

[0067] S204. Based on the standardized demand tree, predict the service duration of customers at bank branches.

[0068] Optionally, based on the standardized demand tree, the service duration of customers at bank branches is predicted, specifically including: extracting business features, customer features and environmental features based on the standardized demand tree, and constructing a feature vector; inputting the feature vector into a pre-trained service duration prediction model, and outputting the predicted value of the service duration.

[0069] Optionally, the service duration prediction model includes at least one of the following models: XGBoost regression model, LSTM time series network, and random forest model. Specifically, XGBoost regression model refers to a regression algorithm based on gradient boosting decision trees, LSTM time series network refers to a long short-term memory network used to process time series data, and random forest model refers to an ensemble learning model based on multiple decision trees.

[0070] Optionally, business characteristics refer to features related to business type (e.g., business type coding, complexity score); customer characteristics refer to features related to customer attributes (e.g., customer star rating, APP usage proficiency); environmental characteristics refer to features related to service environment (e.g., peak daily customer flow, teller on-duty rate); and the service duration prediction model refers to a model that integrates XGBoost regression and LSTM time series network.

[0071] For example, business features (e.g., business type code), customer features (e.g., customer star rating), and environmental features (e.g., peak daily passenger flow) are extracted from a standardized demand tree to generate a feature vector. Then, the feature vector is input into a service duration prediction model (e.g., XGBoost and LSTM) to output a predicted service duration.

[0072] In one possible embodiment, business features (e.g., business type coding, complexity rating), customer features (e.g., customer star rating, APP usage proficiency), and environmental features (e.g., daily peak customer flow, teller on-duty rate) are extracted from a standardized demand tree. These multi-dimensional business features are then normalized to construct a unified feature vector. This unified feature vector is input into a pre-trained service duration prediction model (e.g., a service duration prediction model combining an XGBoost regression model and an LSTM temporal network) to obtain the prediction result of the model (i.e., the predicted service duration), such as "22 minutes ± 3 minutes". Based on the predicted service duration obtained in this embodiment, the value is directly displayed to the customer through the appointment system interface. Based on this information, the customer's itinerary is arranged, and the information is simultaneously synchronized to the bank branch's back-end management system, providing data support for branch resource scheduling and manpower allocation.

[0073] This embodiment achieves intelligent prediction of service duration through the integrated application of multi-dimensional feature extraction and prediction models. Specifically, by fusing extracted business features, customer features, and environmental features, the service duration prediction model can comprehensively perceive all dimensions of the service scenario. The machine learning-based prediction algorithm captures the complex nonlinear relationship between features and duration, improving the scenario adaptability and accuracy of service duration prediction and effectively reducing prediction bias caused by incomplete feature consideration.

[0074] Optionally, after updating the integrity status marker of business materials in the standardized requirements tree, the integrity status marker is used as an environmental feature; the constructed feature vector is updated based on the environmental feature; and the service duration is re-predicted based on the updated feature vector.

[0075] Optionally, the feature vector can be processed using an optional service duration prediction model (such as XGBoost regression, LSTM temporal network, or random forest) to output the predicted service duration. The prediction system dynamically adjusts the prediction algorithm or model based on business needs to adapt to the prediction accuracy requirements of different scenarios.

[0076] In one possible embodiment, a suitable service duration prediction model is selected from a model library based on the characteristics of the business scenario. For example, in scenarios with obvious time-series characteristics, a service duration prediction model based on an LSTM time-series network is selected; in scenarios with predominantly structured features, a service duration prediction model based on an XGBoost regression model is selected. Then, the input features are preprocessed and feature-engineered according to the characteristics of the selected service duration prediction model to obtain feature data. The processed feature data is then input into the selected service duration prediction model for calculation to obtain the prediction result. The prediction results of one or more service duration prediction models are obtained, and the predicted service duration value is output based on the prediction results. Finally, the service resource configuration module performs service resource scheduling based on the obtained predicted service duration value, and simultaneously evaluates and optimizes the performance of each service duration prediction model based on the prediction results of each model.

[0077] This embodiment achieves flexible allocation of prediction algorithm resources by establishing a multi-model selection mechanism. Specifically, the XGBoost regression model is selected when handling regression prediction tasks with structured features; the LSTM time series network is selected when capturing time series patterns; and the random forest model is selected when ensemble learning of nonlinear features is required. Therefore, the above technical solution enables the service duration prediction model to select a matching prediction algorithm according to the characteristics of different business scenarios, improving the adaptability and accuracy of the service duration prediction model and meeting the prediction needs of diverse business scenarios.

[0078] The service duration determination method provided in this application improves the completeness of business requirement parsing by automatically loading the corresponding structured field configuration based on the customer's selected business type, avoiding incomplete information issues caused by missing fields. Simultaneously, it utilizes a natural language processing model to semantically parse the fuzzy descriptions input by the customer, accurately mapping them to standardized business codes, thereby eliminating misjudgments of business types due to non-standard expressions. Finally, by integrating the structured fields and standardized business codes, a unified standardized requirement tree is formed, providing high-quality, structured input features for subsequent service duration prediction models. This method achieves automated and accurate parsing of business requirements, eliminating the need for customers to repeat descriptions or supplement information, thus improving the overall system processing efficiency and user experience.

[0079] Figure 3 A schematic diagram of the device for determining service duration provided in this application is shown below. Figure 3 As shown, the service duration determination device 30 provided in this embodiment includes:

[0080] The acquisition module 301 is used to receive the business type entered by the customer in the appointment system; and dynamically load the corresponding structured fields according to the business type.

[0081] The processing module 302 is used to perform natural language processing on the fuzzy business description input by the customer, parse it and generate standardized business codes;

[0082] The processing module 302 is also used to generate a standardized requirement tree that represents the complete business requirements of the customer by integrating structured fields with standardized business codes;

[0083] The processing module 302 is also used to predict the service duration of customers at bank branches based on a standardized demand tree.

[0084] Optionally, the acquisition module 301 is also used to map the business type to a predefined business type code;

[0085] Based on the field mapping table, query and retrieve the field configuration information corresponding to the business type code;

[0086] By rendering field configuration information into an interactive form component, user interface elements are generated based on the interactive form component for customer input.

[0087] Optionally, the processing module 302 is further configured to use a natural language processing model to perform semantic understanding on the fuzzy business description and extract the semantic feature vector of the fuzzy business description;

[0088] Perform similarity matching between the semantic feature vector and the feature vectors corresponding to each standard service in the standard service coding library;

[0089] Based on the matching results, the standard business with the highest similarity is selected as the standardized business code.

[0090] Optionally, the processing module 302 is also used to extract text from the images of business materials uploaded by the customer using optical character recognition technology after generating a standardized requirement tree that represents the customer's complete business requirements.

[0091] The extracted text is compared with a pre-defined compliance rule base to verify the completeness and compliance of business materials;

[0092] Based on the verification results, update the integrity status markers of business materials in the standardized requirements tree.

[0093] Optionally, the processing module 302 is also used to extract business features, customer features and environmental features based on the standardized demand tree, and construct feature vectors;

[0094] Input the feature vector into the pre-trained service duration prediction model and output the predicted service duration value.

[0095] Optionally, the processing module 302 is also used to receive the customer's voice consultation information through the voice recognition module and convert the voice consultation information into text information before receiving the business type entered by the customer in the appointment system;

[0096] Determine fuzzy business descriptions based on textual information.

[0097] Optionally, the processing module 302 is further configured to, after updating the integrity status marker of the business material in the standardized requirement tree, treat the integrity status marker as an environmental feature; and update the constructed feature vector based on the environmental feature.

[0098] Based on the updated feature vectors, the service duration is re-predicted.

[0099] The device for determining service duration provided in this embodiment can execute the method provided in the above method embodiment. Its implementation principle and technical effect are similar, and will not be described in detail here.

[0100] Figure 4 A schematic diagram of the structure of the electronic device provided in this application. Figure 4 As shown, the electronic device 40 provided in this embodiment includes at least one processor 401 and a memory 402. Optionally, the device 40 further includes a communication component 403. The processor 401, memory 402, and communication component 403 are connected via a bus 404.

[0101] In a specific implementation, at least one processor 401 executes computer execution instructions stored in memory 402, causing at least one processor 401 to perform the above-described method.

[0102] The specific implementation process of processor 401 can be found in the above method embodiments, and its implementation principle and technical effect are similar. It will not be repeated here.

[0103] In the above embodiments, it should be understood that the processor can be a Central Processing Unit (CPU), or other general-purpose processors, digital signal processors (DSPs), application-specific integrated circuits (ASICs), etc. The general-purpose processor can be a microprocessor or any conventional processor. The steps of the method disclosed in this invention can be directly implemented by a hardware processor, or implemented by a combination of hardware and software modules within the processor.

[0104] The memory may include random access memory (RAM) and may also include non-volatile memory (NVM), such as at least one disk storage device.

[0105] The bus can be an Industry Standard Architecture (ISA) bus, a Peripheral Component Interconnect (PCI) bus, or an Extended Industry Standard Architecture (EISA) bus, etc. Buses can be categorized as address buses, data buses, control buses, etc. For ease of illustration, the buses shown in the accompanying drawings are not limited to a single bus or a single type of bus.

[0106] This application also provides a computer program product, including a computer program that, when executed by a processor, implements the above-described method.

[0107] This application also provides a computer-readable storage medium storing computer-executable instructions, which, when executed by a processor, implement the above-described method.

[0108] The aforementioned readable storage medium can be implemented by any type of volatile or non-volatile storage device or a combination thereof, such as static random access memory (SRAM), electrically erasable programmable read-only memory (EEPROM), erasable programmable read-only memory (EPROM), programmable read-only memory (PROM), read-only memory (ROM), magnetic storage, flash memory, magnetic disk, or optical disk. The readable storage medium can be any available medium accessible to a general-purpose or special-purpose computer.

[0109] An exemplary readable storage medium is coupled to a processor, enabling the processor to read information from and write information to the readable storage medium. Of course, the readable storage medium can also be a component of the processor. The processor and the readable storage medium can reside in an Application Specific Integrated Circuit (ASIC). Alternatively, the processor and the readable storage medium can exist as discrete components in the device.

[0110] It should be noted that, for the sake of simplicity, the foregoing method embodiments are all described as a series of actions. However, those skilled in the art should understand that this application is not limited to the described order of actions, as some steps may be performed in other orders or simultaneously according to this application. Furthermore, those skilled in the art should also understand that the embodiments described in the specification are all optional embodiments, and the actions and modules involved are not necessarily essential to this application.

[0111] It should be further noted that although the steps in the flowchart are shown sequentially according to the arrows, these steps are not necessarily executed in the order indicated by the arrows. Unless explicitly stated herein, there is no strict order restriction on the execution of these steps, and they can be executed in other orders. Moreover, at least some steps in the flowchart may include multiple sub-steps or multiple stages. These sub-steps or stages are not necessarily completed at the same time, but can be executed at different times. The execution order of these sub-steps or stages is not necessarily sequential, but can be performed alternately or in turn with other steps or at least some of the sub-steps or stages of other steps.

[0112] It should be understood that the above-described device embodiments are merely illustrative, and the device of this application can also be implemented in other ways. For example, the division of units / modules in the above embodiments is only a logical functional division, and there may be other division methods in actual implementation. For example, multiple units, modules, or components may be combined, or integrated into another system, or some features may be ignored or not executed.

[0113] Furthermore, unless otherwise specified, the functional units / modules in the various embodiments of this application can be integrated into one unit / module, or each unit / module can exist physically separately, or two or more units / modules can be integrated together. The integrated units / modules described above can be implemented in hardware or as software program modules.

[0114] When integrated units / modules are implemented in hardware, the hardware can be digital circuits, analog circuits, etc. The physical implementation of the hardware structure includes, but is not limited to, transistors, memristors, etc. Unless otherwise specified, the processor can be any suitable hardware processor, such as a CPU, GPU, FPGA, DSP, and ASIC, etc. Unless otherwise specified, the storage unit can be any suitable magnetic or magneto-optical storage medium, such as Resistive Random Access Memory (RRAM), Dynamic Random Access Memory (DRAM), Static Random Access Memory (SRAM), Enhanced Dynamic Random Access Memory (EDRAM), High-Bandwidth Memory (HBM), Hybrid Memory Cube (HMC), etc.

[0115] If the integrated unit / module is implemented as a software program module and sold or used as an independent product, it can be stored in a computer-readable storage device (CMD). Based on this understanding, the technical solution of this application, in essence, or the part that contributes to the prior art, or all or part of the technical solution, can be embodied in the form of a software product. This computer software product is stored in a memory and includes several instructions to cause a computer device (which may be a personal computer, server, or network device, etc.) to execute all or part of the steps of the methods of the various embodiments of this application. The aforementioned memory includes various media capable of storing program code, such as a USB flash drive, read-only memory (ROM), random access memory (RAM), portable hard drive, magnetic disk, or optical disk.

[0116] In the above embodiments, the descriptions of each embodiment have their own emphasis. For parts not described in detail in a certain embodiment, please refer to the relevant descriptions of other embodiments. The technical features of the above embodiments can be combined arbitrarily. For the sake of brevity, not all possible combinations of the technical features in the above embodiments are described. However, as long as the combination of these technical features does not contradict each other, it should be considered within the scope of this specification.

[0117] Other embodiments of this application will readily occur to those skilled in the art upon consideration of the specification and practice of the invention disclosed herein. This application is intended to cover any variations, uses, or adaptations of this application that follow the general principles of this application and include common knowledge or customary techniques in the art not disclosed herein. The specification and examples are to be considered exemplary only, and the true scope and spirit of this application are indicated by the following claims.

[0118] It should be understood that this application is not limited to the precise structure described above and shown in the accompanying drawings, and various modifications and changes can be made without departing from its scope. The scope of this application is limited only by the appended claims.

Claims

1. A service duration determination method, characterized in that, receiving a service type input by a customer in a reservation system; dynamically loading a corresponding structured field according to the service type; performing natural language processing on a fuzzy service description input by the customer, and analyzing and generating a standardized service code; integrating the structured field and the standardized service code to generate a standardized demand tree representing the complete service demand of the customer; predicting the service duration of the customer at a bank outlet according to the standardized demand tree.

2. The method of claim 1, wherein, The dynamic loading of the corresponding structured field according to the service type specifically includes: mapping the service type to a predefined service type code; querying and obtaining the field configuration information corresponding to the service type code based on a field mapping table; rendering the field configuration information into an interactive form component, and generating user interface elements based on the interactive form component for customer input.

3. The method of claim 1, wherein, The natural language processing on the fuzzy service description input by the customer, and the analysis and generation of the standardized service code specifically include: using a natural language processing model to perform semantic understanding on the fuzzy service description, and extracting a semantic feature vector of the fuzzy service description; performing similarity matching between the semantic feature vector and the feature vector corresponding to each standard service in a standard service code library; selecting the standard service with the highest similarity as the standardized service code according to the matching result.

4. The method of claim 1, wherein, Further comprising: after generating the standardized demand tree representing the complete service demand of the customer, performing text extraction on the business material image uploaded by the customer through optical character recognition technology; comparing the extracted text with a preset compliance rule library to verify the completeness and compliance of the business material; updating the completeness status mark of the business material in the standardized demand tree according to the verification result.

5. The method of claim 1, wherein, The prediction of the service duration of the customer at the bank outlet according to the standardized demand tree specifically includes: extracting business features, customer features and environmental features based on the standardized demand tree, and constructing a feature vector; inputting the feature vector into a pre-trained service duration prediction model to output a predicted value of the service duration.

6. The method of claim 1, wherein, Further comprising: before receiving the service type input by the customer in the reservation system, receiving the voice consultation information of the customer through a voice recognition module, and converting the voice consultation information into text information; determining the fuzzy service description based on the text information.

7. The method of claim 4, wherein, Further comprising: after updating the completeness status mark of the business material in the standardized demand tree, taking the completeness status mark as an environmental feature; updating the constructed feature vector based on the environmental feature; re-predicting the service duration based on the updated feature vector.

8. A service duration determination apparatus characterized by comprising: Comprising: a receiving module configured to receive a service type input by a customer in a reservation system; and dynamically load a corresponding structured field according to the service type; a processing module configured to perform natural language processing on a fuzzy service description input by the customer, and analyze and generate a standardized service code; the processing module is further configured to integrate the structured field and the standardized service code to generate a standardized demand tree representing the complete service demand of the customer; and The processing module is further configured to predict the service duration of the customer at the bank outlet according to the standardized demand tree.

9. An electronic device, comprising: The method comprises: a processor, and a memory connected to the processor in communication; the memory stores computer-executable instructions; the processor executes the computer-executable instructions stored in the memory to implement the method of 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, and the computer-executable instructions are executed by the processor to implement the method of any one of claims 1 to 7.

11. A computer program product, characterised in that, The computer program is executed by the processor to implement the method of any one of claims 1 to 7.

Images