Service processing method and device, storage medium and electronic equipment

Abstract

The invention discloses a service processing method and device, a storage medium and electronic equipment, and the method comprises the steps: obtaining real-time service index data through a service index calculation data flow corresponding to a target service evaluation index in at least one target service scene, calculating benchmark service index data according to offline service index calculation data, and carrying out the calculation of the benchmark service index data according to the benchmark service index data; calculating online service index data according to the online service index calculation data, and calling an index detection large model to carry out index prediction processing to obtain predicted service index data in a target future time period, and generating service index analysis data based on the real-time service index data, the reference service index data, the online service index data and the prediction service index data, performing association storage on the service index analysis data and the target service evaluation index, and outputting and displaying the service index analysis data. Accurate actual index data with high real-time performance and future index trends are summarized and displayed.

Description

Technical Field

[0001] This specification relates to the field of computer technology, and in particular to a service processing method, device, storage medium, and electronic device. Background Art

[0002] With the rapid development of the internet, the Internet of Things, finance, healthcare, and other fields, data volumes are growing exponentially. This is particularly true on platforms like search engines, social media, and finance, where the demand for real-time or near-real-time processing of billions of data points continues to grow. This is because statistical analysis of billions of data points can minimize latency while ensuring data consistency, thereby providing more timely and accurate decision support.

[0003] In related technologies, statistical analysis of data volumes exceeding 100 million involves analyzing and calculating event streams generated by these volumes to obtain service metrics at different time units, such as real-time or near-real-time metrics. Therefore, for those skilled in the art, efficiently and accurately analyzing these metrics is a pressing technical challenge. Summary of the Invention

[0004] This specification provides a service processing method, device, storage medium, and electronic device. The technical solution is as follows: In a first aspect, this specification provides a service processing method, the method comprising: Monitor a service indicator calculation data stream corresponding to a target service evaluation indicator in at least one target service scenario, and perform stream computing on newly added service indicator calculation data in the service indicator calculation data stream to obtain real-time service indicator data; Calculate the benchmark service indicator data based on the offline service indicator calculation data corresponding to the target service evaluation indicator in the offline data warehouse, and calculate the online service indicator data based on the online service indicator calculation data corresponding to the target service evaluation indicator in the online data warehouse; Based on the real-time service indicator data, the benchmark service indicator data and the online service indicator data, calling the indicator detection model to perform indicator prediction processing to obtain predicted service indicator data within a target future time period; Service indicator parsing data is generated based on the real-time service indicator data, the benchmark service indicator data, the online service indicator data and the predicted service indicator data, the service indicator parsing data is associated with the target service evaluation indicator and stored, and the service indicator parsing data is output and displayed.

[0005] In a second aspect, this specification provides a service processing device, the device comprising: A first indicator calculation module is used to monitor a service indicator calculation data stream corresponding to a target service evaluation indicator in at least one target service scenario, and perform stream computing on newly added service indicator calculation data in the service indicator calculation data stream to obtain real-time service indicator data; A second indicator calculation module is configured to calculate the benchmark service indicator data based on the offline service indicator calculation data corresponding to the target service evaluation indicator in the offline data warehouse, and calculate the online service indicator data based on the online service indicator calculation data corresponding to the target service evaluation indicator in the online data warehouse; A third indicator calculation module is used to call the indicator detection model to perform indicator prediction processing based on the real-time service indicator data, the benchmark service indicator data and the online service indicator data to obtain predicted service indicator data within a target future time period; The indicator analysis and processing module is used to generate service indicator analysis data based on the real-time service indicator data, the benchmark service indicator data, the online service indicator data and the predicted service indicator data, associate the service indicator analysis data with the target service evaluation indicator and store it, and output and display the service indicator analysis data.

[0006] In a third aspect, this specification provides a computer storage medium having a plurality of instructions, wherein the instructions are suitable for being loaded by a processor and executing the above method.

[0007] In a fourth aspect, this specification provides a computer program product, wherein the computer program product stores at least one instruction, and the at least one instruction is loaded by a processor to execute the above method.

[0008] In a fifth aspect, this specification provides an electronic device, which may include: a memory and a processor; wherein the memory stores a computer program, and the computer program is suitable for being loaded by the memory and executing the above method.

[0009] The beneficial effects of the technical solutions provided in this specification include at least: The service processing method provided in the embodiments of this specification first monitors the service indicator calculation data stream corresponding to the target service evaluation indicator in at least one target service scenario, performs stream calculation processing on the newly added service indicator calculation data in the service indicator calculation data stream to obtain real-time service indicator data, calculates benchmark service indicator data based on the offline service indicator calculation data corresponding to the target service evaluation indicator in the offline data warehouse, and calculates online service indicator data based on the online service indicator calculation data corresponding to the target service evaluation indicator in the online data warehouse. Afterwards, based on the real-time service indicator data, the benchmark service indicator data and the online service indicator data, the indicator detection large model is called to perform indicator prediction processing to obtain predicted service indicator data in the target future time period. In this way, offline calculation provides benchmark service indicator data, and online calculation and stream calculation ensure the accuracy and real-time performance of the indicator data. Timeliness, the hybrid computing mode composed of three indicator calculation methods improves the indicator calculation efficiency in big data processing. The large model can predict the indicator data in the future time period, that is, the indicator trend in the future time period can be predicted, thereby realizing the function of providing accurate and high-real-time indicator data and predicting future indicator data. Finally, based on the real-time service indicator data, benchmark service indicator data, online service indicator data and predicted service indicator data, the service indicator analysis data is generated, the service indicator analysis data is associated with the target service evaluation indicator and stored, and the service indicator analysis data is output and displayed. Therefore, by summarizing and displaying the accurate and high-real-time actual indicator data and future indicator trends, the efficient display of the indicator data is guaranteed, which helps relevant personnel understand and use the indicator data and improves the operability and practicality of the indicator data. BRIEF DESCRIPTION OF THE DRAWINGS

[0010] In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the following briefly introduces the drawings required for use in the embodiments or the description of the prior art. Obviously, the drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without paying any creative work.

[0011] Figure 1 This is a scenario diagram of a service processing system provided by an embodiment of this specification; Figure 2 This is a flowchart of a service processing method provided by an embodiment of this specification; Figure 3 This is a flowchart of another service processing method provided in an embodiment of this specification; Figure 4 This is a structural diagram of a service processing device provided in an embodiment of this specification; Figure 5This is a schematic diagram of the structure of an electronic device provided in an embodiment of this specification. DETAILED DESCRIPTION

[0012] In order to make the invention objectives, features, and advantages of the embodiments of this specification more obvious and easy to understand, the technical solutions in the embodiments of this specification will be clearly and completely described below in conjunction with the drawings in the embodiments of this specification. Obviously, the embodiments described are only part of the embodiments of this specification, not all of the embodiments. Based on the embodiments in this specification, all other embodiments obtained by those skilled in the art without making creative efforts shall fall within the scope of protection of this specification.

[0013] In the description of this specification, it should be understood that the terms "first", "second", etc. are used for descriptive purposes only and should not be understood as indicating or implying relative importance. In the description of this specification, it should be noted that, unless otherwise clearly specified and limited, "including" and "having" and any variations thereof are intended to cover non-exclusive inclusions. For example, a process, method, system, product or device that includes a series of steps or units is not limited to the listed steps or units, but optionally also includes steps or units that are not listed, or optionally also includes other steps or units inherent to these processes, methods, products or devices. For those of ordinary skill in the art, the specific meanings of the above terms in this specification can be understood according to the specific circumstances. In addition, in the description of this specification, unless otherwise specified, "multiple" refers to two or more. "And / or" describes the association relationship of associated objects, indicating that three relationships can exist. For example, A and / or B can represent: A exists alone, A and B exist at the same time, and B exists alone. The character " / " generally indicates that the associated objects before and after are in an "or" relationship.

[0014] See Figure 1 , is a schematic diagram of a scenario of a service processing system provided in an embodiment of this specification. Figure 1 As shown, the scenario diagram may include at least a terminal cluster and a server 100.

[0015] In some embodiments, the terminal cluster may include at least one terminal, such as Figure 1 As shown, it specifically includes terminal 1 corresponding to user 1, terminal 2 corresponding to user 2, ..., terminal n corresponding to user n, where n is an integer greater than 0.

[0016] Each terminal in the terminal cluster can be a smart device with communication capabilities, including but not limited to wearable devices, handheld devices, personal computers, tablet computers, smartphones, computing devices, or other processing devices connected to a wireless modem. Smart devices may be called different names in different networks, such as user equipment, access terminal, subscriber unit, subscriber station, mobile station, mobile station, remote station, remote terminal, mobile device, user terminal, terminal, wireless communication device, user agent or user device, cellular phone, cordless phone, personal digital assistant (PDA), and electronic devices in 5G networks or future evolution networks.

[0017] In some embodiments, the server 100 is a hardware device with strong computing capabilities. Specifically, the server 100 can use a separate server device, such as a rack-mounted, blade, tower, or cabinet-mounted server device, or a workstation, mainframe computer, or other hardware device; or a server cluster composed of multiple servers can be used. The servers in the service cluster can be symmetrically arranged, wherein each server has equivalent functions and status in the service link, and each server can provide services to the outside world independently. Providing services independently can be understood as not requiring the assistance of another server.

[0018] In some embodiments, the electronic device that executes the service processing method may be a server 100, and the server 100 may establish a communication connection with the terminal to complete data interaction during the service processing based on the communication connection. For example, in a service processing method, a terminal interacts with a service processing system to generate service indicator calculation data corresponding to a target service evaluation indicator. The service indicator calculation data generated in different time periods can constitute a service indicator calculation data stream, offline service indicator calculation data, and online service indicator calculation data. The server corresponding to the service processing system monitors the service indicator calculation data stream corresponding to the target service evaluation indicator in at least one target service scenario, performs stream computing on the newly added service indicator calculation data in the service indicator calculation data stream to obtain real-time service indicator data. The server calculates benchmark service indicator data based on the offline service indicator calculation data corresponding to the target service evaluation indicator in the offline data warehouse, and calculates online service indicator data based on the online service indicator calculation data corresponding to the target service evaluation indicator in the online data warehouse. The server calls an indicator detection model based on the real-time service indicator data, the benchmark service indicator data, and the online service indicator data to perform indicator prediction processing to obtain predicted service indicator data in the target future time period. Service indicator parsing data is generated based on the real-time service indicator data, the benchmark service indicator data, the online service indicator data, and the predicted service indicator data. The service indicator parsing data is associated and stored with the target service evaluation indicator, and the service indicator parsing data is output and displayed.

[0019] It should be noted that the server 100 and the terminal establish a communication connection through a network for interactive communication, wherein the network can be a wireless network or a wired network. Wireless networks include but are not limited to cellular networks, wireless local area networks, infrared networks, or Bluetooth networks, and wired networks include but are not limited to Ethernet, universal serial bus (USB), or controller area network. In one or more embodiments of the specification, technologies and / or formats including Hypertext Markup Language (HTML) and Extensible Markup Language (XML) are used to represent data exchanged over the network (such as target compressed packages). In addition, conventional encryption technologies such as Secure Sockets Layer (SSL), Transport Layer Security (TLS), Virtual Private Network (VPN), and Internet Protocol Security (IPsec) can also be used to encrypt all or some links. In other embodiments, customized and / or dedicated data communication technologies can also be used to replace or supplement the above data communication technologies.

[0020] The service processing system embodiments provided in this specification share the same concept as the service processing methods described in one or more embodiments. The execution entity corresponding to the service processing methods described in one or more embodiments of this specification may be an electronic device, which may be the aforementioned server 100. The specific implementation process of the service processing system embodiments can be found in the following method embodiments and will not be further elaborated here.

[0021] In one embodiment, Figure 2 As shown, a service processing method is proposed. The method can be implemented by a computer program and can be run on a service processing device based on the von Neumann architecture. The computer program can be integrated into an application or run as an independent tool application.

[0022] Specifically, the service processing method includes: S202: Monitor a service indicator calculation data stream corresponding to a target service evaluation indicator in at least one target service scenario, and perform stream computing on newly added service indicator calculation data in the service indicator calculation data stream to obtain real-time service indicator data.

[0023] Among them, the target service evaluation indicators are used to measure the service processing effect in the service activities involved in providing specific services on the target service platform.

[0024] It can be understood that the target service scenario refers to the specific service activity scenario involved in the target service platform when providing specific services. For example, the target service platform may include but is not limited to a search engine platform, a financial platform, a social media platform, etc. For example, in a search engine platform, the target service scenario may specifically be the search service scenario involved in the search engine platform. For example, in a financial platform, the target service scenario may specifically be the financial service scenario involved in the financial platform. For example, in a social media platform, the target service scenario may specifically be the social service scenario involved in the social media platform. In different target service scenarios, there can usually be at least one target service evaluation indicator. These target service evaluation indicators are an indicator used to measure the service processing effect in the service activities involved when the target service platform provides specific services.

[0025] Taking the target service scenario as an example of a search service scenario, the target service evaluation indicators may include but are not limited to search volume, number of clicks, page views, query conversion rate, etc. For example, the search volume can be used to reflect the intensity of users' demand for the search service provided by the search platform, and can then be used to measure the service processing effect of the search service provided by the search platform. For another example, the number of clicks can measure the user's interest in a certain page, and can reflect the attractiveness of the page or content to the user. The page views can be used to measure the user's attention to the content of a specific page. The number of clicks and page views can be used to analyze which pages are popular and which content users are more inclined to view, thereby being used to measure the content layout effect and page structure optimization effect in the search service provided by the search platform, which helps the platform improve the content layout, page structure, etc. in the search service.

[0026] Taking the target service scenario as a financial services scenario as an example, target service evaluation indicators may include, but are not limited to, the number of user registrations, the number of user visits, the number of active users, transaction volume, and transaction amount. For example, at least one of the following indicators can be used to measure the intensity of user demand for financial products or services, and thus the service processing effectiveness of financial services. Specifically, for newly launched financial services on a financial platform, the number of user registrations, the number of user visits, the number of active users, the transaction volume, and the transaction amount can be used to measure the service innovation effectiveness of the newly launched financial services.

[0027] Taking the target service scenario of a social service as an example, target service evaluation indicators may include, but are not limited to, content readership, video play volume, and the amount of user-generated content. For example, content readership and video play volume can be used to reflect the exposure of content or videos and can be used to measure the effectiveness of social service delivery services. Another example is the amount of user-generated content that can reflect users' fondness for the platform or their popularity, and can be used to measure the effectiveness of internal publishing services within social services.

[0028] The service indicator calculation data stream includes the continuously generated calculation data used to calculate the target service evaluation indicators. For example, in the target service scenario, the calculation data required to calculate the target service evaluation indicators will be continuously generated. These continuously generated calculation data constitute the service indicator calculation data.

[0029] New service indicator calculation data can be understood as service indicator calculation data located after the reference time in the service indicator calculation data stream. The reference time can be the time when the previous real-time service indicator data is obtained. Since the real-time service indicator data is the near-real-time value of the target service evaluation indicator, when new real-time service indicator data is obtained, the real-time service indicator data adjacent to the new real-time service indicator data and generated before the new real-time service indicator data is the previous real-time service indicator data. The aforementioned previous real-time service indicator data and new real-time service indicator data can be understood as the values ​​of the same service evaluation indicator at different times. For example, if the target service evaluation indicator is transaction volume, the previous real-time service indicator data can be a transaction volume with a value of 10,000, and the new real-time service indicator data can be a transaction volume with a value of 12,000.

[0030] In some embodiments, the implementation method of monitoring the service indicator calculation data flow corresponding to the target service evaluation indicator in at least one target service scenario can be specifically as follows: for each target service scenario, at least one data source associated with at least one target service scenario can be monitored through a data collection tool, and the original service indicator calculation data flow corresponding to the target service evaluation indicator can be collected from at least one data source, and the original service indicator calculation data flow can be cleaned to obtain the service indicator calculation data flow.

[0031] Specifically, the at least one data source associated with the target service scenario may be at least one service processing system associated with the target service scenario. For example, taking the target service scenario as a financial services scenario, the at least one data source associated with the financial services scenario may include other service processing systems such as transaction systems and user behavior systems.

[0032] The original service indicator calculation data stream is cleansed to obtain the service indicator calculation data stream. This can be achieved by using a preset data access tool to perform data cleaning operations such as deduplication, filling missing values, and detecting outliers on the original service indicator calculation data stream to ensure a high-quality service indicator calculation data stream. The preset data access tool can support multiple data formats and protocols and can perform data cleaning on original service indicator calculation data streams from different data sources.

[0033] Performing stream computing on the newly added service indicator calculation data in the service indicator calculation data stream to obtain real-time service indicator data can be understood as performing stream computing on the newly added service indicator calculation data in the service indicator calculation data stream using a preset stream computing method to obtain real-time service indicator data.

[0034] S204 , calculating benchmark service indicator data based on offline service indicator calculation data corresponding to the target service evaluation indicator in the offline data warehouse, and calculating online service indicator data based on online service indicator calculation data corresponding to the target service evaluation indicator in the online data warehouse.

[0035] An offline data warehouse refers to a data warehouse system that stores and analyzes data in batches. An online data warehouse refers to a data warehouse system that can process and query data online in real time. Typically, data in an offline data warehouse is not updated in real time, but rather processed and loaded in batches periodically. In contrast, data in an online data warehouse is updated in real time or near real time. In some embodiments, offline service indicator calculation data for a target service evaluation indicator within a first preset time interval can be obtained from an offline data warehouse, and then offline indicator calculation processing can be performed on the offline service indicator calculation data to obtain baseline service indicator data. The baseline service indicator data can be understood as the value of the target service evaluation indicator within the first preset time interval. Online service indicator calculation data for the target service evaluation indicator within a second preset time interval can be obtained from an online data warehouse, and then online indicator calculation processing can be performed on the online service indicator calculation data to obtain online service indicator data. The online service indicator data can be understood as the value of the target service evaluation indicator within the second preset time interval. The first preset time interval is greater than the second preset time interval.

[0036] S206: Based on the real-time service indicator data, the benchmark service indicator data and the online service indicator data, the indicator detection large model is called to perform indicator prediction processing to obtain predicted service indicator data within the target future time period.

[0037] It's understood that the indicator detection large model can directly use the basic Large Language Model (LLM) or a large language model obtained by training the basic LLM for the indicator detection scenario. An LLM is an artificial intelligence content generation model designed to understand and generate human language. Trained on large amounts of data, LLM can perform a wide range of tasks, including text summarization, translation, sentiment analysis, and more. Alternatively, an LLM can be a generative artificial intelligence (AIGC) model.

[0038] In some embodiments, a prompt word is generated to indicate that the indicator detection large model predicts the predicted service indicator data of the target service evaluation indicator in the target future time period through the real-time service indicator data, the benchmark service indicator data and the online service indicator data. By inputting the prompt word into the indicator detection large model, the indicator detection large model can obtain the predicted service indicator data. The predicted service indicator data can be understood as the numerical value of the target service evaluation indicator in the target future time period. It is understandable that the predicted service indicator data can include the indicator numerical value of the service evaluation indicator in one or more target future time periods. For example, the predicted service indicator data can include the numerical value of the target service evaluation indicator in the next hour, the numerical value in the next two hours, the numerical value in the next day, the numerical value in the next week, and so on.

[0039] S208, generating service indicator analysis data based on the real-time service indicator data, the benchmark service indicator data, the online service indicator data and the predicted service indicator data, associating the service indicator analysis data with the target service evaluation indicator and storing the service indicator analysis data, and outputting and displaying the service indicator analysis data.

[0040] In some embodiments, the real-time service indicator data, the benchmark service indicator data, the online service indicator data, and the target service indicator data can be merged according to the timestamp to obtain the merged indicator data, obtain the preset service indicator display form, and use the preset service indicator display form to perform indicator analysis on the merged indicator data to obtain service indicator parsing data. After the service indicator parsing data is generated, the service indicator parsing data can be associated with the target service evaluation indicator and stored. Specifically, the mapping relationship between the storage service indicator parsing data and the storage service indicator parsing data and the target service evaluation indicator can be stored in the preset storage space. After the service indicator parsing data is generated, the service indicator parsing data can also be output and displayed.

[0041] Optionally, in one embodiment of the service processing method provided in this specification, taking the target service scenario as a search service scenario as an example, executing the service processing method may specifically include: monitoring the service indicator calculation data stream corresponding to the search service evaluation indicator in at least one search service scenario, performing stream calculation processing on the newly added service indicator calculation data in the service indicator calculation data stream to obtain real-time service indicator data; calculating benchmark service indicator data based on the offline service indicator calculation data corresponding to the search service evaluation indicator in the offline data warehouse, and calculating online service indicator data based on the online service indicator calculation data corresponding to the search service evaluation indicator in the online data warehouse; based on the real-time service indicator data, the benchmark service indicator data, and the online service indicator data, calling the indicator detection large model to perform indicator prediction processing to obtain predicted service indicator data for the target future time period; generating service indicator parsing data based on the real-time service indicator data, the benchmark service indicator data, the online service indicator data, and the predicted service indicator data, associating the service indicator parsing data with the search service evaluation indicator for storage, and outputting and displaying the service indicator parsing data. In this way, by generating service indicator parsing data associated with the search service evaluation indicator, it is helpful to evaluate the effectiveness of a specific service in the search service.

[0042] Optionally, in another embodiment of the service processing method provided in this specification, taking the target service scenario as a financial service scenario as an example, executing the service processing method may specifically include: monitoring the service indicator calculation data stream corresponding to the financial service evaluation indicator in at least one financial service scenario, performing stream computing processing on the newly added service indicator calculation data in the service indicator calculation data stream to obtain real-time service indicator data; calculating benchmark service indicator data based on the offline service indicator calculation data corresponding to the financial service evaluation indicator in the offline data warehouse, and calculating online service indicator data based on the online service indicator calculation data corresponding to the financial service evaluation indicator in the online data warehouse; based on the real-time service indicator data, the benchmark service indicator data, and the online service indicator data, calling the indicator detection large model to perform indicator prediction processing to obtain predicted service indicator data for the target future time period; generating service indicator parsing data based on the real-time service indicator data, the benchmark service indicator data, the online service indicator data, and the predicted service indicator data, associating the service indicator parsing data with the search service evaluation indicator for storage, and outputting and displaying the service indicator parsing data. In this way, by generating service indicator parsing data associated with the financial service evaluation indicator, it is helpful to evaluate the effectiveness of a specific service in the financial service.

[0043] Optionally, in another embodiment of the service processing method provided in this specification, taking the target service scenario as a social service scenario as an example, executing the service processing method may specifically include: monitoring a service indicator calculation data stream corresponding to a social service evaluation indicator in at least one social service scenario, performing stream computing processing on newly added service indicator calculation data in the service indicator calculation data stream to obtain real-time service indicator data; calculating benchmark service indicator data based on the offline service indicator calculation data corresponding to the social service evaluation indicator in the offline data warehouse, and calculating online service indicator data based on the online service indicator calculation data corresponding to the social service evaluation indicator in the online data warehouse; based on the real-time service indicator data, the benchmark service indicator data, and the online service indicator data, calling an indicator detection model to perform indicator prediction processing to obtain predicted service indicator data for a target future time period; generating service indicator parsing data based on the real-time service indicator data, the benchmark service indicator data, the online service indicator data, and the predicted service indicator data, associating the service indicator parsing data with the search service evaluation indicator for storage, and outputting and displaying the service indicator parsing data. In this way, by generating service indicator parsing data associated with the social service evaluation indicator, it is helpful to evaluate the effectiveness of a specific service in the social service.

[0044] The service processing method provided in the embodiments of this specification first monitors the service indicator calculation data stream corresponding to the target service evaluation indicator in at least one target service scenario, performs stream calculation processing on the newly added service indicator calculation data in the service indicator calculation data stream to obtain real-time service indicator data, calculates benchmark service indicator data based on the offline service indicator calculation data corresponding to the target service evaluation indicator in the offline data warehouse, and calculates online service indicator data based on the online service indicator calculation data corresponding to the target service evaluation indicator in the online data warehouse. Then, based on the real-time service indicator data, the benchmark service indicator data and the online service indicator data, the indicator detection large model is called to perform indicator prediction processing to obtain predicted service indicator data in the target future time period. In this way, offline calculation provides benchmark service indicator data, and online calculation and stream calculation ensure the accuracy and real-time performance of the indicator data. The hybrid computing mode composed of the three indicator calculation methods improves the indicator calculation efficiency in big data processing. The large model can be used to predict the indicator data in the future time period, that is, the indicator trend in the future time period can be predicted, thereby realizing the function of providing accurate and high-real-time actual indicator data and predicting future indicator data. Finally, service indicator analysis data is generated based on real-time service indicator data, benchmark service indicator data, online service indicator data and predicted service indicator data, the service indicator analysis data is associated with the target service evaluation indicator and stored, and the service indicator analysis data is output and displayed. Therefore, by summarizing and displaying the accurate and high-real-time actual indicator data and future indicator trends, the efficient display of indicator data is guaranteed, which helps relevant personnel understand and use indicator data and improves the operability and practicality of indicator data.

[0045] See Figure 3 , is a flow chart of another embodiment of a service processing method provided in the embodiments of this specification. Specifically, the method may include the following steps: S302: Monitor a service indicator calculation data flow corresponding to a target service evaluation indicator in at least one target service scenario.

[0046] Specifically, the implementation of step S302 can be found in Figure 2 The description of the relevant parts in the illustrated embodiment will not be repeated in detail here.

[0047] S304: Monitor service indicator delay backlog parameters and determine service indicator calculation priorities.

[0048] Among them, the service indicator delay backlog parameter refers to a parameter used to characterize the delay of real-time target service evaluation indicators. Specifically, the service indicator delay backlog parameter may include the service indicator delay data volume and service indicator delay duration corresponding to the real-time service indicator data. The service indicator delay data volume refers to the data volume of service indicator calculation data waiting to be calculated when there is a calculation delay in the real-time service indicator data. The service indicator delay duration refers to the length of time that the real-time service indicator data is delayed in calculation.

[0049] For example, taking the target evaluation service indicator as transaction volume, the real-time service indicator data is the new transaction volume in the current hour. Due to the sudden increase in a large amount of transaction data in the past two hours, the calculation of the real-time service indicator data consumes a long computing time. If the current time is 15:00, it is still stuck in calculating the real-time service indicator data in the time period from 13:00 to 14:00. There is currently a delay in the calculation of the real-time service indicator data, and the real-time service indicator data in the time period from 14:00 to 15:00 should be calculated. The amount of service indicator calculation data waiting to be calculated in the time period from 14:00 to 15:00 is 1 million data items. Therefore, there is a one-hour delay in calculating the real-time service indicator data, and the amount of data waiting to be calculated is 1 million data items. It can be confirmed that the service evaluation indicator delay corresponding to the real-time service indicator data is one hour, and it can be confirmed that the amount of service indicator calculation data waiting to be calculated is 1 million data items.

[0050] The service indicator calculation priority is a parameter used to represent the importance of the service evaluation indicator. The higher the service indicator calculation priority, the more important the service evaluation indicator is.

[0051] In some embodiments, monitoring of a service indicator delay backlog parameter may be implemented by monitoring the service indicator delay data volume and service indicator delay duration corresponding to real-time service indicator data in real time, and performing data normalization processing on the service indicator delay data volume and service indicator delay duration to obtain the service indicator delay backlog parameter. For example, the service indicator delay data volume may be adjusted to a service indicator delay data volume in a first standard format, and the service indicator delay duration may be adjusted to a service indicator delay duration in a second standard format, and the service indicator delay backlog parameter may be generated based on the service indicator delay data volume in the first standard format and the service indicator delay duration in the second standard format.

[0052] The service indicator calculation priority may be determined by obtaining a service indicator identifier and querying the service indicator calculation priority corresponding to the service indicator identifier from a preset priority mapping relationship. The preset priority mapping relationship may include at least one reference service indicator identifier and the service indicator calculation priority corresponding to each reference service indicator identifier.

[0053] S306 , detecting whether a service indicator delay backlog parameter is greater than or equal to a preset delay backlog threshold, and detecting whether a service indicator calculation priority is greater than or equal to a preset priority.

[0054] It can be understood that since the service indicator delay backlog parameter includes the service indicator delay data volume and service indicator delay duration corresponding to the real-time service indicator data, the preset delay backlog threshold may include a first preset delay backlog threshold and a second preset delay backlog threshold. When detecting whether the service indicator delay backlog parameter is greater than or equal to the preset delay backlog threshold, it may be to detect whether the service indicator delay data volume is greater than or equal to the first preset delay backlog threshold, or to detect whether the service indicator delay duration is greater than or equal to the second preset delay backlog threshold.

[0055] Specifically, executing step S306 may include: detecting whether the service indicator delay backlog parameter is greater than or equal to the preset delay backlog threshold, and at the same time, detecting whether the service indicator calculation priority is greater than or equal to the preset priority. Executing step S306 may also include: first detecting whether the service indicator delay backlog parameter is greater than or equal to the preset delay backlog threshold. Then detecting whether the service indicator calculation priority is greater than or equal to the preset priority. Executing step S306 may also include: first detecting whether the service indicator calculation priority is greater than or equal to the preset priority, and then detecting whether the service indicator delay backlog parameter is greater than or equal to the preset delay backlog threshold.

[0056] Specifically, when the amount of service indicator delayed data is greater than or equal to the first preset delay backlog threshold, or the service indicator delay duration is greater than or equal to the second preset delay backlog threshold, it is confirmed that the service indicator delay backlog parameter is greater than or equal to the preset delay backlog threshold. It is understandable that when the amount of service indicator delayed data is greater than or equal to the first preset delay backlog threshold, and the service indicator delay duration is less than the second preset delay backlog threshold, it is confirmed that the service indicator delay backlog parameter is greater than or equal to the preset delay backlog threshold. When the amount of service indicator delayed data is less than the first preset delay backlog threshold, and the service indicator delay duration is greater than or equal to the second preset delay backlog threshold, it is confirmed that the service indicator delay backlog parameter is greater than or equal to the preset delay backlog threshold. When the amount of service indicator delayed data is less than the first preset delay backlog threshold, and the service indicator delay duration is less than the second preset delay backlog threshold, it is confirmed that the service indicator delay backlog parameter is less than the preset delay backlog threshold.

[0057] S308, if the service indicator delay backlog parameter is greater than or equal to the preset delay backlog threshold, and the service indicator calculation priority is greater than or equal to the preset priority, then the newly added service indicator calculation data in the service indicator calculation data stream is processed for data clipping to obtain the target service indicator calculation data, and the target service indicator calculation data is processed for stream computing to obtain real-time service indicator data.

[0058] In some embodiments, data clipping processing is performed on the newly added service indicator calculation data in the service indicator calculation data stream to obtain target service indicator calculation data, which may specifically include the following steps: A2: determining the stream calculation data processing range in the newly added service indicator calculation data in the service indicator calculation data stream, A4: performing data clipping processing on the newly added service indicator calculation data based on the stream calculation data processing range to obtain target service indicator calculation data.

[0059] In step A2, a service indicator identifier is obtained, and a data length limit mapping relationship is obtained. The data length limit mapping relationship includes the data volume of the reference service indicator calculation data corresponding to the reference service indicator identifier and the reference service indicator delay backlog parameter. The data volume of the service indicator calculation data corresponding to the service indicator identifier and the service indicator delay backlog parameter is queried from the data length limit mapping relationship. According to the data volume of the service indicator calculation data, the stream calculation data processing range is determined in the newly added service indicator calculation data in the service indicator calculation data stream. Among them, the stream calculation data processing range includes the data starting position and data ending position determined in the newly added service indicator calculation data for stream calculation. Specifically, after determining the data volume of the service indicator calculation data, the latest data collected in the newly added service indicator calculation data can be used as the data ending position, the data ending position can be used as the end point of the stream calculation data processing range, the data volume of the service indicator calculation data can be used as the interval length of the stream calculation data processing range, and the data starting position can be determined in the newly added service indicator calculation data.

[0060] Specifically, the method for determining the new service indicator calculation data in the service indicator calculation data stream can be: using the service indicator calculation data located after the reference time in the service indicator calculation data stream as the new service indicator calculation data, and the reference time can be the time when the previous real-time service indicator is obtained. Since the real-time service indicator is a near-real-time service indicator, when a new real-time service indicator is obtained, the real-time service indicator adjacent to the new real-time service indicator and generated before the new real-time service indicator is the previous real-time service indicator.

[0061] In step A4, the target data interval indicated by the stream computing data processing range is determined in the newly added service index calculation data, and the data in the target data interval in the newly added service index calculation data is used as the target service index calculation data.

[0062] In some embodiments, stream computing is performed on the target service indicator calculation data to obtain real-time service indicator data, which may specifically include the following steps: B2: If the target service indicator calculation data belongs to all the data in the newly added service indicator calculation data, the stream computing hardware resources are expanded, and the expanded stream computing hardware resources are used to perform stream computing on the target service indicator calculation data to obtain real-time service indicator data; B4: If the target service indicator calculation data belongs to part of the data in the newly added service indicator calculation data, the original stream computing hardware resources are used to perform stream computing on the target service indicator calculation data to obtain real-time service indicator data.

[0063] In step B2, when the service indicator delay backlog parameter is greater than or equal to the preset delay backlog threshold, and the service evaluation indicator calculation priority is greater than or equal to the preset priority, it indicates that there is a certain delay in the calculation process of the real-time service indicator data of the important target service evaluation indicator. At this time, in order to solve the delay, if the target service indicator calculation data belongs to all the data in the newly added service indicator calculation data, the solution of expanding hardware resources is adopted, that is, the stream computing hardware resources are expanded, and the expanded stream computing hardware resources are used to perform stream computing on the target service indicator calculation data to obtain real-time service indicator data. Optionally, the hardware resources can be expanded by increasing the number of high-computing power devices participating in the stream computing, so that more high-computing power devices can be used to complete the stream computing together to solve the delay in obtaining real-time service indicator data by the stream computing.

[0064] In step B4, to address delays in calculating the real-time service indicator data for important service evaluation metrics, if the target service indicator calculation data is part of the newly added service indicator calculation data, this means that some of the newly added service indicator calculation data has been discarded, reducing the amount of data required for stream computing. In this case, the delay can be addressed by using existing hardware resources. Specifically, the original stream computing hardware resources are used to perform stream computing on the target service indicator calculation data to obtain the real-time service indicator data. The original stream computing hardware resources refer to the original hardware resources configured for stream computing to obtain the real-time service indicator data.

[0065] Optionally, if the service indicator delay backlog parameter is greater than or equal to the preset delay backlog threshold, and the service indicator calculation priority is less than the preset priority, the original stream computing hardware resources are used to perform stream computing on the newly added service indicator calculation data in the service indicator calculation data stream to obtain real-time service indicator data. It is understandable that in this scenario, there is a delay in the calculation process of the real-time service indicator data of non-important service evaluation indicators. Since non-important service evaluation indicators are temporarily delayed, the real-time requirements of non-important service evaluation indicators are not high, so as to avoid wasting hardware resources, the original hardware resources can be used for stream computing to obtain real-time service indicator data.

[0066] Optionally, if the service indicator delay backlog parameter is less than a preset delay backlog threshold, and the service indicator calculation priority is greater than or equal to the preset priority, the original stream computing hardware resources are used to perform stream computing on the newly added service indicator calculation data in the service indicator calculation data stream to obtain real-time service indicator data. It is understandable that in this scenario, the calculation process of the real-time service indicator data, which is an important service evaluation indicator, does not experience any delay. Since there is no delay, hardware resources are avoided from being wasted, and the original hardware resources can be used to perform stream computing to obtain the real-time service indicator data.

[0067] Optionally, if the service indicator delay backlog parameter is less than a preset delay backlog threshold, and the service indicator calculation priority is less than a preset priority, the original stream computing hardware resources are used to perform stream computing on the newly added service indicator calculation data in the service indicator calculation data stream to obtain real-time service indicator data. It is understandable that in this scenario, the calculation process of real-time service indicator data for non-critical service evaluation indicators does not experience any delay. Since there is no delay, hardware resources are avoided from being wasted, and the original hardware resources can be used to perform stream computing to obtain real-time service indicator data.

[0068] S310: Obtain offline service indicator calculation data of the target service evaluation indicator within a first preset time interval from the offline data warehouse, and perform offline indicator calculation processing on the offline service indicator calculation data to obtain benchmark service indicator data.

[0069] S312: Obtain online service indicator calculation data of the target service evaluation indicator within a second preset time interval from the online data warehouse, and perform online indicator calculation processing on the online service indicator calculation data to obtain online service indicator data.

[0070] The first preset time interval is greater than the second preset time interval.

[0071] In step S310, the first preset time interval can be a time interval of one day, three days, five days, seven days, fourteen days, etc. For example, if the first preset time interval is one day, the benchmark service indicator data of the target service evaluation indicator on that day can be calculated every day, and multiple benchmark service indicator data can be calculated within a week. For a certain target service evaluation indicator, multiple indicator values ​​(i.e., multiple benchmark service indicator data) of the service evaluation indicator within a week can be obtained. These multiple indicator values ​​(i.e., multiple benchmark service indicator data) can reflect the change trend of the service evaluation indicator within a week. In the process of calculating the benchmark service indicator data, the offline service indicator calculation data can be cleaned first, and then the offline service indicator calculation data after data cleaning can be processed by offline indicator calculation to obtain the benchmark service indicator data. Specifically, data cleaning mainly includes data operations such as removing duplicate values, filling missing values, and detecting outliers. Data quality is improved through data cleaning. Optionally, the offline service indicator calculation data obtained from the offline data warehouse can also be data that has undergone data cleaning processing before being stored in the offline data warehouse. In this way, after obtaining it from the offline data warehouse, there is no need to perform data cleaning processing again, and offline indicator calculation processing can be directly performed to obtain the benchmark service indicator data.

[0072] In step S312, the second preset time interval can be one hour, two hours, three hours, five hours, etc. For example, if the second preset time interval is one hour, starting from midnight, online service indicator data of the target service evaluation indicator can be calculated every hour. Multiple online service indicator data can be calculated within a day. For a certain service evaluation indicator, multiple indicator values ​​(i.e., multiple online service indicator data) of the service indicator within a day can be obtained. These multiple indicator values ​​(i.e., multiple online service indicator data) can reflect the changing trend of the certain service evaluation indicator within a day.

[0073] In the process of calculating the online service indicator data, the online service indicator calculation data can be cleansed first, and then the cleaned online service indicator calculation data can be processed online to obtain the online service indicator data. Optionally, the offline service indicator calculation data obtained from the online data warehouse can also be cleansed before being stored in the online data warehouse. In this way, after obtaining the offline service indicator data from the online data warehouse, no further data cleaning is required, and the online indicator calculation can be directly performed to obtain the online service indicator data.

[0074] It is understandable that in Figure 3In the illustrated embodiment, step S310 is executed before step S312. In other embodiments not shown in this specification, step S312 may be executed first and then step S310, or step S310 and step S312 may be executed simultaneously.

[0075] S314, based on the real-time service indicator data, the benchmark service indicator data and the online service indicator data, the indicator detection prompt words are generated, the indicator detection prompt words are input into the indicator detection big model, and the indicator detection big model is used to perform indicator prediction processing to obtain the predicted service indicator data in the target future time period.

[0076] In some embodiments, the step of generating an indicator detection prompt word based on the real-time service indicator data, the benchmark service indicator data, and the online service indicator data may specifically include: extracting service indicator features based on the real-time service indicator data, the benchmark service indicator data, and the online service indicator data; selecting a prompt word template that matches the service indicator features; and performing prompt word combination processing on the real-time service indicator data, the benchmark service indicator data, the online service indicator data, and the prompt word template to obtain an indicator detection prompt word. The indicator detection prompt word includes task description information indicating that the indicator detection large model predicts service indicator data of a target service evaluation indicator within a target future time period based on the real-time service indicator data, the benchmark service indicator data, and the online service indicator data. For example, taking the target service scenario as the financial service processing scenario and the target service evaluation indicator as transaction volume, the real-time service indicator data is the real-time transaction volume, the benchmark service indicator data is the daily transaction volume, and the online service indicator data is the online transaction volume within each hour of the day. The generated indicator detection prompt can be: "You are a technical expert involved in the financial service processing scenario. Please predict the transaction volume in the next 24 hours based on the daily transaction volume of the past week, the online transaction volume within each hour of the day, and the real-time transaction volume. The transaction volume in the next 24 hours will be output with a one-hour time window."

[0077] The steps of executing indicator prediction processing through the indicator detection big model to obtain predicted service indicator data in the target future time period may specifically include: B2: extracting indicator time fluctuation characteristics and indicator scenario fluctuation characteristics through the indicator detection big model; B4: generating short-term fluctuation information, medium-term change information and long-term trend information based on the indicator time fluctuation characteristics and indicator scenario fluctuation characteristics, and performing service indicator prediction processing in the target future time period based on the short-term fluctuation information, medium-term change information and long-term trend information to obtain predicted service indicator data.

[0078] In step B2, the indicator time fluctuation characteristics are used to reflect the changes in service indicators over different time periods. Indicator time fluctuation characteristics may include hourly fluctuation characteristics, daily fluctuation characteristics, weekly fluctuation characteristics, monthly fluctuation characteristics, and so on. The indicator scenario fluctuation characteristics are used to reflect the changes in service indicators under different specific scenarios within the same target service scenario. For example, indicator scenario fluctuation characteristics may include marketing campaign characteristics, regional characteristics, seasonal characteristics, and so on. The indicator detection large model performs time fluctuation feature extraction processing on the indicator detection prompt word to obtain the indicator time fluctuation feature, and performs scenario fluctuation feature extraction processing to obtain the indicator scenario fluctuation feature.

[0079] In step B4, short-term fluctuation information refers to the immediate fluctuations of service indicators within an hour, day, or several days. This can be obtained by dividing the data into smaller time windows (e.g., hourly or daily) and analyzing and observing them through sliding windows to obtain current and past short-term fluctuation information. Changes in the amplitude of fluctuations over a period of time can also be analyzed, for example, by using statistics such as standard deviation, kurtosis, and skewness to quantify short-term fluctuations and obtain short-term fluctuation information. It is understood that short-term fluctuation information can help capture rapid changes in service indicators over the next few hours or days. For example, if the service evaluation indicator is trading volume, the trading volume of a particular day may increase sharply during peak hours, and short-term fluctuation information can provide a forecast of this change.

[0080] Medium-term variation information refers to fluctuations in service metrics over a period of weeks to months. This information typically reflects trends or seasonal variations in service metrics. This information can be obtained through methods such as trend decomposition, moving averages, and seasonal pattern recognition. Trend decomposition involves using trend decomposition methods (such as STL decomposition) to extract cyclical fluctuations, trend components, and residuals from historical metric data; this allows analysis of medium-term variation patterns in service metrics. Moving averages involve applying a moving average to medium-term metric data to smooth out minor fluctuations and help identify major trends. Regarding seasonal pattern recognition, for example, the Autoregressive Integrated Moving Average (ARIMA) model can be used to identify seasonal patterns and cyclical fluctuations. It can be understood that medium-term variation information helps identify medium-term growth or decline trends in service metrics. For example, taking transaction volume as a service metric, transaction volume in a given quarter may be affected by seasonality or promotional activities.

[0081] Long-term trend information reflects the overall changing trends of service evaluation indicators over a longer timeframe, typically spanning months or even years. At this stage, attention is focused on the impact of long-term growth trends, market saturation, and macroeconomic factors. Specifically, methods such as trend modeling, external factor modeling, and long-term stability can be used to obtain long-term trend information for service evaluation indicators. Trend modeling involves using methods such as linear regression and exponential smoothing to model long-term trends. These methods can capture long-term upward or downward trends in the data. Regarding external factor modeling, for example, causal inference models can be used to analyze the impact of external factors (such as economic cycles and policy changes) on long-term trends. Regarding long-term stability, historical indicator data can be analyzed to determine whether service evaluation indicators exhibit a stable growth pattern over a specific period. It is understood that long-term trend information helps identify the overall direction of service evaluation indicators. For example, if the target service scenario is a financial services scenario and the service evaluation indicator is user number, the number of users of a particular financial service within the financial services scenario may be declining, or after years of growth, the financial service may be approaching market saturation.

[0082] Furthermore, after obtaining short-term fluctuation information, medium-term change information, and long-term trend information, the indicator detection model combines this information for comprehensive prediction, and finally predicts the forecast service indicator data for the future time period.

[0083] Optionally, the training method of the indicator detection big model is: using the basic big language model to create an initial indicator detection big model for the service indicator detection scenario; obtaining sample data, the sample data including real-time sample service indicator data, online sample service indicator data, and benchmark sample service indicator data corresponding to the sample service evaluation indicator in at least one target service scenario, and labeling the sample data with a sample service indicator label, the sample service indicator label is the actual service indicator data generated by the sample service evaluation indicator in the future time period corresponding to the real-time sample service indicator data, the online sample service indicator data, and the benchmark sample service indicator; using the sample data to perform at least one round of model training on the initial indicator detection big model, during the model training process, using the initial indicator detection big model to perform indicator prediction processing on the sample data to obtain predicted sample service indicator data for the sample in the future time period, adjusting the model parameters of the initial indicator detection big model based on the predicted sample service indicator data and the sample service indicator label to obtain the indicator detection big model after model training.

[0084] Specifically, the execution of using the basic big language model to create an initial indicator detection big model for the service indicator detection scenario can be: obtaining the basic big language model, creating an indicator detection scenario adaptation module for the service indicator detection scenario and a big language generation module based on the basic big language model, and forming the initial indicator detection big model based on the big language generation module and the indicator detection scenario adaptation module.

[0085] Specifically, the model training process of the initial indicator detection large model can be as follows: sample data is input into the initial indicator detection large model for at least one round of model training to obtain predicted service indicators; the comprehensive model loss is calculated based on the predicted service indicators and the sample service indicator labels; the model parameters of the indicator detection scenario adaptation module in the initial indicator detection large model are adjusted based on the comprehensive model loss, and the model parameters of the large language generation module are controlled to remain unchanged. This process continues until the model training end conditions are met, the large language generation module and the indicator detection scenario adaptation module are obtained, and the model fusion of the large language generation module and the indicator detection scenario adaptation module is completed, resulting in the trained indicator detection large model.

[0086] Specifically, the model fusion of the large language generation module and the indicator detection scenario adaptation module can be: the model structure layer weight of the indicator detection scenario adaptation module and the large language generation module are weighted and fused, and the model structure layer parameters of the target model structure layer are parameter-fused with the model structure layer weight by determining the target model structure layer corresponding to the model structure layer weight in the large language generation module. The model structure layer weight of the indicator detection scenario adaptation module may only partially correspond to and have model structure layer weights in all model structure layers in the basic large language model. By completing the parameter update of the model structure layer based on the model structure layer weight for this part of the target model structure layer, and so on, the reference update process of all model structure layer weights is completed, thereby obtaining the indicator detection large model.

[0087] Optionally, the model training termination conditions may include, for example, the loss function value being less than or equal to a preset loss function threshold, the number of iterations reaching a preset number threshold, etc. Specific model training termination conditions may be determined based on actual conditions and are not specifically limited here.

[0088] S316, performing indicator merging processing on the data real-time service indicator, benchmark service indicator data, online service indicator data and predicted service indicator data to obtain merged indicator data.

[0089] In some embodiments, the service indicator data are merged in chronological order according to the timestamps of the service indicator data to obtain merged indicator data. For example, the benchmark service indicator data is yesterday's indicator data, the online service indicator data is the indicator data for each hour within the past 12 hours of today, the real-time service indicator data is the indicator data for the past hour, and the predicted service indicator data is the predicted indicator data for the next 12 hours of today. After sorting by timestamp, the order of these indicators is: benchmark service indicator data, online service indicator data, real-time service indicator data, and predicted service indicator data. After merging according to this order, the merged indicator data is obtained.

[0090] S318: Obtain a preset service indicator display format, and use the preset service indicator display format to perform indicator analysis on the combined indicator data to obtain service indicator analysis data.

[0091] The preset service indicator display form may include but is not limited to other display forms such as charts, text, etc. Specifically, the chart may include other types of charts such as trend charts, bar charts, and pie charts.

[0092] In some embodiments, the combined indicator data may be subjected to an indicator trend analysis, an indicator growth rate analysis, or other analytical processing to obtain an indicator analysis result. The combined indicator data may be converted into an indicator display result using a preset service indicator display format. The indicator display result and the indicator analysis result may be combined to obtain service indicator analysis data. It is understood that the service indicator analysis data may include the changing trends of the service indicators within historical time periods and future time periods, and the service indicator analysis data may provide support for service decision-making.

[0093] S320: associate the service indicator parsing data with the target service evaluation indicator and store them, and output and display the service indicator parsing data.

[0094] The implementation of step S320 can be found in Figure 2 The relevant description of S208 in the illustrated embodiment will not be repeated here in detail.

[0095] The service processing method provided in the embodiments of this specification monitors the service indicator calculation data flow corresponding to the target service evaluation indicator in at least one target service scenario, monitors the service indicator delay backlog parameter, determines the service indicator calculation priority, detects whether the service indicator delay backlog parameter is greater than or equal to the preset delay backlog threshold, and detects whether the service indicator calculation priority is greater than or equal to the preset priority. If the service indicator delay backlog parameter is greater than or equal to the preset delay backlog threshold, and the service indicator calculation priority is greater than or equal to the preset priority, then data clipping processing is performed on the newly added service indicator calculation data in the service indicator calculation data flow to obtain target service indicator calculation data, and stream calculation processing is performed on the target service indicator calculation data to obtain real-time service indicator. In this way, in the process of obtaining real-time service indicator data through stream computing, the indicator delay situation is determined and a corresponding strategy is adopted according to the indicator delay situation to solve the indicator delay problem; then, offline service indicator calculation data of the target service evaluation indicator within a first preset time interval is obtained from the offline data warehouse, and offline indicator calculation processing is performed on the offline service indicator calculation data to obtain benchmark service indicator data, and online service indicator calculation data of the target service evaluation indicator within a second preset time interval is obtained from the online data warehouse, and online service indicator calculation data is performed on the online service indicator calculation data to obtain online service indicator data, and indicator detection prompt words are generated based on the real-time service indicator data, the benchmark service indicator data and the online service indicator data, and the indicators are converted into The detection prompt word is input into the indicator detection big model, and the indicator prediction processing is performed through the indicator detection big model to obtain the predicted service indicator data in the target future time period, and the real-time service indicator data, benchmark service indicator data, online service indicator data and predicted service indicator data are merged to obtain the merged indicator data, and the preset service indicator display form is obtained. The preset service indicator display form is used to perform indicator analysis processing on the merged indicator data to obtain service indicator parsing data. In this way, the indicator data of different time dimensions are calculated through a hybrid computing method composed of stream computing, offline computing and online computing, which improves the indicator calculation efficiency and indicator calculation accuracy in big data processing. The indicator data in the future time period can be predicted through the big model, that is, The indicator trends in the future time period are predicted, thereby realizing the function of providing accurate actual indicator data and predicting future indicator data. Finally, service indicator analysis data is generated based on real-time service indicator data, benchmark service indicator data, online service indicator data and predicted service indicator data. The actual indicator data and future indicator trends are summarized and displayed through the preset service indicator display format, the service indicator analysis data are associated with the target service indicator and stored, and the service indicator analysis data is output and displayed. Therefore, by summarizing and displaying accurate and highly real-time actual indicator data and future indicator trends, efficient display of indicator data is guaranteed, which helps relevant personnel understand and use indicator data and improves the operability and practicality of indicator data.

[0096] The following will be combined Figure 4 , the service processing device provided in the embodiment of this specification is introduced in detail. It should be noted that, Figure 4 The service processing device shown is used to execute this instruction Figures 1 to 3 For the convenience of explanation, only the part related to the embodiment of this specification is shown. For the specific technical details not disclosed, please refer to this specification. Figures 1 to 3 The embodiment shown.

[0097] See Figure 4 , which shows a schematic diagram of the structure of the service processing device of an embodiment of this specification. The service processing device 1 can be implemented as all or part of the device through software, hardware, or a combination of both. According to some embodiments, the service processing device 1 includes a first indicator calculation module 11, a second indicator calculation module 12, a third indicator calculation module 13, and an indicator analysis and processing module 14, which are specifically used to: The first indicator calculation module 11 is used to monitor the service indicator calculation data stream corresponding to the target service evaluation indicator in at least one target service scenario, and perform stream computing on the newly added service indicator calculation data in the service indicator calculation data stream to obtain real-time service indicator data; A second indicator calculation module 12 is configured to calculate benchmark service indicator data based on offline service indicator calculation data corresponding to the target service evaluation indicator in the offline data warehouse, and to calculate online service indicator data based on online service indicator calculation data corresponding to the target service evaluation indicator in the online data warehouse; The third indicator calculation module 13 is used to call the indicator detection model to perform indicator prediction processing based on the real-time service indicator data, the benchmark service indicator data and the online service indicator data to obtain predicted service indicator data within a target future time period; The indicator analysis and processing module 14 is used to generate service indicator parsing data based on the real-time service indicator data, the benchmark service indicator data, the online service indicator data and the predicted service indicator data, associate the service indicator parsing data with the target service indicator and store it, and output and display the service indicator parsing data.

[0098] Optionally, the first indicator calculation module 11 includes: A data monitoring unit is used to monitor the service indicator delay backlog parameters and determine the service indicator calculation priority; a data detection unit, configured to detect whether the service indicator delay backlog parameter is greater than or equal to a preset delay backlog threshold, and to detect whether the service indicator calculation priority is greater than or equal to a preset priority; A data calculation unit is used to perform data clipping processing on the newly added service indicator calculation data in the service indicator calculation data stream to obtain target service indicator calculation data if the service indicator delay backlog parameter is greater than or equal to the preset delay backlog threshold and the service indicator calculation priority is greater than or equal to the preset priority, and to perform stream calculation processing on the target service indicator calculation data to obtain real-time service indicator data.

[0099] Optional data computing unit, specifically used for: A stream computing data processing range is determined in the newly added service indicator calculation data in the service indicator calculation data stream, and data clipping processing is performed on the newly added service indicator calculation data based on the stream computing data processing range to obtain target service indicator calculation data.

[0100] Optional data computing unit, specifically used for: If the target service indicator calculation data belongs to all the data in the newly added service indicator calculation data, then performing resource expansion processing on the stream computing hardware resources, and using the expanded stream computing hardware resources to perform stream computing processing on the target service indicator calculation data to obtain real-time service indicator data; If the target service indicator calculation data is part of the newly added service indicator calculation data, the original stream computing hardware resources are used to perform stream computing processing on the target service indicator calculation data to obtain real-time service indicator data.

[0101] Optionally, the third indicator calculation module 13 includes: An indicator prediction unit is used to generate indicator detection prompt words based on the real-time service indicator data, the benchmark service indicator data and the online service indicator data, input the indicator detection prompt words into the indicator detection large model, and perform indicator prediction processing through the indicator detection large model to obtain predicted service indicator data in the target future time period.

[0102] Optional indicator prediction unit, specifically used for: Extracting the indicator time fluctuation characteristics and the indicator scene fluctuation characteristics through the indicator detection large model; Based on the indicator time fluctuation characteristics and the indicator scenario fluctuation characteristics, short-term fluctuation information, medium-term change information and long-term trend information are generated. Based on the short-term fluctuation information, the medium-term change information and the long-term trend information, the service indicator prediction processing in the target future time period is performed to obtain predicted service indicator data.

[0103] Optionally, the indicator analysis and processing module 14 is specifically configured to: Performing indicator merging processing on the real-time service indicator data, the benchmark service indicator data, the online service indicator data, and the predicted service indicator data to obtain merged indicator data; A preset service indicator display format is obtained, and the preset service indicator display format is used to perform indicator analysis processing on the combined indicator data to obtain service indicator parsing data.

[0104] Optionally, the second indicator calculation module 12 is specifically configured to: Obtaining offline service indicator calculation data of the target service evaluation indicator within a first preset time interval from an offline data warehouse, and performing offline indicator calculation processing on the offline service indicator calculation data to obtain benchmark service indicator data; Obtaining online service indicator calculation data of the target service evaluation indicator within a second preset time interval from an online data warehouse, and performing online indicator calculation processing on the online service indicator calculation data to obtain online service indicator data; Wherein, the first preset time interval is greater than the second preset time interval.

[0105] Please refer to Figure 5 , which shows a schematic diagram of the structure of an electronic device provided by an exemplary embodiment of this specification. The electronic device described in this specification may include one or more of the following components: a processor 110, a memory 120, an input device 130, an output device 140, and a bus 150. The processor 110, the memory 120, the input device 130, and the output device 140 may be connected via the bus 150.

[0106] Processor 110 may include one or more processing cores. Using various interfaces and circuits, processor 110 connects various components within the terminal. It executes instructions, programs, code sets, or instruction sets stored in memory 120, as well as accesses data stored in memory 120, to perform various functions and process data for terminal 100. Optionally, processor 110 may be implemented in hardware using at least one of a digital signal processing (DSP), a field-programmable gate array (FPGA), and a programmable logic array (PLA). Processor 110 may integrate one or a combination of a central processing unit (CPU), a graphics processing unit (GPU), and a modem. The CPU primarily handles the operating system, user interface, and application programs; the GPU is responsible for rendering and drawing display content; and the modem handles wireless communications. It is understood that the modem may also be implemented independently of the processor 110 via a separate communications chip.

[0107] The memory 120 may include a random access memory (RAM) or a read-only memory (ROM). Optionally, the memory 120 includes a non-transitory computer-readable storage medium. The memory 120 may be used to store instructions, programs, codes, code sets, or instruction sets. The memory 120 may include a program storage area and a data storage area, wherein the program storage area may store instructions for implementing an operating system, instructions for implementing at least one function (e.g., a touch function, a sound playback function, an image playback function, etc.), instructions for implementing the following various method embodiments, etc. The operating system may be an Android system, including systems deeply developed based on the Android system, an iOS system developed by Apple, including systems deeply developed based on the iOS system, or other systems.

[0108] In order for the operating system to distinguish the specific application scenarios of third-party applications, it is necessary to open up data communication between third-party applications and the operating system so that the operating system can obtain the current scenario information of third-party applications at any time, and then perform targeted system resource adaptation based on the current scenario.

[0109] The input device 130 is used to receive input commands or data and includes, but is not limited to, a keyboard, a mouse, a camera, a microphone, or a touch-sensitive device. The output device 140 is used to output commands or data and includes, but is not limited to, a display device and a speaker. In one example, the input device 130 and the output device 140 may be combined, and the input device 130 and the output device 140 may be a touch-sensitive display.

[0110] The touch display screen can be designed as a full screen, a curved screen or a special-shaped screen. The touch display screen can also be designed as a combination of a full screen and a curved screen, or a combination of a special-shaped screen and a curved screen, which is not limited in the embodiments of this specification.

[0111] In addition, those skilled in the art will understand that the structures of the electronic devices shown in the above figures do not limit the electronic devices. The electronic devices may include more or fewer components than shown, or may combine certain components or arrange the components differently. For example, the electronic devices may also include radio frequency circuits, input units, sensors, audio circuits, wireless fidelity (WiFi) modules, power supplies, Bluetooth modules, and other components, which will not be described in detail here.

[0112] In some embodiments, Figure 5 In the electronic device shown, the processor 110 may be configured to call a program of a service processing method stored in the memory 120 and specifically perform the following operations: Monitor a service indicator calculation data stream corresponding to a target service evaluation indicator in at least one target service scenario, and perform stream computing on newly added service indicator calculation data in the service indicator calculation data stream to obtain real-time service indicator data; Calculate the benchmark service indicator data based on the offline service indicator calculation data of the target service evaluation indicator in the offline data warehouse, and calculate the online service indicator data based on the online service indicator calculation data of the target service evaluation indicator in the online data warehouse; Based on the real-time service indicator data, the benchmark service indicator data and the online service indicator data, calling the indicator detection model to perform indicator prediction processing to obtain predicted service indicator data within a target future time period; Service indicator parsing data is generated based on the real-time service indicator data, the benchmark service indicator data, the online service indicator data and the predicted service indicator data, the service indicator parsing data is associated with the target service indicator and stored, and the service indicator parsing data is output and displayed.

[0113] Optionally, when executing the step of performing stream computing on the newly added service indicator calculation data in the service indicator calculation data stream to obtain real-time service indicator data, the processor 110 specifically performs the following operations: Monitor service indicator delay backlog parameters and determine service indicator calculation priorities; Detecting whether the service indicator delay backlog parameter is greater than or equal to a preset delay backlog threshold, and detecting whether the service indicator calculation priority is greater than or equal to a preset priority; If the service indicator delay backlog parameter is greater than or equal to the preset delay backlog threshold, and the service indicator calculation priority is greater than or equal to the preset priority, the newly added service indicator calculation data in the service indicator calculation data stream is subjected to data trimming processing to obtain target service indicator calculation data, and the target service indicator calculation data is subjected to stream computing processing to obtain real-time service indicator data.

[0114] Optionally, when executing the step of performing data trimming processing on the newly added service indicator calculation data in the service indicator calculation data stream to obtain target service indicator calculation data, the processor 110 specifically performs the following operations: A stream computing data processing range is determined in the newly added service indicator calculation data in the service indicator calculation data stream, and data clipping processing is performed on the newly added service indicator calculation data based on the stream computing data processing range to obtain target service indicator calculation data.

[0115] Optionally, when executing the step of performing stream computing on the target service indicator calculation data to obtain the real-time service indicator, the processor 110 specifically performs the following operations: If the target service indicator calculation data belongs to all the data in the newly added service indicator calculation data, then performing resource expansion processing on the stream computing hardware resources, and using the expanded stream computing hardware resources to perform stream computing processing on the target service indicator calculation data to obtain real-time service indicator data; If the target service indicator calculation data is part of the newly added service indicator calculation data, the original stream computing hardware resources are used to perform stream computing processing on the target service indicator calculation data to obtain real-time service indicator data.

[0116] Optionally, when executing the step of calling the indicator detection large model to perform indicator prediction processing based on the real-time service indicator data, the benchmark service indicator data, and the online service indicator data to obtain predicted service indicator data for a target future time period, the processor 110 specifically performs the following operations: Based on the real-time service indicator data, the benchmark service indicator data and the online service indicator data, an indicator detection prompt word is generated, the indicator detection prompt word is input into the indicator detection big model, and the indicator prediction processing is performed by the indicator detection big model to obtain the predicted service indicator data in the target future time period.

[0117] Optionally, when executing the step of performing indicator prediction processing using the indicator detection large model to obtain predicted service indicator data within a target future time period, the processor 110 specifically performs the following operations: Extracting the indicator time fluctuation characteristics and the indicator scene fluctuation characteristics through the indicator detection large model; Based on the indicator time fluctuation characteristics and the indicator scenario fluctuation characteristics, short-term fluctuation information, medium-term change information and long-term trend information are generated. Based on the short-term fluctuation information, the medium-term change information and the long-term trend information, the service indicator in the target future time period is predicted and processed to obtain the target service indicator.

[0118] Optionally, when executing the step of generating the service indicator parsing data based on the real-time service indicator data, the benchmark service indicator data, the online service indicator data, and the predicted service indicator data, the processor 110 specifically performs the following operations: Performing indicator merging processing on the real-time service indicator data, the benchmark service indicator data, the online service indicator data, and the predicted service indicator data to obtain merged indicator data; A preset service indicator display format is obtained, and the preset service indicator display format is used to perform indicator analysis processing on the combined indicator data to obtain service indicator parsing data.

[0119] Optionally, when executing the steps of calculating the benchmark service indicator based on the offline service indicator calculation data corresponding to the target service evaluation indicator in the offline data warehouse, and calculating the online service indicator based on the online service indicator calculation data corresponding to the target service evaluation indicator in the online data warehouse, the processor 110 specifically performs the following operations: Obtaining offline service indicator calculation data of the target service evaluation indicator within a first preset time interval from an offline data warehouse, and performing offline indicator calculation processing on the offline service indicator calculation data to obtain benchmark service indicator data; Obtaining online service indicator calculation data of the target service evaluation indicator within a second preset time interval from an online data warehouse, and performing online indicator calculation processing on the online service indicator calculation data to obtain online service indicator data; Wherein, the first preset time interval is greater than the second preset time interval.

[0120] It should be noted that the information (including but not limited to user device information, user personal information, etc.), data (including but not limited to data used for analysis, stored data, displayed data, etc.) and signals involved in the embodiments of this specification are all authorized by the user or fully authorized by all parties, and the collection, use and processing of relevant data must comply with the relevant laws, regulations and standards of the relevant countries and regions. For example, the service indicator calculation data flow, new service indicator calculation data, offline service indicator calculation data, and online service indicator calculation data involved in this specification.

[0121] An embodiment of this specification further provides a computer-readable storage medium, which stores at least one instruction, and the at least one instruction is used to be executed by a processor to implement the service processing method described in the above embodiments.

[0122] The embodiments of this specification also provide a computer program product, which stores at least one instruction, and the at least one instruction is loaded and executed by the processor to implement the service processing method described in the above embodiments.

[0123] Those skilled in the art will appreciate that in one or more of the above examples, the functions described in the embodiments of this specification can be implemented using hardware, software, firmware, or any combination thereof. When implemented using software, these functions can be stored in a computer-readable medium or transmitted as one or more instructions or codes on a computer-readable medium. Computer-readable media include computer storage media and communication media, wherein communication media include any medium that facilitates the transmission of computer programs from one place to another. The storage medium can be any available medium that can be accessed by a general-purpose or special-purpose computer.

[0124] The above description is only an optional embodiment of this specification and is not intended to limit this specification. Any modifications, equivalent replacements, improvements, etc. made within the spirit and principles of this specification should be included in the scope of protection of this specification.

[0125] The foregoing description of this specification describes specific embodiments. Other embodiments are within the scope of the appended claims. In some cases, the actions or steps recited in the claims can be performed in an order different from that described in the embodiments and still achieve the desired results. Furthermore, the processes depicted in the accompanying drawings do not necessarily require the specific order shown or the sequential order to achieve the desired results. In certain embodiments, multitasking and parallel processing are also possible or may be advantageous.

Claims

1. A service processing method, the method comprising: Monitor a service indicator calculation data stream corresponding to a target service evaluation indicator in at least one target service scenario, and perform stream computing on newly added service indicator calculation data in the service indicator calculation data stream to obtain real-time service indicator data; Calculate the benchmark service indicator data based on the offline service indicator calculation data corresponding to the target service evaluation indicator in the offline data warehouse, and calculate the online service indicator data based on the online service indicator calculation data corresponding to the target service evaluation indicator in the online data warehouse; Based on the real-time service indicator data, the benchmark service indicator data and the online service indicator data, calling the indicator detection model to perform indicator prediction processing to obtain predicted service indicator data within a target future time period; Service indicator parsing data is generated based on the real-time service indicator data, the benchmark service indicator data, the online service indicator data and the predicted service indicator data, the service indicator parsing data is associated with the target service evaluation indicator and stored, and the service indicator parsing data is output and displayed.

2. The method according to claim 1, wherein performing stream computing on the newly added service indicator calculation data in the service indicator calculation data stream to obtain real-time service indicator data comprises: Monitor service indicator delay backlog parameters and determine service indicator calculation priorities; Detecting whether the service indicator delay backlog parameter is greater than or equal to a preset delay backlog threshold, and detecting whether the service indicator calculation priority is greater than or equal to a preset priority; If the service indicator delay backlog parameter is greater than or equal to the preset delay backlog threshold, and the service indicator calculation priority is greater than or equal to the preset priority, the newly added service indicator calculation data in the service indicator calculation data stream is subjected to data trimming processing to obtain target service indicator calculation data, and the target service indicator calculation data is subjected to stream computing processing to obtain real-time service indicator data.

3. The method according to claim 2, wherein the step of performing data trimming processing on the newly added service indicator calculation data in the service indicator calculation data stream to obtain target service indicator calculation data comprises: A stream computing data processing range is determined in the newly added service indicator calculation data in the service indicator calculation data stream, and data clipping processing is performed on the newly added service indicator calculation data based on the stream computing data processing range to obtain target service indicator calculation data.

4. The method according to claim 2 or 3, wherein the step of performing stream computing on the target service indicator calculation data to obtain real-time service indicator data comprises: If the target service indicator calculation data belongs to all the data in the newly added service indicator calculation data, then performing resource expansion processing on the stream computing hardware resources, and using the expanded stream computing hardware resources to perform stream computing processing on the target service indicator calculation data to obtain real-time service indicator data; If the target service indicator calculation data is part of the newly added service indicator calculation data, the original stream computing hardware resources are used to perform stream computing processing on the target service indicator calculation data to obtain real-time service indicator data.

5. The method according to claim 1, wherein the step of calling a large indicator detection model to perform indicator prediction processing based on the real-time service indicator data, the benchmark service indicator data, and the online service indicator data to obtain predicted service indicator data for a target future time period comprises: Based on the real-time service indicator data, the benchmark service indicator data and the online service indicator data, an indicator detection prompt word is generated, the indicator detection prompt word is input into the indicator detection big model, and the indicator prediction processing is performed by the indicator detection big model to obtain the predicted service indicator data in the target future time period.

6. The method according to claim 5, wherein the step of performing indicator prediction processing using the indicator detection large model to obtain predicted service indicator data for a target future time period comprises: Extracting the indicator time fluctuation characteristics and the indicator scene fluctuation characteristics through the indicator detection large model; Based on the indicator time fluctuation characteristics and the indicator scenario fluctuation characteristics, short-term fluctuation information, medium-term change information and long-term trend information are generated. Based on the short-term fluctuation information, the medium-term change information and the long-term trend information, the service indicator prediction processing in the target future time period is performed to obtain predicted service indicator data.

7. The method according to claim 1, wherein generating service indicator parsing data based on the real-time service indicator data, the benchmark service indicator data, the online service indicator data, and the predicted service indicator data comprises: Performing indicator merging processing on the real-time service indicator data, the benchmark service indicator data, the online service indicator data, and the predicted service indicator data to obtain merged indicator data; A preset service indicator display format is obtained, and the preset service indicator display format is used to perform indicator analysis processing on the combined indicator data to obtain service indicator parsing data.

8. The method according to claim 1, wherein the step of calculating the benchmark service indicator data based on the offline service indicator calculation data corresponding to the target service evaluation indicator in the offline data warehouse, and calculating the online service indicator data based on the online service indicator calculation data corresponding to the target service evaluation indicator in the online data warehouse, comprises: Obtaining offline service indicator calculation data of the target service evaluation indicator within a first preset time interval from an offline data warehouse, and performing offline indicator calculation processing on the offline service indicator calculation data to obtain benchmark service indicator data; Obtaining online service indicator calculation data of the target service evaluation indicator within a second preset time interval from an online data warehouse, and performing online indicator calculation processing on the online service indicator calculation data to obtain online service indicator data; Wherein, the first preset time interval is greater than the second preset time interval.

9. A service processing device, comprising: A first indicator calculation module is used to monitor a service indicator calculation data stream corresponding to a target service evaluation indicator in at least one target service scenario, and perform stream computing on newly added service indicator calculation data in the service indicator calculation data stream to obtain real-time service indicator data; A second indicator calculation module is configured to calculate the benchmark service indicator data based on the offline service indicator calculation data corresponding to the target service evaluation indicator in the offline data warehouse, and calculate the online service indicator data based on the online service indicator calculation data corresponding to the target service evaluation indicator in the online data warehouse; A third indicator calculation module is used to call the indicator detection model to perform indicator prediction processing based on the real-time service indicator data, the benchmark service indicator data and the online service indicator data to obtain predicted service indicator data within a target future time period; The indicator analysis and processing module is used to generate service indicator analysis data based on the real-time service indicator data, the benchmark service indicator data, the online service indicator data and the predicted service indicator data, associate the service indicator analysis data with the target service evaluation indicator and store it, and output and display the service indicator analysis data.

10. A computer storage medium storing a plurality of instructions, wherein the instructions are suitable for being loaded by a processor and executing the method according to any one of claims 1 to 8.

11. A computer program product, wherein the computer program product stores at least one instruction, wherein the at least one instruction is loaded by a processor and executes the method according to any one of claims 1 to 8.

12. An electronic device comprising: A processor and a memory; wherein the memory stores a computer program, and the computer program is suitable for being loaded by the processor and executing the method according to any one of claims 1 to 8.

Images