Centralized service agent construction method and device, equipment, medium and program product

Abstract

The embodiment of the invention discloses a centralized service agent construction method and device, equipment, a medium and a program product. The method comprises the following steps: acquiring a plurality of historical appeals and appeal content corresponding to each historical appeal; determining a risk degree corresponding to each historical appeal according to the appeal content; determining a plurality of target appeals according to the risk degrees, and clustering the plurality of target appeals to obtain a target appeal set belonging to the same category; determining a plurality of target items according to the target appeal set; performing permutation and combination on the plurality of target items to obtain a plurality of item sequences to be selected, and verifying each item sequence to be selected to obtain a successfully verified target item sequence; according to the method, the centralized service agent corresponding to the target item sequence is constructed, and centralized service is carried out on all target items in the target item sequence based on the centralized service agent, so that a new'one item 'scene is automatically mined, the corresponding centralized service agent is constructed, and the coverage degree of centralized service is improved.

Description

TECHNICAL FIELD

[0001] Embodiments of the present application relate to computer technology, and particularly relate to a centralized service intelligent agent construction method, device, equipment, medium and program product. BACKGROUND

[0002] "High efficiency to do one thing" is an important way to optimize government service and improve administrative efficiency. "One thing" refers to the centralized service of "one thing once" and "one type of thing one-stop" for enterprises and the public, which is to centrally handle multiple matters that need to be handled by multiple departments or across levels, have strong correlation, large handling volume and relatively concentrated handling time, from the perspective of enterprises and the public.

[0003] Currently, a batch of high-frequency, wide-ranging and problem-ridden "one thing" scenarios are proposed from the perspective of the needs of the public and enterprises, and integrated development of centralized services for such high-frequency general scenarios is carried out to form complete multi-matter joint business processes, such as the existing "one thing" scenarios focusing on high-frequency general scenarios such as newborn birth, enterprise establishment and real estate registration. However, due to the lack of some niche demand "one thing" scenarios, some groups still need to handle in the traditional way. Therefore, there is an urgent need for a way to mine various types of "one thing" scenarios and provide corresponding centralized services. SUMMARY

[0004] Embodiments of the present application provide a centralized service intelligent agent construction method, device, equipment, medium and program product to automatically mine new "one thing" scenarios and construct corresponding centralized service intelligent agents, thereby realizing centralized services for various types of "one thing" scenarios and improving the coverage of centralized services.

[0005] In a first aspect, embodiments of the present application provide a centralized service intelligent agent construction method, comprising:

[0006] obtaining a plurality of historical demands and demand content corresponding to each historical demand;

[0007] determining the risk degree corresponding to each historical demand according to the demand content corresponding to each historical demand;

[0008] determining a plurality of target demands from the plurality of historical demands according to the risk degree corresponding to each historical demand, and clustering the plurality of target demands to obtain a target demand set belonging to the same category;

[0009] determining a plurality of target matters corresponding to the target demand set according to the demand content of each target demand in the target demand set;

[0010] The multiple target items are arranged and combined to obtain multiple candidate item sequences, and each candidate item sequence is verified to obtain a target item sequence that has been successfully verified.

[0011] Construct a centralized service agent corresponding to the target item sequence, so as to provide centralized services for all target items in the target item sequence based on the centralized service agent.

[0012] Secondly, embodiments of the present invention also provide a centralized service agent construction apparatus, comprising:

[0013] The historical request acquisition module is used to acquire multiple historical requests and the content of each historical request.

[0014] The risk level determination module is used to determine the risk level of each historical request based on the content of the request.

[0015] The target demand set determination module is used to determine multiple target demands from the multiple historical demands based on the risk level corresponding to each historical demand, and to cluster the multiple target demands to obtain a set of target demands belonging to the same category.

[0016] The target item determination module is used to determine multiple target items corresponding to the target request set based on the request content of each target request in the target request set;

[0017] The target item sequence determination module is used to arrange and combine the multiple target items to obtain multiple candidate item sequences, and to verify each candidate item sequence to obtain a target item sequence that has been successfully verified.

[0018] A centralized service agent construction module is used to construct a centralized service agent corresponding to the target item sequence, so as to provide centralized services for all target items in the target item sequence based on the centralized service agent.

[0019] Thirdly, embodiments of the present invention also provide an electronic device, the electronic device comprising:

[0020] At least one processor; and

[0021] A memory communicatively connected to the at least one processor; wherein,

[0022] The memory stores a computer program that can be executed by the at least one processor, which enables the at least one processor to execute the centralized service agent construction method provided in any embodiment of the present invention.

[0023] Fourthly, embodiments of the present invention also provide a computer-readable storage medium storing computer instructions, which are used to cause a processor to execute and implement the centralized service agent construction method provided in any embodiment of the present invention.

[0024] Fifthly, embodiments of the present invention also provide a computer program product, including a computer program that, when executed by a processor, implements the centralized service agent construction method provided in any embodiment of the present invention.

[0025] One embodiment of the above invention has the following advantages or beneficial effects:

[0026] By acquiring multiple historical requests and their corresponding content, determining the risk level of each historical request based on its content, identifying high-risk target requests from all historical requests based on their risk levels, and clustering all target requests to obtain a set of target requests belonging to the same category, and determining multiple target items corresponding to each target request set based on the content of each target request in the target request set, and arranging and combining all target items to obtain multiple candidate item sequences, and validating each candidate item sequence to obtain a sequence of successfully validated target items, which constitutes the new "one-thing" scenario, new "one-thing" scenarios with personalized needs are automatically extracted from historical requests. By constructing a centralized service agent corresponding to the target item sequence, centralized services are provided for all target items in the target item sequence, thereby achieving centralized services for various types of "one-thing" scenarios and improving the coverage of centralized services.

[0027] It should be understood that the description in this section is not intended to identify key or essential features of the embodiments of the present invention, nor is it intended to limit the scope of the invention. Other features of the invention will become readily apparent from the following description. Attached Figure Description

[0028] To more clearly illustrate the technical solutions in the embodiments of the present invention or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0029] Figure 1 This is a flowchart of a centralized service agent construction method provided in one embodiment of the present invention;

[0030] Figure 2This is an example diagram illustrating a centralized service agent construction process according to an embodiment of the present invention;

[0031] Figure 3 This is a flowchart of another centralized service agent construction method provided in one embodiment of the present invention;

[0032] Figure 4 This is a schematic diagram of a centralized service agent construction device provided in one embodiment of the present invention;

[0033] Figure 5 This is a schematic diagram of the structure of an electronic device that implements the centralized service agent construction method of the present invention. Detailed Implementation

[0034] To enable those skilled in the art to better understand the present invention, the technical solutions of the present invention will be clearly and completely described below with reference to the accompanying drawings of the embodiments of the present invention. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort should fall within the scope of protection of the present invention.

[0035] It should be noted that the terms "first," "second," etc., in the specification, claims, and accompanying drawings of this invention are used to distinguish similar objects and are not necessarily used to describe a specific order or sequence. It should be understood that such data can be interchanged where appropriate so that the embodiments of the invention described herein can be implemented in orders other than those illustrated or described herein. Furthermore, the terms "comprising" and "having," and any variations thereof, are intended to cover a non-exclusive inclusion; for example, a process, method, system, product, or apparatus that comprises a series of steps or units is not necessarily limited to those steps or units explicitly listed, but may include other steps or units not explicitly listed or inherent to such processes, methods, products, or apparatus.

[0036] Figure 1 This is a flowchart illustrating a centralized service agent construction method according to an embodiment of the present invention. This embodiment is applicable to situations where new "one-thing" scenarios are identified and corresponding centralized service agents are constructed. The method can be executed by a centralized service agent construction device, which can be implemented in software and / or hardware and can be configured in an electronic device. Figure 1 As shown, the method specifically includes the following steps:

[0037] S110. Obtain multiple historical requests and the content of each historical request.

[0038] Among these, "requests" refer to issues raised by the public or employees through government service hotlines and other means. "Historical requests" refer to requests received within a specific historical period. For example, historical requests refer to requests received within the last month. The content of historical requests can include specific information about the issues that need to be resolved.

[0039] In this embodiment, multiple historical requests and the content of each historical request can be obtained from the historical request work order data accumulated by the government service hotline. This allows for the subsequent extraction of personalized "one-stop" scenarios from the historical requests, thereby increasing the richness of "one-stop" scenarios and meeting the need to provide centralized services to various groups.

[0040] S120. Determine the risk level corresponding to each historical demand based on the content of the demand.

[0041] The risk level can be used to characterize the importance of handling historical demands. Each historical demand is assigned a risk score based on its content, and the score is used as the risk level for that historical demand.

[0042] In some optional implementations, step S120 may include: determining the urgency, impact, and difficulty of resolution of each historical request based on the content of the request; and determining the risk level of each historical request based on the urgency, impact, and difficulty of resolution.

[0043] The urgency level can be used to measure the timeliness sensitivity of handling historical demands. The impact level can be used to assess the scope and affected groups of historical demands. The difficulty of resolution can be used to assess the resource consumption and coordination complexity required to handle historical demands. For example, see... Figure 2 This method scores each historical request based on its content, considering three risk dimensions: urgency, impact, and difficulty of resolution. This yields the corresponding urgency, impact, and difficulty of resolution for each historical request. By merging these factors, a merged risk level is obtained, improving the accuracy of risk assessment and further enhancing the accuracy of "one-event" scenario discovery. For example, the urgency, impact, and difficulty of resolution for each historical request can be multiplied, and the result can be used as the corresponding risk level. Alternatively, these factors can be added together, and the sum can be used as the corresponding risk level. This embodiment does not specifically limit the merging method. Figure 2As shown, the risk level (i.e., score) of each historical claim (i.e., claim 1, claim 2, claim 3, etc.) can be a value between 0 and 100.

[0044] S130. Based on the risk level corresponding to each historical demand, determine multiple target demands from multiple historical demands, and cluster the multiple target demands to obtain a set of target demands belonging to the same category.

[0045] Here, a target demand can refer to a high-risk demand that requires constructing a scenario for "one thing". For example, a target demand can be a high-risk demand that is strongly reflected, has a serious problem, and requires immediate resolution. A set of target demands includes at least one target demand whose content belongs to the same category. All target demands in a set can be related in content and can be processed collectively. The number of target demand sets can be one or more.

[0046] Specifically, the risk levels of historical requests can be compared with preset risk levels. Each historical request exceeding the preset risk level is identified as a target request. Alternatively, based on the risk level of each historical request, all historical requests can be sorted in descending order, and the top preset number of such requests are identified as target requests. This yields multiple high-risk target requests, allowing for the identification of more urgent "one-thing" scenarios from among them. Based on the content of each target request, all target requests are clustered, with each cluster representing a target request set. The clustering method can be density-based, allowing for clustering without specifying the number of clusters, thus improving the accuracy of determining the target request set.

[0047] S140. Based on the content of each target demand in the target demand set, determine the multiple target items corresponding to the target demand set.

[0048] Among these, "matters" can be those explicitly stipulated to be handled in policy documents. Policy documents are legally binding formal documents formulated by the government to regulate social behavior and manage public affairs. "Target matters" refer to matters related to the target demands, that is, the matters requested to be handled within the target demands. Different target demands may correspond to different target matters, or they may correspond to the same target matters.

[0049] Specifically, for each set of target demands obtained in step S130, the content of each target demand in the set is matched with the content of all matters in all policy documents. The successfully matched matters are identified as the target matters corresponding to the respective target demands, thereby obtaining all target matters corresponding to that set of target demands. All target matters are then deduplicated to avoid processing duplicate target matters in subsequent steps. See also... Figure 2 By referring to policy documents, the target items associated with each target demand are extracted to obtain multiple standardized target items, namely Item 1, Item 2, ..., Item N.

[0050] In some optional implementations, step S140 may include: for each target demand in the target demand set, inputting the demand content of the target demand and the target policy document corresponding to the target demand into the language processing model to identify the target items related to the target demand from the target policy document; and performing deduplication processing on all target items corresponding to the target demand set to obtain all deduplicated target items.

[0051] The language processing model can be a neural network model capable of understanding and processing natural language. For example, it can be, but is not limited to, any Large Language Model (LLM). The target policy document refers to the policy document to which the target appeal is mapped. Based on the keywords in the content of the target appeal and the corresponding keywords in each policy document, the target appeal can be mapped to a specific policy document to obtain the target policy document corresponding to the target appeal. By inputting only the content of the target appeal and the corresponding target policy document into the language processing model, it is unnecessary to input all policy documents into the language processing model for recognition. Therefore, the language processing model can more quickly and accurately determine the target items corresponding to each target appeal, and deduplicate all target items, using the deduplicated list of all target items as the final determined list of target items.

[0052] S150. Arrange and combine multiple target items to obtain multiple candidate item sequences, and verify each candidate item sequence to obtain a target item sequence that has been successfully verified.

[0053] Here, a candidate item sequence refers to the sequence obtained by arranging at least two target items from all target items. There can be multiple candidate item sequences, the exact number of which is determined based on the number of permutations and combinations. Each candidate item sequence represents a possible business process, or a possible "one-thing" scenario. The target item sequence refers to the successfully validated candidate item sequence, which is the newly discovered "one-thing" scenario. There can be one or more target item sequences.

[0054] In some alternative implementations, step S150, "permuting and combining multiple target items to obtain multiple candidate item sequences", may include: enumerating and permuting at least two target items from the multiple target items, and taking each permutation and combination as a candidate item sequence.

[0055] Specifically, to avoid overlooking possible "one thing" scenarios, every permutation and combination of at least two target items from all target items is enumerated, and each permutation and combination is treated as a sequence of candidate items. For example, see Figure 2 From N target items, select M target items, where M is an integer greater than or equal to 2 and less than or equal to N. Perform a full permutation of each selected M target items, for example, by recursively generating the permutations. The number of possible sequences of items corresponding to each of the M target items is: The number of permutations and combinations generated through enumeration, i.e., the number of sequences of candidate items, is: .

[0056] In some optional implementations, step S150, "verifying each candidate item sequence to obtain a successfully verified target item sequence", may include: verifying the validity, compliance, and feasibility of the inter-item dependencies for each candidate item sequence to obtain a successfully verified candidate item sequence; and determining the target item sequence to be constructed from the successfully verified candidate item sequences based on the existing item sequences corresponding to the constructed centralized service agent.

[0057] By validating and filtering each candidate item sequence, the target item sequence corresponding to the new "one-stop" scenario to be constructed is obtained. Specifically, the dependency relationship between items can refer to the order of the target items in the candidate item sequence. The validity verification process of the dependency relationship between items can be as follows: the language processing model is guided by prompt words to identify whether the dependency relationship between items in each candidate item sequence conforms to the dependency relationship stipulated in the policy document. For example, the "newborn household registration" item needs to precede the "medical insurance enrollment" item. If the dependency relationship between all items in the candidate item sequence conforms to the dependency relationship stipulated in the policy document, then the candidate item sequence is determined to be a valid candidate item sequence. The compliance verification process can be as follows: the target items in each candidate item sequence are checked for compliance by matching each candidate item sequence with relevant policy basis. The feasibility verification process can be as follows: a prediction model is trained in advance using historical work order data to predict the success rate of combining all items in the item sequence. For example, the "enterprise registration" and "tax registration" items have a high material reuse rate, making joint processing more feasible in actual scenarios. In practical applications, each candidate item sequence is input into the prediction model to predict its success rate, thus obtaining the success rate corresponding to each candidate item sequence. Candidate item sequences with a success rate greater than a preset probability are identified as feasible. Through validation of the validity, compliance, and feasibility of inter-item dependencies, valid, compliant, and feasible candidate item sequences are selected from all candidate item sequences; these are the validated candidate item sequences. The existing item sequences corresponding to the constructed centralized service agent are compared with the validated candidate item sequences. Existing item sequences are removed from the validated candidate item sequences, and the remaining candidate item sequences are used as target item sequences to be constructed. This process eliminates existing "one-thing" scenarios and uncovers new "one-thing" scenarios. See also... Figure 2 Through compliance verification and feasibility assessment, the final selected permutations and combinations are obtained, i.e., the target item sequence is: {Item 1, Item 2, Item 3}.

[0058] S160. Construct a centralized service agent corresponding to the target item sequence, so as to provide centralized services for all target items in the target item sequence based on the centralized service agent.

[0059] The centralized service agent can be an intelligent module that integrates the services of all target items in a target item sequence. This allows the centralized service agent to centrally provide services for all target items in the sequence. Each target item sequence corresponds to a new "one-stop" scenario. By constructing a corresponding centralized service agent for each target item sequence, centralized services for personalized "one-stop" scenarios can also be achieved, improving the coverage of centralized services and thus ensuring efficiency for various groups.

[0060] For example, taking the "one-stop service for newborn birth" scenario as an example, before the centralized service for this scenario was implemented, users needed to make 7 trips, submit 29 documents, go through 10 processing steps, and complete the process in 30 working days. Furthermore, there were instances of duplicate document submissions when handling different matters. After the centralized service for this scenario was implemented, users only need to fill in all the required information on a single webpage. The system can then automatically complete the process according to the dependencies between the matters, reducing the number of trips to 0, the number of documents submitted to 4, the processing step to 1, and the processing time to 3 working days. The problem of duplicate document submissions is also eliminated. It is evident that the centralized service for the "one-stop service" scenario significantly improves the efficiency of government services and facilitates business for the public and businesses.

[0061] In some optional implementations, step S160 may include: determining the target business process corresponding to the target item sequence based on the business system interface corresponding to each target item in the target item sequence and the dependencies between target items; constructing a centralized service agent corresponding to the target item sequence based on the target business process; and testing the constructed centralized service agent based on the target test cases corresponding to the target item sequence to obtain a normally functioning centralized service agent.

[0062] Specifically, the business system interface corresponding to the target matter can refer to the interface of the business system used to implement the target matter. Through this interface, data and tools can be exchanged between the centralized intelligent agent and the business system, thereby enabling the centralized intelligent agent to process the target matter. See also Figure 2For each target item sequence, the process is analyzed based on the business system interface corresponding to each target item in the sequence and the dependencies between target items to obtain the target business process corresponding to the new "one thing" scenario. Based on this target business process, a centralized service agent is developed and tested to obtain the centralized service agent corresponding to the new "one thing" scenario. The business system interface corresponding to each target item can achieve data and tool interaction between the centralized agent and the interface through the Model Context Protocol (MCP). If the centralized service agent corresponding to the target item sequence needs to call other centralized service agents, communication and tool invocation functions between different agents are achieved through the Agent-to-Agent Protocol (A2A). After constructing the centralized service agent corresponding to the target item sequence, a batch of target test cases that need to be processed in the new "one thing" scenario can be obtained. All forms and files are submitted to the centralized service agent at once to observe whether the centralized service agent can complete the normal full-process processing. If it can, the centralized service agent is running normally, meaning the "one thing" scenario is successfully constructed; otherwise, it continues to iterate until it is completely normal. After successful testing, the normally functioning centralized service agent will be integrated into the government service business system so that it can provide centralized services for "one-stop" scenarios, thereby improving efficiency.

[0063] The technical solution of this embodiment obtains multiple historical requests and the content of each historical request; determines the risk level of each historical request based on its content; identifies multiple high-risk target requests from all historical requests based on their risk levels, and clusters all target requests to obtain a set of target requests belonging to the same category; determines multiple target items corresponding to the target request set based on the content of each target request in the target request set; arranges and combines all target items to obtain multiple candidate item sequences, and verifies each candidate item sequence to obtain a sequence of target items that has been verified. The sequence of target items is the new "one-thing" scenario, thereby automatically mining new "one-thing" scenarios with personalized needs from historical requests. By constructing a centralized service agent corresponding to the target item sequence, centralized services are provided for all target items in the target item sequence based on the centralized service agent, thereby realizing centralized services for various types of "one-thing" scenarios and improving the coverage of centralized services.

[0064] Figure 3This is a flowchart illustrating another centralized service agent construction method according to an embodiment of the present invention. Based on the above embodiments, this embodiment describes in detail the process of determining the urgency, impact, and difficulty of resolution for each historical request. Explanations of terms that are the same as or corresponding to those in the above embodiments are not repeated here.

[0065] See Figure 3 Another centralized service agent construction method provided in this embodiment specifically includes the following steps:

[0066] S310. Obtain multiple historical requests and the content of each historical request.

[0067] S320. Based on the content of each historical request, determine the degree of risk of time limit failure, emotional intensity, and frequency of repeated complaints for each historical request, and determine the degree of urgency for each historical request based on the degree of risk of time limit failure, emotional intensity, and frequency of repeated complaints.

[0068] In this embodiment, the urgency level of each historical request is measured and calculated from three dimensions: the risk of time-limit expiration, the intensity of emotions, and the frequency of repeated complaints, thereby improving the accuracy of urgency calculation. After determining the risk of time-limit expiration, the intensity of emotions, and the frequency of repeated complaints for each historical request, these factors can be weighted and summed to obtain the urgency level. For example, urgency level = w1 × risk of time-limit expiration + w2 × intensity of emotions + w3 × frequency of repeated complaints, where w1, w2, and w3 are the weights of the risk of time-limit expiration, the intensity of emotions, and the frequency of repeated complaints, respectively, and w1 + w2 + w3 = 1. Since the risk of time-limit expiration is more important, w1 ≥ 0.4 can be set.

[0069] In some optional implementations, step S320, "determining the risk level of time limit expiration, emotional intensity, and frequency of repeated complaints for each historical request based on the content of the request," may include: for each historical request, determining the risk level of time limit expiration based on the required completion time and the time when the historical request was submitted; extracting emotional keywords from the content of the historical request and determining the emotional intensity based on the emotional intensity corresponding to the emotional keywords; determining the number of complaints for the historical request within a preset time period and defining the number of complaints as the frequency of repeated complaints for the historical request.

[0070] Specifically, for each historical request, if the required completion time for the historical request is N days, and X days have passed since the request was submitted, the ratio of X to N can be determined as the risk level of the time limit lapse for that historical request. If the risk level of the time limit lapse is a value between 0 and 100, then the result of multiplying the ratio by 100 can be used as the risk level of the time limit lapse for that historical request. As X increases from 0 to N, the risk level of the time limit lapse also gradually increases from 0 to 100. Extract emotional keywords from the content of the historical request, such as words like "immediately," "right away," "must," "strongly," and "multiple complaints unresolved." The presence of these emotional keywords in the request indicates that the complainant is emotionally agitated and may require urgent handling. Based on a pre-defined correspondence between emotional keywords and emotional intensity, the emotional intensity corresponding to the emotional keywords in the content of the historical request can be determined as the emotional intensity of that historical request. This emotional intensity can also be a value between 0 and 100. The system can count the number of complaints for a historical request within a preset time period, such as the number of times a historical request has been repeatedly complained about within the past year. This number of complaints can be directly determined as the repeat complaint frequency for that historical request. If the number of complaints is greater than 100, the repeat complaint frequency can be set to 100 to ensure that the repeat complaint frequency is also a value between 0 and 100.

[0071] S330. Based on the content of each historical demand, determine the spatial impact scale, group vulnerability, and affected object scale of each historical demand, and determine the degree of impact of each historical demand based on the spatial impact scale, group vulnerability, and affected object scale.

[0072] In this embodiment, the impact of each historical claim is measured and calculated from three dimensions: spatial impact scale, group vulnerability, and the scale of affected objects, thereby improving the accuracy of the impact calculation. After determining the spatial impact scale, group vulnerability, and the scale of affected objects for each historical claim, these three dimensions can be weighted and summed, and the sum is taken as the impact degree. For example, impact degree = w4 × spatial impact scale + w5 × group vulnerability + w6 × scale of affected objects, where w4, w5, and w6 are the weights of spatial impact scale, group vulnerability, and the scale of affected objects, respectively, and w4 + w5 + w6 = 1. Since spatial impact scale is more important, w4 ≥ 0.5 can be set.

[0073] In some optional implementations, step S330, "determining the spatial impact scale, group vulnerability, and affected object scale of each historical demand based on the content of each historical demand," may include: for each historical demand, determining the target area corresponding to that historical demand and the number of demands belonging to the same category as that historical demand appearing in the target area, and determining the spatial impact scale corresponding to that historical demand based on the number of demands; detecting whether there are keywords corresponding to vulnerable groups in the content of the historical demand, and determining the group vulnerability corresponding to that historical demand based on the detection results; and determining the affected object scale corresponding to that historical demand based on the number of people affected by that historical demand and the total number of people in the area to which that historical demand belongs.

[0074] Specifically, for each historical demand, by locating the region to which it belongs, the target region corresponding to that historical demand is obtained. The number of demands of the same type appearing within the target region can then be counted, and this number can be directly determined as the spatial impact scale corresponding to that historical demand. If the number of demands is greater than 100, the spatial impact scale can be set to 100, ensuring that the spatial impact scale is a value between 0 and 100. As the number of demands increases, it indicates a gradually increasing likelihood of the demands covering a large area, which requires attention. The process involves checking if the content of the historical appeal contains keywords related to vulnerable groups. For example, keywords related to "elderly" include "elderly," "pension," and "medical insurance reimbursement," while keywords related to "disabled persons" include "sign language." If the appeal contains keywords related to vulnerable groups, the appealing group is considered relatively vulnerable and may require priority handling. In this case, the vulnerability of the group can be determined as the maximum value, such as 100. If the appeal does not contain keywords related to vulnerable groups, the vulnerability of the group can be determined as the minimum value, such as 0. The ratio of the number of people affected by the historical appeal to the total population of the region to which the appeal is located can be used to determine the scale of the affected group. If the scale of the affected group is between 0 and 100, the ratio can be multiplied by 100 to obtain the scale of the affected group. For example, a company's appeal may be related to its registered capital and number of employees. If the appeal is not resolved in a timely manner, it may affect all employees of the company, thus requiring consideration of the scale of the affected group.

[0075] S340. Based on the content of each historical demand, determine the departmental collaboration coefficient, policy clarity, and resource accessibility for each historical demand, and determine the difficulty of resolving each historical demand based on the departmental collaboration coefficient, policy clarity, and resource accessibility.

[0076] In this embodiment, the difficulty of resolving each historical request is measured and calculated from three dimensions: departmental collaboration coefficient, policy clarity, and resource accessibility, thereby improving the accuracy of the difficulty calculation. After determining the departmental collaboration coefficient, policy clarity, and resource accessibility for each historical request, these factors can be weighted and summed, and the sum is used as the difficulty of resolving the request. For example, the degree of impact = w7 × departmental collaboration coefficient + w8 × policy clarity + w9 × resource accessibility, where w7, w8, and w9 are the weights of the departmental collaboration coefficient, policy clarity, and resource accessibility, respectively, and w7 + w8 + w9 = 1. Since the departmental collaboration coefficient is more important, w7 can be set to ≥ 0.6.

[0077] In some optional implementations, step S340, "determining the departmental collaboration coefficient, policy clarity, and resource accessibility for each historical request based on the content of each historical request," may include: for each historical request, determining the departmental collaboration coefficient based on the number of departments involved in processing the historical request; detecting whether there are relevant policy bases for processing the historical request, and determining the policy clarity for the historical request based on the detection results; and determining the resource accessibility for the historical request based on the cost information for processing the historical request.

[0078] Specifically, for each historical demand, handling it may involve the collaborative efforts of multiple departments. For example, a demand regarding noise from entertainment venues might require the cooperation of departments such as culture and tourism, environmental protection, urban management, and market supervision. The more departments involved, the higher the cross-departmental collaboration cost for that historical demand. Therefore, the number of departments involved in handling a historical demand can be directly determined as the departmental collaboration coefficient corresponding to that historical demand. Since the number of government departments at the same level generally does not exceed 100, the departmental collaboration coefficient ranges from 1 to 100. It is also necessary to check whether there is a relevant policy basis for handling the historical demand. If no relevant policy basis exists, it may make resolving the demand difficult. In this case, the policy clarity corresponding to the historical demand can be determined to be the minimum value, such as 0. If there is a relevant policy basis, the policy clarity corresponding to the historical demand can be determined to be the maximum value, such as 100. For requests that arise at night or during holidays, it may be necessary to urgently allocate human and material resources, resulting in corresponding costs. This cost information can be normalized, and the result of multiplying the normalized cost information by 100 can be used as the resource accessibility corresponding to the historical request, so that the resource accessibility is a value between 0 and 100.

[0079] S350. Determine the risk level of each historical demand based on its urgency, impact, and difficulty of resolution.

[0080] By integrating the urgency, impact, and difficulty of resolution for each historical request, a combined risk level is obtained, thereby improving the accuracy of risk level determination and further enhancing the accuracy of "one-event" scenario mining. For example, the urgency, impact, and difficulty of resolution for each historical request can be multiplied together, and the result can be used as the corresponding risk level. Alternatively, the urgency, impact, and difficulty of resolution for each historical request can be added together, and the result can be used as the corresponding risk level. This embodiment does not specifically limit the integration method.

[0081] S360. Based on the risk level corresponding to each historical demand, determine multiple target demands from multiple historical demands, and cluster the multiple target demands to obtain a set of target demands belonging to the same category.

[0082] S370. Based on the content of each target demand in the target demand set, determine the multiple target items corresponding to the target demand set.

[0083] S380. Arrange and combine multiple target items to obtain multiple candidate item sequences, and verify each candidate item sequence to obtain a target item sequence that has been successfully verified.

[0084] S390. Construct a centralized service agent corresponding to the target item sequence, so as to provide centralized services for all target items in the target item sequence based on the centralized service agent.

[0085] The technical solution of this embodiment can more accurately determine the urgency of each historical request by considering the risk level of time-limit failure, emotional intensity, and frequency of repeated complaints. It can also more accurately determine the impact of each historical request by considering its spatial impact scale, group vulnerability, and the scale of affected individuals. Furthermore, it can more accurately determine the difficulty of resolving each historical request by considering its departmental collaboration coefficient, policy clarity, and resource accessibility. This further improves the accuracy of risk assessment and enables automated discovery of "one-stop" scenarios, eliminating the need for manual analysis by business personnel and significantly enhancing the level of intelligence.

[0086] The following are embodiments of the centralized service agent construction device provided in this invention. This device and the centralized service agent construction method of the above embodiments belong to the same inventive concept. For details not described in detail in the embodiments of the centralized service agent construction device, please refer to the embodiments of the centralized service agent construction method described above.

[0087] Figure 4 This is a schematic diagram of a centralized service agent construction device provided in an embodiment of the present invention. This embodiment can be applied to situations where new "one-thing" scenarios are explored and corresponding centralized service agents are constructed. Figure 4 As shown, the device specifically includes: a historical demand acquisition module 410, a risk level determination module 420, a target demand set determination module 430, a target item determination module 440, a target item sequence determination module 450, and a centralized service intelligent agent construction module 460.

[0088] The system includes the following modules: a historical request acquisition module 410, used to acquire multiple historical requests and the content of each historical request; a risk level determination module 420, used to determine the risk level of each historical request based on its content; a target request set determination module 430, used to determine multiple target requests from the multiple historical requests based on their risk levels and cluster them to obtain a set of target requests belonging to the same category; a target item determination module 440, used to determine multiple target items corresponding to the target request set based on the content of each target request in the target request set; a target item sequence determination module 450, used to arrange and combine the multiple target items to obtain multiple candidate item sequences and verify each candidate item sequence to obtain a sequence of successfully verified target items; and a centralized service agent construction module 460, used to construct a centralized service agent corresponding to the target item sequence, so as to provide centralized services for all target items in the target item sequence based on the centralized service agent.

[0089] The technical solution of this embodiment obtains multiple historical requests and the content of each historical request; determines the risk level of each historical request based on its content; identifies multiple high-risk target requests from all historical requests based on their risk levels, and clusters all target requests to obtain a set of target requests belonging to the same category; determines multiple target items corresponding to the target request set based on the content of each target request in the target request set; arranges and combines all target items to obtain multiple candidate item sequences, and verifies each candidate item sequence to obtain a sequence of target items that has been verified. The sequence of target items is the new "one-thing" scenario, thereby automatically mining new "one-thing" scenarios with personalized needs from historical requests. By constructing a centralized service agent corresponding to the target item sequence, centralized services are provided for all target items in the target item sequence based on the centralized service agent, thereby realizing centralized services for various types of "one-thing" scenarios and improving the coverage of centralized services.

[0090] Optionally, the risk level determination module 420 includes:

[0091] The historical demands analysis unit is used to determine the urgency, impact, and difficulty of resolution of each historical demand based on its content.

[0092] The risk level determination unit is used to determine the risk level of each historical demand based on its urgency, impact, and difficulty of resolution.

[0093] Optionally, the historical demands analysis unit includes:

[0094] The first analysis subunit is used to determine the degree of time limit failure risk, emotional intensity and frequency of repeated complaints for each historical request based on the content of the request, and to determine the urgency level for each historical request based on the degree of time limit failure risk, the emotional intensity and the frequency of repeated complaints.

[0095] The second analysis subunit is used to determine the spatial impact scale, group vulnerability, and affected object scale of each historical demand based on the content of the demand, and to determine the impact degree of each historical demand based on the spatial impact scale, the group vulnerability, and the affected object scale.

[0096] The third analysis subunit is used to determine the departmental collaboration coefficient, policy clarity, and resource accessibility corresponding to each historical demand based on the content of the demand, and to determine the difficulty of solving each historical demand based on the departmental collaboration coefficient, the policy clarity, and the resource accessibility.

[0097] Optionally, the first analysis subunit is specifically used for:

[0098] For each historical request, the degree of risk of the time limit becoming invalid is determined based on the required completion time of the historical request and the time when the historical request was submitted.

[0099] Extract emotional keywords from the content of the historical appeal and determine the emotional intensity of the historical appeal based on the intensity of the emotions corresponding to the emotional keywords.

[0100] Determine the number of complaints made regarding the historical request within a preset time period, and define the number of complaints as the frequency of repeated complaints corresponding to the historical request.

[0101] Optionally, the second analysis subunit is specifically used for:

[0102] For each historical claim, determine the target area corresponding to the historical claim and the number of claims that belong to the same category as the historical claim appear in the target area, and determine the spatial impact scale corresponding to the historical claim based on the number of claims.

[0103] The system detects whether keywords corresponding to vulnerable groups exist in the content of the historical demands, and determines the vulnerability of the group corresponding to the historical demands based on the detection results.

[0104] The scale of those affected by the historical claim is determined based on the number of people affected by the historical claim and the total number of people in the area to which the historical claim belongs.

[0105] Optionally, the third analysis subunit is specifically used for:

[0106] For each historical request, the departmental collaboration coefficient corresponding to that historical request is determined based on the number of departments involved in handling it.

[0107] The investigation will determine whether there are relevant policy bases for addressing the historical request, and based on the investigation results, determine the policy clarity corresponding to the historical request.

[0108] The availability of resources corresponding to the historical request is determined based on the cost information for processing the historical request.

[0109] Optionally, the target item determination module 440 is specifically used for:

[0110] For each target demand in the target demand set, the content of the target demand and the corresponding target policy document are input into the language processing model to identify target matters related to the target demand from the target policy document;

[0111] The set of target requests is deduplicated to obtain all target items after deduplication.

[0112] Optionally, the target item sequence determination module 450 is specifically used for:

[0113] At least two of the multiple target items are enumerated and combined, and each of the resulting permutations and combinations is used as a sequence of candidate items.

[0114] Optionally, the target item sequence determination module 450 is specifically used for:

[0115] For each candidate item sequence, the validity, compliance, and feasibility of the inter-item dependencies are verified to obtain the candidate item sequence that has passed the verification.

[0116] Based on the existing item sequences corresponding to the constructed centralized service agent, the target item sequence to be constructed is determined from the successfully verified candidate item sequences.

[0117] Optionally, the centralized service agent building module 460 is specifically used for:

[0118] Based on the business system interface corresponding to each target item in the target item sequence and the dependencies between target items, the target business process corresponding to the target item sequence is determined;

[0119] Based on the target business process, a centralized service agent corresponding to the target item sequence is constructed, and the constructed centralized service agent is tested based on the target test cases corresponding to the target item sequence to obtain a normally functioning centralized service agent.

[0120] The centralized service agent construction device provided in the embodiments of the present invention can execute the centralized service agent construction method provided in any embodiment of the present invention, and has the corresponding functional modules and beneficial effects of executing the centralized service agent construction method.

[0121] It is worth noting that in the embodiments of the above-mentioned centralized service intelligent agent construction device, the various units and modules included are only divided according to functional logic, but are not limited to the above division, as long as the corresponding functions can be realized; in addition, the specific names of each functional unit are only for easy differentiation and are not used to limit the scope of protection of the present invention.

[0122] Figure 5A schematic diagram of an electronic device 10, which can be used to implement embodiments of the present invention, is shown. The electronic device is intended to represent various forms of digital computers, such as laptop computers, desktop computers, workstations, personal digital assistants, servers, blade servers, mainframe computers, and other suitable computers. The electronic device can also represent various forms of mobile devices, such as personal digital processors, cellular phones, smartphones, wearable devices (e.g., helmets, glasses, watches, etc.), and other similar computing devices. The components shown herein, their connections and relationships, and their functions are merely illustrative and are not intended to limit the implementation of the invention described and / or claimed herein.

[0123] like Figure 5 As shown, the electronic device 10 includes at least one processor 11 and a memory, such as a read-only memory (ROM) 12 or a random access memory (RAM) 13, communicatively connected to the at least one processor 11. The memory stores computer programs executable by the at least one processor. The processor 11 can perform various appropriate actions and processes based on the computer program stored in the ROM 12 or loaded from storage unit 18 into the RAM 13. The RAM 13 can also store various programs and data required for the operation of the electronic device 10. The processor 11, ROM 12, and RAM 13 are interconnected via a bus 14. An input / output (I / O) interface 15 is also connected to the bus 14.

[0124] Multiple components in electronic device 10 are connected to I / O interface 15, including: input unit 16, such as keyboard, mouse, etc.; output unit 17, such as various types of displays, speakers, etc.; storage unit 18, such as disk, optical disk, etc.; and communication unit 19, such as network card, modem, wireless transceiver, etc. Communication unit 19 allows electronic device 10 to exchange information / data with other devices through computer networks such as the Internet and / or various telecommunications networks.

[0125] Processor 11 can be a variety of general-purpose and / or special-purpose processing components with processing and computing capabilities. Some examples of processor 11 include, but are not limited to, central processing unit (CPU), graphics processing unit (GPU), various special-purpose artificial intelligence (AI) computing chips, various processors running machine learning model algorithms, digital signal processors (DSPs), and any suitable processor, controller, microcontroller, etc. Processor 11 executes the various centralized service agent construction methods and processes described above.

[0126] In some embodiments, the centralized service agent construction method may be implemented as a computer program tangibly contained in a computer-readable storage medium, such as storage unit 18. In some embodiments, part or all of the computer program may be loaded and / or installed on electronic device 10 via ROM 12 and / or communication unit 19. When the computer program is loaded into RAM 13 and executed by processor 11, one or more steps of the centralized service agent construction method described above may be performed. Alternatively, in other embodiments, processor 11 may be configured to execute the centralized service agent construction method by any other suitable means (e.g., by means of firmware).

[0127] Various embodiments of the systems and techniques described above herein can be implemented in digital electronic circuit systems, integrated circuit systems, field-programmable gate arrays (FPGAs), application-specific integrated circuits (ASICs), application-specific standard products (ASSPs), systems-on-a-chip (SoCs), payload-programmable logic devices (CPLDs), computer hardware, firmware, software, and / or combinations thereof. These various embodiments may include implementations in one or more computer programs that can be executed and / or interpreted on a programmable system including at least one programmable processor, which may be a dedicated or general-purpose programmable processor, capable of receiving data and instructions from a storage system, at least one input device, and at least one output device, and transmitting data and instructions to the storage system, the at least one input device, and the at least one output device.

[0128] Computer programs used to implement the centralized service agent construction method of the present invention can be written in any combination of one or more programming languages. These computer programs can be provided to a processor of a general-purpose computer, a special-purpose computer, or other programmable data processing device, such that when executed by the processor, the computer programs cause the functions / operations specified in the flowcharts and / or block diagrams to be implemented. The computer programs can be executed entirely on a machine, partially on a machine, as a standalone software package partially on a machine and partially on a remote machine, or entirely on a remote machine or server.

[0129] In the context of this invention, a computer-readable storage medium can be a tangible medium that may contain or store a computer program for use by or in conjunction with an instruction execution system, apparatus, or device. A computer-readable storage medium may include, but is not limited to, electronic, magnetic, optical, electromagnetic, infrared, or semiconductor systems, apparatus, or devices, or any suitable combination thereof. Alternatively, a computer-readable storage medium may be a machine-readable signal medium. More specific examples of machine-readable storage media include electrical connections based on one or more wires, portable computer disks, hard disks, random access memory (RAM), read-only memory (ROM), erasable programmable read-only memory (EPROM or flash memory), optical fibers, portable compact disk read-only memory (CD-ROM), optical storage devices, magnetic storage devices, or any suitable combination thereof.

[0130] To provide interaction with a user, the systems and techniques described herein can be implemented on an electronic device having: a display device (e.g., a CRT (cathode ray tube) or LCD (liquid crystal display) monitor) for displaying information to the user; and a keyboard and pointing device (e.g., a mouse or trackball) through which the user provides input to the electronic device. Other types of devices can also be used to provide interaction with the user; for example, feedback provided to the user can be any form of sensory feedback (e.g., visual feedback, auditory feedback, or tactile feedback); and input from the user can be received in any form (including sound input, voice input, or tactile input).

[0131] The systems and technologies described herein can be implemented in computing systems that include backend components (e.g., as data servers), or middleware components (e.g., application servers), or frontend components (e.g., user computers with graphical user interfaces or web browsers through which users can interact with implementations of the systems and technologies described herein), or any combination of such backend, middleware, or frontend components. The components of the system can be interconnected via digital data communication of any form or medium (e.g., communication networks). Examples of communication networks include local area networks (LANs), wide area networks (WANs), blockchain networks, and the Internet.

[0132] A computing system can include clients and servers. Clients and servers are generally located far apart and typically interact through communication networks. The client-server relationship is created by computer programs running on the respective computers and having a client-server relationship with each other. The server can be a cloud server, also known as a cloud computing server or cloud host, which is a hosting product within the cloud computing service system to address the shortcomings of traditional physical hosts and VPS services, such as high management difficulty and weak business scalability.

[0133] This invention also provides a computer program product, including a computer program that, when executed by a processor, implements the centralized service agent construction method provided in any embodiment of this invention.

[0134] In the implementation of a computer program product, computer program code for performing the operations of this invention can be written in one or more programming languages ​​or a combination thereof. Programming languages ​​include object-oriented programming languages ​​as well as conventional procedural programming languages. The program code can be executed entirely on the user's computer, partially on the user's computer, as a standalone software package, partially on the user's computer and partially on a remote computer, or entirely on a remote computer or server. In cases involving remote computers, the remote computer can be connected to the user's computer via any type of network—including a local area network (LAN) or a wide area network (WAN)—or can be connected to an external computer (e.g., via the Internet using an Internet service provider).

[0135] It should be understood that the various forms of processes shown above can be used, with steps reordered, added, or deleted. For example, the steps described in this invention can be executed in parallel, sequentially, or in different orders, as long as the desired result of the technical solution of this invention can be achieved, and this is not limited herein.

[0136] The specific embodiments described above do not constitute a limitation on the scope of protection of this invention. Those skilled in the art should understand that various modifications, combinations, sub-combinations, and substitutions can be made according to design requirements and other factors. Any modifications, equivalent substitutions, and improvements made within the spirit and principles of this invention should be included within the scope of protection of this invention.

Claims

1. A centralized service agent construction method, characterized in that, The method comprises the following steps: obtaining a plurality of historical appeals and appeal content corresponding to each historical appeal; determining the risk degree corresponding to each historical appeal according to the appeal content corresponding to each historical appeal; determining a plurality of target appeals from the plurality of historical appeals according to the risk degree corresponding to each historical appeal, and clustering the plurality of target appeals to obtain a target appeal set belonging to the same category; determining a plurality of target matters corresponding to the target appeal set according to the appeal content of each target appeal in the target appeal set; arranging and combining the plurality of target matters to obtain a plurality of candidate matter sequences, and verifying each candidate matter sequence to obtain a target matter sequence that passes the verification; constructing a centralized service agent corresponding to the target matter sequence to centrally serve all target matters in the target matter sequence based on the centralized service agent.

2. The method of claim 1, wherein, The method further comprises the following steps: determining the risk degree corresponding to each historical appeal according to the appeal content corresponding to each historical appeal, comprising: determining the urgency, impact and difficulty of solving corresponding to each historical appeal according to the appeal content corresponding to each historical appeal; 3. The method of claim 2, wherein, determining the risk degree corresponding to each historical appeal according to the urgency, impact and difficulty of solving corresponding to each historical appeal. The method further comprises the following steps: determining the urgency, impact and difficulty of solving corresponding to each historical appeal according to the appeal content corresponding to each historical appeal, comprising: determining the time limit failure risk degree, emotional intensity and repeated complaint frequency corresponding to each historical appeal according to the appeal content corresponding to each historical appeal, and determining the urgency corresponding to each historical appeal according to the time limit failure risk degree, the emotional intensity and the repeated complaint frequency; 4. The method of claim 3, wherein, determining the spatial impact scale, group vulnerability and affected object scale corresponding to each historical appeal according to the appeal content corresponding to each historical appeal, and determining the impact corresponding to each historical appeal according to the spatial impact scale, the group vulnerability and the affected object scale; determining the department cooperation coefficient, policy clarity and resource accessibility corresponding to each historical appeal according to the appeal content corresponding to each historical appeal, and determining the difficulty of solving corresponding to each historical appeal according to the department cooperation coefficient, the policy clarity and the resource accessibility. The method further comprises the following steps: determining the time limit failure risk degree corresponding to each historical appeal according to the required handling completion time of each historical appeal and the time when the historical appeal is proposed; 5. The method of claim 3, wherein, extracting emotional keywords in the appeal content corresponding to the historical appeal, and determining the emotional intensity corresponding to the historical appeal based on the emotional intensity corresponding to the emotional keywords; determining the number of complaints of the historical appeal within a preset time period, and determining the number of complaints as the repeated complaint frequency corresponding to the historical appeal. The method further comprises the following steps: determining the spatial impact scale, group vulnerability and affected object scale corresponding to each historical appeal according to the appeal content corresponding to each historical appeal, comprising: For each historical appeal, determine the target area corresponding to the historical appeal and the number of appeals that appear in the target area and belong to the same type of appeal as the historical appeal, and determine the spatial influence scale of the historical appeal based on the number of appeals; Detect whether there is a keyword corresponding to a vulnerable group in the appeal content corresponding to the historical appeal, and determine the group vulnerability corresponding to the historical appeal based on the detection result; According to the number of affected personnel of the historical appeal and the total number of personnel in the region to which the historical appeal belongs, determine the affected object scale corresponding to the historical appeal.

6. The method of claim 3, wherein, According to the appeal content corresponding to each historical appeal, determine the department cooperation coefficient, policy clarity and resource accessibility corresponding to each historical appeal, including: For each historical appeal, determine the department cooperation coefficient corresponding to the historical appeal according to the number of processing departments involved in processing the historical appeal; Detect whether there is a relevant policy basis for processing the historical appeal, and determine the policy clarity corresponding to the historical appeal based on the detection result; According to the cost information of processing the historical appeal, determine the resource accessibility corresponding to the historical appeal.

7. The method of claim 1, wherein, According to the appeal content of each target appeal in the target appeal set, determine a plurality of target matters corresponding to the target appeal set, including: For each target appeal in the target appeal set, input the appeal content of the target appeal and the target policy file corresponding to the target appeal into a language processing model to identify target matters associated with the target appeal from the target policy file; All target matters corresponding to the target appeal set are processed to obtain all target matters after deduplication.

8. The method of claim 1, wherein, The arrangement and combination of the plurality of target matters to obtain a plurality of candidate matter sequences, including: At least two target matters in the plurality of target matters are enumerated and combined, and each obtained arrangement and combination is taken as a candidate matter sequence.

9. The method of claim 1, wherein, The verification of each candidate matter sequence to obtain a target matter sequence that passes the verification, including: Each candidate matter sequence is verified for the validity, compliance and feasibility of the inter-matter dependency relationship to obtain a candidate matter sequence that passes the verification; According to the existing matter sequence corresponding to the constructed centralized service agent, determine the target matter sequence to be constructed from the candidate matter sequence that passes the verification.

10. The method according to any one of claims 1 to 9, characterized in that, The construction of the centralized service agent corresponding to the target matter sequence, including: Based on the business system interface and the inter-target matter dependency relationship of each target matter in the target matter sequence, determine the target business process corresponding to the target matter sequence; Based on the target business process, construct the centralized service agent corresponding to the target matter sequence, and based on the target test case corresponding to the target matter sequence, test the constructed centralized service agent to obtain a normally operating centralized service agent. 11.A centralized service intelligent agent construction apparatus characterized by comprising: Including: A historical appeal acquisition module is configured to acquire a plurality of historical appeals and appeal content corresponding to each historical appeal; A risk degree determination module is configured to determine the risk degree corresponding to each historical appeal according to the appeal content corresponding to each historical appeal; The target appeal set determination module is configured to determine a plurality of target appeals from the plurality of historical appeals according to a risk degree corresponding to each historical appeal, and cluster the plurality of target appeals to obtain a target appeal set belonging to a same category; The target matter determination module is configured to determine a plurality of target matters corresponding to the target appeal set according to appeal content of each target appeal in the target appeal set; The target matter sequence determination module is configured to arrange and combine the plurality of target matters to obtain a plurality of candidate matter sequences, and check each candidate matter sequence to obtain a target matter sequence that passes the check; The centralized service agent construction module is configured to construct a centralized service agent corresponding to the target matter sequence, and perform centralized service on all target matters in the target matter sequence based on the centralized service agent.

12. An electronic device, comprising: The electronic device comprises: at least one processor; and a memory connected with the at least one processor in communication; wherein the memory stores a computer program that can be executed by the at least one processor, and the computer program is executed by the at least one processor to enable the at least one processor to perform the centralized service agent construction method in any one of claims 1-10.

13. A computer-readable storage medium, characterized in that, The computer readable storage medium stores computer instructions for enabling the processor to perform the centralized service agent construction method in any one of claims 1-10 when executed by the processor.

14. A computer program product comprising a computer program, characterized in that, The computer program, when executed by the processor, implements the centralized service agent construction method in any one of claims 1-10.

Images