Insurance agent team management strategy generation method and device, equipment and medium

Abstract

The embodiment of the invention provides an insurance seat team management strategy generation method and device, equipment and a medium, and belongs to the technical field of artificial intelligence, and the method comprises the steps: obtaining a team portrait of an insurance seat team; constructing multi-dimensional features of the insurance seat team based on the multi-source data corresponding to the team portrait; inputting the multi-dimensional features into a preset multi-layer analysis model to obtain a model analysis result of the multi-dimensional features by the multi-layer analysis model; and generating a management strategy of the insurance agent team based on the model analysis result. By adopting the technical scheme provided by the embodiment of the invention, the management strategy of the insurance seat team can be intelligently generated based on the team portrait of the insurance seat team.

Description

Technical Field

[0001] This application relates to the field of artificial intelligence technology, and in particular to a method, apparatus, device, and medium for generating insurance agent team management strategies. Background Technology

[0002] Competition in the insurance industry is becoming increasingly fierce, and customer needs are becoming more diverse. The importance of the insurance customer service team, as a key link between the company and its customers, is self-evident. The insurance customer service team is not only responsible for answering customer inquiries and handling claims, but also bears the heavy responsibility of sales conversion, directly impacting company performance.

[0003] In related technologies, the management of insurance agent teams still relies on traditional models. This lack of timely understanding of team structure, personnel characteristics, skill distribution, and potential problems leads to a series of issues. For example, some teams have structural deficiencies, with an imbalance in the ratio of new to experienced employees, resulting in poor business transitions. Furthermore, the varying skill levels of employees in some teams make it difficult to provide accurate and efficient customer service.

[0004] Therefore, how to improve team management methods to enhance the overall business level of insurance agent teams has become a pressing technical problem in the field. Summary of the Invention

[0005] The main objective of this application is to propose a method, apparatus, device, and medium for generating insurance agent team management strategies, aiming to intelligently generate management strategies for insurance agent teams based on team profiles.

[0006] To achieve the above objectives, a first aspect of this application proposes a method for generating an insurance agent team management strategy, the method comprising:

[0007] Obtain a team profile of the insurance agent team;

[0008] The multidimensional features of the insurance agent team are constructed based on the multi-source data corresponding to the team profile.

[0009] The multidimensional features are input into a preset multi-level analysis model to obtain the model analysis results of the multi-level analysis model on the multidimensional features;

[0010] The management strategy for the insurance agent team is generated based on the analysis results of the model.

[0011] In some embodiments, the multi-source data refers to the multi-type data of the insurance agent team collected from multiple data sources to construct the team profile;

[0012] The construction of multi-dimensional features of the insurance agent team based on the multi-source data corresponding to the team profile includes:

[0013] The multi-source data is cleaned according to the preset data cleaning specifications to obtain cleaned multi-source data.

[0014] The insurance agent team is constructed with multidimensional structured features based on the cleaned multi-source data; and multidimensional derived features are constructed with the insurance agent team based on the cleaned multi-source data; the multidimensional features include the multidimensional structured features and the multidimensional derived features.

[0015] In some embodiments, the multi-layer analysis model includes at least a basic analysis model, a deep mining model, and a prediction and early warning model;

[0016] The step of inputting the multidimensional features into a preset multi-layer analysis model includes:

[0017] A standardized multidimensional feature matrix is ​​generated based on the first part of the multidimensional features, and the multidimensional feature matrix is ​​input into the basic analysis model.

[0018] Based on the second part of the multidimensional features, a discretized feature combination is generated, and the feature combination is input into the deep mining model.

[0019] A time-series feature matrix is ​​generated based on the third part of the multidimensional features, and the time-series feature matrix is ​​input into the prediction and early warning model.

[0020] In some embodiments, after inputting the multidimensional feature matrix into the basic analysis model, the method further includes:

[0021] The multidimensional feature matrix is ​​subjected to cluster analysis using the basic analysis model to divide the insurance agent team into multiple groups.

[0022] Obtain the team classification labels corresponding to each of the multiple groups output by the basic analysis model; the model analysis results include multiple team classification labels.

[0023] In some embodiments, after inputting the feature combination into the deep mining model, the method further includes:

[0024] The feature combination is subjected to frequent itemset mining processing by the deep mining model to obtain the association rules corresponding to the feature combination.

[0025] The feature combination is sorted by feature importance using the deep mining model to obtain the key influencing factors in the feature combination;

[0026] Obtain the association rules and key influencing factors output by the deep mining model; the model analysis results include the association rules and key influencing factors.

[0027] In some embodiments, after inputting the time-series feature matrix into the prediction and early warning model, the method further includes:

[0028] The time-series feature matrix is ​​subjected to binary classification prediction processing by the prediction and early warning model to obtain short-term early warning prompts for the insurance agent team.

[0029] The time-series feature matrix is ​​subjected to survival analysis by the prediction and early warning model to obtain the medium- and long-term early warning prompts for the insurance agent team.

[0030] Obtain the short-term and medium-to-long-term early warning prompts output by the prediction and early warning model; the model analysis results include the short-term and medium-to-long-term early warning prompts.

[0031] In some embodiments, obtaining the team profile of the insurance agent team includes:

[0032] The insurance agent team collects various types of data from a pre-defined data source matrix; the data source matrix includes multiple data source systems.

[0033] Based on the collected multi-type data, a multi-dimensional team profile of the insurance agent team is constructed;

[0034] The team profile includes multiple main dimensions, which include at least two of the following: team structure dimension, capability distribution dimension, performance dimension, stability analysis dimension, and collaboration effectiveness dimension; each main dimension includes at least two sub-dimensions.

[0035] To achieve the above objectives, a second aspect of this application provides an apparatus for generating an insurance agent team management strategy, the apparatus comprising:

[0036] The acquisition module is used to obtain team profiles of insurance agent teams;

[0037] The feature construction module is used to construct multi-dimensional features of the insurance agent team based on the multi-source data corresponding to the team profile;

[0038] The model analysis module is used to input the multidimensional features into a preset multi-level analysis model to obtain the model analysis results of the multi-level analysis model on the multidimensional features;

[0039] The strategy generation module is used to generate management strategies for the insurance agent team based on the analysis results of the model.

[0040] To achieve the above objectives, a third aspect of this application provides a computer device, the computer device including a memory and a processor, the memory storing a computer program, and the processor executing the computer program to implement the method for generating insurance agent team management strategies as described in the first aspect.

[0041] To achieve the above objectives, a fourth aspect of this application provides a computer-readable storage medium storing a computer program that, when executed by a processor, implements the method for generating insurance agent team management strategies as described in the first aspect.

[0042] To achieve the above objectives, a fifth aspect of this application provides a computer program product storing a computer program that, when executed by a processor, implements the method for generating insurance agent team management strategies as described in the first aspect.

[0043] This application proposes a method, apparatus, computer device, computer-readable storage medium, and computer program product for generating insurance agent team management strategies. The method involves: acquiring a team profile of the insurance agent team; constructing multi-dimensional features of the insurance agent team based on multi-source data corresponding to the team profile; inputting the multi-dimensional features into a preset multi-level analysis model to obtain the model analysis results of the multi-level analysis model on the multi-dimensional features; and generating a management strategy for the insurance agent team based on the model analysis results.

[0044] Compared to traditional management methods, this application's embodiments utilize team profiling of insurance agent teams, then construct multi-dimensional characteristics of the insurance agent teams based on multi-source data corresponding to these profilings. These multi-layered analytical models are then used to analyze these multi-dimensional characteristics, yielding model analysis results. Based on these results, management strategies for the insurance agent teams are generated. Thus, by employing a data-driven approach using team profiling to improve team management, and by intelligently generating management strategies based on these profilings, managers can optimize team management with a deep understanding of the team's situation, thereby improving the overall business performance, work efficiency, and customer satisfaction of the insurance agent teams. Attached Figure Description

[0045] Figure 1 The flowchart of the method for generating insurance agent team management strategy provided in this application embodiment is shown in some embodiments.

[0046] Figure 2 for Figure 1 A detailed flowchart of step S101;

[0047] Figure 3 for Figure 1 A detailed flowchart of step S102;

[0048] Figure 4 The method for generating insurance agent team management strategies provided in the embodiments of this application is illustrated in some embodiments with a schematic diagram of a four-layer analysis model architecture.

[0049] Figure 5 for Figure 1 A detailed flowchart of step S103;

[0050] Figure 6 The method for generating insurance agent team management strategies provided in the embodiments of this application includes a schematic diagram of the basic analysis model involved in some embodiments;

[0051] Figure 7 The method for generating insurance agent team management strategies provided in the embodiments of this application is illustrated in the model architecture diagram of the deep mining model involved in some embodiments;

[0052] Figure 8 The method for generating insurance agent team management strategies provided in the embodiments of this application includes a schematic diagram of the model architecture of the prediction and early warning model involved in some embodiments;

[0053] Figure 9 A schematic diagram of the structure of the device for generating insurance agent team management strategies provided in the embodiments of this application;

[0054] Figure 10 This is a schematic diagram of the hardware structure of a computer device provided in an embodiment of this application. Detailed Implementation

[0055] To make the objectives, technical solutions, and advantages of this application clearer, the following detailed description is provided in conjunction with the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are merely illustrative and not intended to limit the scope of this application.

[0056] It should be noted that although functional modules are divided in the device schematic diagram and a logical order is shown in the flowchart, in some cases, the steps shown or described may be performed in a different order than the module division in the device or the order in the flowchart. The terms "first," "second," etc., in the specification, claims, and the aforementioned drawings are used to distinguish similar objects and are not necessarily used to describe a specific order or sequence.

[0057] Unless otherwise defined, all technical and scientific terms used herein have the same meaning as commonly understood by one of ordinary skill in the art to which this application belongs. The terminology used herein is for the purpose of describing embodiments of this application only and is not intended to limit this application.

[0058] First, let's analyze some of the terms used in the embodiments of this application:

[0059] Artificial intelligence (AI) is a new technical science that studies and develops theories, methods, technologies, and application systems for simulating, extending, and expanding human intelligence. AI is a branch of computer science that attempts to understand the essence of intelligence and produce intelligent machines that can react in a way similar to human intelligence. Research in this field includes robotics, speech recognition, image recognition, natural language processing, and expert systems. AI can simulate the information processes of human consciousness and thought. AI also utilizes digital computers or machines controlled by digital computers to simulate, extend, and expand human intelligence, perceiving the environment, acquiring knowledge, and using that knowledge to achieve optimal results. Basic AI technologies generally include sensors, dedicated AI chips, cloud computing, distributed storage, big data processing technology, operating / interactive systems, and mechatronics. AI software technologies mainly include computer vision, robotics, biometrics, speech processing, natural language processing, and machine learning / deep learning.

[0060] Natural Language Processing (NLP): NLP uses computers to process, understand, and utilize human language (such as Chinese and English). NLP is a branch of artificial intelligence and an interdisciplinary field of computer science and linguistics, often referred to as computational linguistics. NLP includes syntactic analysis, semantic analysis, and discourse understanding. It is commonly used in machine translation, handwritten and printed character recognition, speech recognition and text-to-speech conversion, intent recognition, information extraction and filtering, text classification and clustering, sentiment analysis, and opinion mining. It involves data mining, machine learning, knowledge acquisition, knowledge engineering, artificial intelligence research, and linguistic research related to language computation.

[0061] Information extraction is a text processing technique that extracts factual information such as entities, relationships, and events from natural language text and outputs it as structured data. Information extraction is a technique for extracting specific information from text data. Text data is composed of specific units, such as sentences, paragraphs, and chapters. Text information is composed of smaller, specific units, such as characters, words, phrases, sentences, paragraphs, or combinations of these units. Extracting noun phrases, names of people, and place names from text data is an example of text information extraction. Of course, text information extraction techniques can extract information of various types.

[0062] K-means clustering: A classic unsupervised learning algorithm used to divide a dataset into K distinct clusters (categories), such that data points within the same cluster have high similarity, while data points between different clusters have significant differences. Its core idea is to iteratively optimize and find the center (called the "centroid") of each cluster, then assign each data point to the cluster containing the nearest centroid.

[0063] The Apriori algorithm for association rule mining is a classic frequent itemset mining algorithm used to discover frequently occurring item combinations (i.e., frequent itemsets) in transactional databases and generate association rules from them. The Apriori algorithm is a landmark algorithm in the field of association rule mining and is widely used in market analysis, recommender systems, bioinformatics, and other fields.

[0064] Random Forest: An ensemble learning algorithm that builds a powerful classification or regression model by combining multiple decision trees. Random Forest combines the ideas of bagging and random feature selection, effectively reducing overfitting and improving the model's generalization ability and stability.

[0065] Logistic Regression: A statistical learning method widely used in classification problems. Despite its name containing "regression," it is actually a classification algorithm. The core idea of ​​logistic regression is to map the output of linear regression to probability values ​​(between 0 and 1) using the logistic function (Sigmoid function), thereby predicting the probability that a sample belongs to a certain class.

[0066] The Cox Proportional Hazards Model (CPH) is a statistical model used for survival analysis. Its core function is to analyze the factors influencing the time of event occurrence (survival time) and assess the extent of these factors' impact on the risk of event occurrence.

[0067] Next, the overall concept of the embodiments of this application will be briefly described.

[0068] Competition in the insurance industry is becoming increasingly fierce, and customer needs are becoming more diverse. The importance of the insurance customer service team, as a key link between the company and its customers, is self-evident. The insurance customer service team is not only responsible for answering customer inquiries and handling claims, but also bears the heavy responsibility of sales conversion, directly impacting company performance.

[0069] In related technologies, the management of insurance agent teams still relies on traditional models. This lack of timely understanding of team structure, personnel characteristics, skill distribution, and potential problems leads to a series of issues. For example, some teams have structural deficiencies, with an imbalance in the ratio of new to experienced employees, resulting in poor business transitions. Furthermore, the varying skill levels of employees in some teams make it difficult to provide accurate and efficient customer service.

[0070] Therefore, how to improve team management methods to enhance the overall business level of insurance agent teams has become a pressing technical problem in the field.

[0071] Based on this, embodiments of this application provide a method for generating insurance agent team management strategies, an apparatus for generating insurance agent team management strategies, a computer device, a computer-readable storage medium, and a computer program product, aiming to intelligently generate management strategies for insurance agent teams based on the team profiles of insurance agent teams.

[0072] Considering that in order to optimize team management and improve the work efficiency and customer satisfaction of insurance agent teams, team managers need to understand the team's structure, personnel characteristics, ability distribution and potential problems, this application introduces a data-driven method of team profiling to help managers gain a deeper understanding of the team's situation, thereby providing a basis for management strategies, training resource investment and performance improvement plans.

[0073] This application embodiment obtains a team profile of the insurance agent team; constructs multi-dimensional features of the insurance agent team based on multi-source data corresponding to the team profile; inputs the multi-dimensional features into a preset multi-level analysis model to obtain the model analysis results of the multi-level analysis model on the multi-dimensional features; and generates a management strategy for the insurance agent team based on the model analysis results.

[0074] Compared to traditional management methods, this application's embodiments utilize team profiling of insurance agent teams, then construct multi-dimensional characteristics of the insurance agent teams based on multi-source data corresponding to these profilings. These multi-layered analytical models are then used to analyze these multi-dimensional characteristics, yielding model analysis results. Based on these results, management strategies for the insurance agent teams are generated. Thus, by employing a data-driven approach using team profiling to improve team management, and by intelligently generating management strategies based on these profilings, managers can optimize team management with a deep understanding of the team's situation, thereby improving the overall business performance, work efficiency, and customer satisfaction of the insurance agent teams.

[0075] Next, specific embodiments of the method for generating insurance agent team management strategies, the apparatus for generating insurance agent team management strategies, the computer equipment, and the computer storage medium provided in the embodiments of this application will be described, and the method for generating insurance agent team management strategies in the embodiments of this application will be described first.

[0076] It should be noted that the embodiments of this application can acquire and process relevant data based on artificial intelligence (AI) technology.

[0077] It should be noted that in all specific embodiments of this application, when processing data related to user identity or characteristics, such as user information, user behavior data, user historical data, and user location information, user permission or consent is obtained first. Furthermore, the collection, use, and processing of this data comply with relevant laws, regulations, and standards. In addition, when embodiments of this application require access to sensitive personal information of users, separate permission or consent from the user is obtained through pop-ups or redirection to confirmation pages. Only after obtaining the user's separate permission or consent is the necessary user-related data required for the proper functioning of these embodiments acquired.

[0078] The method for generating insurance agent team management strategies provided in this application relates to the field of data statistics and analysis processing technology. This method can be applied to a terminal, a server, or software running on either a terminal or a server. In some embodiments, the terminal can be a platform device used by an insurance service institution to provide insurance business services, such as a smartphone, tablet, laptop, or desktop computer; the server can be configured as an independent physical server, a server cluster or distributed system composed of multiple physical servers, or a cloud server providing basic cloud computing services such as cloud services, cloud databases, cloud computing, cloud functions, cloud storage, network services, cloud communication, middleware services, domain name services, security services, CDN, and big data and artificial intelligence platforms; the software can be an application implementing the method for generating insurance agent team management strategies, but is not limited to the above forms.

[0079] The embodiments of this application can also be used in numerous general-purpose or special-purpose computer system environments or configurations. Examples include: personal computers, server computers, handheld or portable devices, tablet devices, multiprocessor systems, microprocessor-based systems, set-top boxes, programmable consumer computer devices, network PCs, minicomputers, mainframe computers, distributed computing environments including any of the above systems or devices, etc. This application can be described in the general context of computer-executable instructions executed by a computer, such as program modules. Generally, program modules include routines, programs, objects, components, data structures, etc., that perform specific tasks or implement specific abstract data types. This application can also be practiced in distributed computing environments where tasks are performed by remote processing devices connected via a communication network. In a distributed computing environment, program modules can reside in local and remote computer storage media, including storage devices.

[0080] For ease of understanding and explanation, the following description will use the method for generating insurance agent team management strategies provided in the embodiments of this application on a terminal device as an example. The implementation of the method for generating insurance agent team management strategies provided in the embodiments of this application by any of the aforementioned entities can refer to the process of generating insurance agent team management strategies on a terminal device as described below.

[0081] Please refer to Figure 1 , Figure 1 The flowchart illustrates the steps of the method for generating an insurance agent team management strategy provided in this application embodiment. It should be understood that, although... Figure 1The figure shows the execution order of some method steps, but based on different design needs of actual applications, the method for generating insurance agent team management strategies provided in this application embodiment can of course adopt a different execution order of method steps than that shown in the figure. That is, Figure 1 The order of the method steps shown does not constitute a limitation on the execution logic order of the method for generating insurance agent team management strategies provided in this application embodiment. Any other method based on... Figure 1 Reasonable changes to the sequence of steps shown should be included within the protection scope of the method for generating insurance agent team management strategies provided in the embodiments of this application.

[0082] like Figure 1 As shown, in some embodiments, the method for generating insurance agent team management strategies provided in this application may include, but is not limited to, steps S101 to S104 as shown below.

[0083] Step S101: Obtain the team profile of the insurance agent team.

[0084] It should be noted that team profiling is a comprehensive analytical framework. In this embodiment, the team profile of an insurance agent team can cover multiple core dimensions, such as team structure, capability distribution, performance, stability analysis, and collaboration effectiveness. This can help users (such as managers of insurance agent teams) gain a deeper understanding of the insurance agent team from multiple perspectives.

[0085] In the process of intelligently generating management strategies for insurance agent teams, the terminal device can respond to the intelligent generation operation of management strategies initiated by the user and obtain the team profile of the insurance agent team to which the current operation is directed.

[0086] In some embodiments, the terminal device can pre-build a team profile of the insurance agent group and store the team profile in a local database or an online database. Thus, when a user initiates an intelligent generation operation for management policies targeting the insurance agent team, the terminal device can directly retrieve the pre-built team profile from the local database or the online database.

[0087] In other embodiments, the terminal device can also acquire relevant data in real time to construct a team profile of the insurance agent team when the user initiates an intelligent generation operation for management strategies for the insurance agent team.

[0088] In some embodiments, the terminal device can also obtain the team profile of the insurance agent team from other terminal devices through communication connections, when the user initiates an intelligent generation operation for management strategies targeting the insurance agent team. In this case, the other terminal devices can pre-build the team profile of the insurance agent team and store it in a local database or an online database; alternatively, the other terminal devices can also obtain relevant data in real time to build the team profile of the insurance agent team.

[0089] Step S102: Construct multi-dimensional features of the insurance agent team based on the multi-source data corresponding to the team profile.

[0090] It should be noted that the multi-source data corresponding to the team profile can be data collected from multiple data sources when constructing the team profile. Multi-source data can include basic information of each member of the insurance agent team, such as length of service, job level, education, and age, etc., as well as the overall business data, training data, performance data, and behavioral data of the insurance agent team. Among them, business data can be such as call recordings, work order processing records, etc.; training data can be such as course completion rate, simulation exercise score, etc.; performance data can be such as key performance indicator (KPI) achievement rate, quality inspection deductions, etc.; and behavioral data can be such as knowledge base contributions, cross-team communication frequency, etc.

[0091] After obtaining the team profile of the insurance agent team, the terminal device further constructs the multi-dimensional characteristics of the insurance agent team based on the multi-source data corresponding to the team profile.

[0092] In some embodiments, the terminal device can directly obtain multi-source data corresponding to the insurance agent team's profile by parsing the team profile. For example, the terminal device can obtain the multi-source data corresponding to the team profile by performing information extraction processing on the team profile.

[0093] In other embodiments, the terminal device can also re-collect multi-source data corresponding to the team profile from multiple data sources based on the records when the team profile of the insurance agent team was pre-built.

[0094] Step S103: Input the multidimensional features into a preset multi-level analysis model to obtain the model analysis results of the multi-level analysis model on the multidimensional features.

[0095] It should be noted that the preset multi-layer analysis model can be pre-built and trained for the terminal device and stored in a local database or online database. Alternatively, the multi-layer analysis model can also be built and trained in real time for the terminal device after it has constructed the multi-dimensional features of the insurance agent team based on the multi-source data corresponding to the team profile.

[0096] After constructing multidimensional features of the insurance agent team based on multi-source data corresponding to the team profile, the terminal device further inputs these multidimensional features into a multi-layer analysis model. The multi-layer analysis model then analyzes and processes these multidimensional features to obtain the model analysis results output by the multi-layer analysis model for these multidimensional features.

[0097] In some embodiments, the terminal device can simultaneously perform multi-dimensional feature analysis at multiple levels using a multi-layer analysis model. These multiple levels of analysis may include basic analysis, in-depth analysis, and short-term and medium-to-long-term early warning analysis.

[0098] In other embodiments, the terminal device may also perform multi-dimensional feature analysis at multiple levels sequentially through a multi-layer analysis model.

[0099] Step S104: Generate the management strategy for the insurance agent team based on the analysis results of the model.

[0100] After obtaining the model analysis results of the multi-dimensional features from the multi-layer analysis model, the terminal device can perform a table lookup operation based on the model analysis results to obtain the management strategy corresponding to the model analysis results in the preset management strategy table. Then, the management strategy is used as the current intelligent generation operation for the management strategy initiated by the user. The management strategy of the insurance agent team is intelligently generated based on the team profile of the insurance agent group.

[0101] In some embodiments, the terminal device can perform a table lookup operation based on the model that analyzes multidimensional features in the multi-layer analysis model adopted, and perform model analysis results based on the output of multidimensional features analysis based on each type of model to obtain the management strategy corresponding to the model analysis result. All management strategies obtained from the table lookup are used as the current management strategies for the insurance agent team intelligently generated based on the team profile in response to user operations.

[0102] For example, the model type, output format, management application scenario, and decision triggering mechanism of the multi-layer analysis model adopted by the terminal device can be shown in Table 1 below.

[0103]

[0104] Table 1

[0105] Among them, when the terminal device uses the clustering model in the multi-layer analysis model to analyze the multi-dimensional characteristics of the insurance agent team and outputs the model analysis results, and performs a table lookup operation to determine the management strategy corresponding to the model analysis results, the terminal device determines the management strategy and corresponding implementation case from Table 2 below based on the different types of the model analysis results, as one of the current management strategies for intelligently generating insurance agent teams based on team profiles in response to user operations.

[0106]

[0107] Table 2

[0108] In this embodiment, a data analysis-based management tool called "team profiling" can characterize the features of insurance agent team members from multiple dimensions, thus providing a scientific basis for personalized management. During the intelligent generation of management strategies for the insurance agent team via a terminal device, in response to a user-initiated intelligent management strategy generation operation, a team profile of the insurance agent team is obtained. Then, based on the multi-source data corresponding to this team profile, multi-dimensional features of the insurance agent team are constructed. Next, the terminal device inputs these multi-dimensional features into a multi-layer analysis model, which analyzes and processes them to obtain the model analysis results. Based on these results, a lookup operation is performed to obtain the management strategy corresponding to the model analysis results in a preset management strategy table. This management strategy is then used as the management strategy intelligently generated for the insurance agent team based on the team profile, in response to the user-initiated intelligent management strategy generation operation.

[0109] Compared to traditional management methods, this application's embodiments utilize team profiling of insurance agent teams, then construct multi-dimensional characteristics of the insurance agent teams based on multi-source data corresponding to these profilings. These multi-layered analytical models are then used to analyze these multi-dimensional characteristics, yielding model analysis results. Based on these results, management strategies for the insurance agent teams are generated. Thus, by employing a data-driven approach using team profiling to improve team management, and by intelligently generating management strategies based on these profilings, managers can optimize team management with a deep understanding of the team's situation, thereby improving the overall business performance, work efficiency, and customer satisfaction of the insurance agent teams.

[0110] Please refer to Figure 2 , Figure 2 for Figure 1 A detailed flowchart of step S101.

[0111] like Figure 2As shown, in some embodiments, step S101 above: obtaining the team profile of the insurance agent team may include steps S201 and S202 as described below.

[0112] Step S201: Collect various types of data from the insurance agent team from a preset data source matrix; the data source matrix includes multiple data source systems.

[0113] When a terminal device responds to a user-initiated intelligent management strategy generation operation and obtains the team profile of the insurance agent team targeted by the current operation, if the terminal device acquires relevant data in real time to construct the team profile, it can collect various types of data about the insurance agent team from multiple data source systems included in a preset data source matrix.

[0114] For example, multiple data source systems may include Human Resource (HR) systems, Customer Relationship Management (CRM) systems, Learning Management System (LMS) learning platforms, performance management systems, and internal collaboration platforms. Furthermore, different data source systems can be configured with different data collection frequencies. In addition, the various types of data may include basic information, business processes, training data, performance data, and behavioral data. In this case, the various types of data collected by the terminal device from multiple data source systems can be shown in Table 3 below.

[0115]

[0116]

[0117] Table 3

[0118] Step S202: Construct a multi-dimensional team profile of the insurance agent team based on the collected multi-type data.

[0119] It should be noted that the team profile dimensions described above include multiple main dimensions, which include at least two of the following: team structure, capability distribution, performance, stability analysis, and collaboration effectiveness. Each main dimension includes at least two sub-dimensions. For example, the team profile dimension design can be shown in Table 4 below.

[0120]

[0121] Table 4

[0122] When the terminal device acquires relevant data in real time to build a team profile of the insurance agent team, it can first perform data cleaning on the collected multi-type data, and then, based on the cleaned multi-type data, design and build a multi-dimensional team profile of the insurance agent team according to the above-mentioned profile dimensions.

[0123] In this embodiment, a team profile of the insurance agent team is constructed through a terminal device. Then, based on this team profile, multi-dimensional characteristics of the insurance agent team are created, and these multi-layered analytical models are used to analyze these characteristics. Based on the model analysis results, management strategies for the insurance agent team are generated. In this way, based on the team profile of the insurance agent team, users can better understand the team structure, personnel characteristics, capability distribution, and potential problems, thereby optimizing management strategies, training resource investment, and performance improvement plans.

[0124] Please refer to Figure 3 , Figure 3 for Figure 1 A detailed flowchart of step S102.

[0125] like Figure 3 As shown, in some embodiments, step S102 above: constructing multi-dimensional features of the insurance agent team based on the multi-source data corresponding to the team profile, may include steps S301 and S302 as shown below.

[0126] Step S301: Clean the multi-source data according to the preset data cleaning specifications to obtain cleaned multi-source data.

[0127] It should be noted that the preset data cleaning specifications can include missing value handling, outlier detection, format standardization, and data anonymization. Missing value handling can be achieved by filling continuous variables with the median and categorical variables with the mode; outlier detection can be performed by reviewing extreme values ​​such as call duration > 12 hours / day and conversion rate > 200%; format standardization can be achieved by standardizing date format to YYYY-MM-DD and monetary unit to ten thousand yuan; and data anonymization can be achieved by encrypting fields such as ID card numbers and contact information.

[0128] When the terminal device constructs the multi-dimensional features of the insurance agent team based on the multi-source data corresponding to the team profile of the insurance agent team, if the multi-source data is multi-type data collected by the terminal device from multiple data sources to construct the team profile of the insurance agent team, the terminal device will clean the multi-source data according to the preset data cleaning specifications to obtain the cleaned multi-source data.

[0129] In some embodiments, if the multi-source data is data obtained by the terminal device from parsing the team profile, since the terminal device has already cleaned the collected data when constructing the team profile, the terminal device may not need to perform cleaning processing on the data directly obtained from parsing the team profile, but may directly use it as cleaned multi-source data for subsequent processing.

[0130] Step S302: Construct multidimensional structured features of the insurance agent team based on the cleaned multi-source data; and construct multidimensional derived features of the insurance agent team based on the cleaned multi-source data; the multidimensional features include the multidimensional structured features and the multidimensional derived features.

[0131] After obtaining the cleaned multi-source data, the terminal device can further construct multi-dimensional features from this data to build the multi-dimensional features of the insurance agent team. This involves standardizing the cleaned multi-source data to construct the multi-dimensional structured features of the insurance agent team. For example, the terminal device can extract feature parameters from the structured feature library shown in Table 5 below from the multi-source data to construct the multi-dimensional structured features of the insurance agent team.

[0132]

[0133]

[0134] Table 5

[0135] Furthermore, when constructing multi-dimensional features, the terminal device can also build derived features based on cleaned multi-source data, thereby constructing multi-dimensional derived features for the insurance agent team. For example, the terminal device can subtract the lowest skill score from the highest skill score of the insurance agent team in the cleaned multi-source data, and then divide by the average score to obtain the team's ability balance index. As another example, the terminal device can also construct the team resilience coefficient and performance volatility of the insurance agent team based on the cleaned multi-source data. The team resilience coefficient = (1 - turnover rate) * average years of service^0.5, and the performance volatility = standard deviation of KPIs over the past 6 months / mean.

[0136] After the terminal device constructs multidimensional structured features and / or multidimensional derived features of the insurance agent team based on the cleaned multi-source data, the terminal device can use the constructed features as the multidimensional features of the insurance agent team constructed based on the multi-source features corresponding to the team profile, for subsequent model analysis and processing.

[0137] In some embodiments, the aforementioned preset multi-layer analysis model may include at least a basic analysis model, a deep mining model, and a prediction and early warning model. In this case, the terminal device can perform the step of "inputting the multi-dimensional features into the preset multi-layer analysis model" in step S103 as shown in method 1 or method 2 below.

[0138] Method 1: Input the multidimensional features into the basic analysis model, the deep mining model, and the prediction and early warning model simultaneously.

[0139] Please refer to Figure 4 , Figure 4 The method for generating insurance agent team management strategies provided in the embodiments of this application is illustrated in some embodiments with a four-layer analysis model architecture.

[0140] like Figure 4 As shown, the terminal device can obtain a team profile of the insurance agent team at the data layer. Based on the multi-source data corresponding to this team profile, it constructs multi-dimensional features of the insurance agent team. These multi-dimensional features are then simultaneously input into the basic analysis model, the deep mining model, and the predictive early warning model. The basic analysis model, the deep mining model, and the predictive early warning model then analyze the multi-dimensional features in parallel and output corresponding model analysis results (such as team type, key influencing factors, and risk warning alerts). The terminal device then generates a management strategy for the insurance agent team based on these model analysis results.

[0141] Method 2: Input the multidimensional features sequentially into the basic analysis model, the deep mining model, and the prediction and early warning model.

[0142] The terminal device can also construct multi-dimensional characteristics of the insurance agent team, and then input these characteristics into a basic analysis model, a deep mining model, and a prediction and early warning model in a certain order. For example, the multi-dimensional characteristics are first input into the basic analysis model, and after the basic analysis model analyzes the multi-dimensional characteristics and outputs the model analysis results (such as team type), the multi-dimensional characteristics are then input into the deep mining model, and after the deep mining model analyzes the multi-dimensional characteristics and outputs the model analysis results (such as key influencing factors), the multi-dimensional characteristics are further input into the prediction and early warning model, which analyzes the multi-dimensional characteristics and outputs the corresponding model analysis results (such as risk warning prompts).

[0143] Please refer to Figure 5 , Figure 5 for Figure 1 A detailed flowchart of step S103.

[0144] like Figure 5As shown, in some embodiments, step S103 above: inputting the multidimensional features into a preset multilayer analysis model may include steps S501 to S503 as described below.

[0145] Step S501: Generate a standardized multidimensional feature matrix based on the first part of the multidimensional features, and input the multidimensional feature matrix into the basic analysis model.

[0146] It should be noted that the first part of the multidimensional features refers to some or all of the multidimensional features of the insurance agent team that require cluster analysis by the basic analysis model to group the insurance agent team.

[0147] When the terminal device performs model analysis on the multidimensional characteristics of the insurance agent team based on the multi-layer analysis model, regardless of the method used to input the multidimensional characteristics into the multi-layer analysis model, when inputting the multidimensional characteristics into the basic analysis model in the multi-layer analysis model, the terminal device first generates a standardized multidimensional feature matrix based on the first part of the multidimensional characteristics, and then inputs the multidimensional feature matrix into the basic analysis model.

[0148] Step S502: Generate a discretized feature combination based on the second part of the multidimensional features, and input the feature combination into the deep mining model.

[0149] It should be noted that the second part of the multidimensional features refers to some or all of the multidimensional features of the insurance agent team that require in-depth mining of key factors by the model.

[0150] When the terminal device performs model analysis on the multidimensional characteristics of the insurance agent team based on the multi-layer analysis model, regardless of the method used to input the multidimensional characteristics into the multi-layer analysis model, when inputting the multidimensional characteristics into the deep mining model in the multi-layer analysis model, the terminal device first generates a discretized feature combination based on the second part of the multidimensional characteristics, and then inputs the feature combination into the deep mining model.

[0151] Step S503: Generate a time-series feature matrix based on the third part of the multidimensional features, and input the time-series feature matrix into the prediction and early warning model.

[0152] It should be noted that the third part of the multidimensional features refers to some or all of the multidimensional features of the insurance agent team that need to be used for logistic regression analysis and survival analysis in the prediction and early warning model.

[0153] When the terminal device performs model analysis on the multidimensional characteristics of the insurance agent team based on the multi-layer analysis model, regardless of the method used to input the multidimensional characteristics into the multi-layer analysis model, when inputting the multidimensional characteristics into the prediction and early warning model in the multi-layer analysis model, the terminal device first generates a time series feature matrix based on the third part of the multidimensional characteristics, and then inputs the time series feature matrix into the prediction and early warning model.

[0154] In some embodiments, after step S501 described above, the method for generating an insurance agent team management strategy provided in this application embodiment may further include the following steps:

[0155] The multidimensional feature matrix is ​​subjected to cluster analysis using the basic analysis model to divide the insurance agent team into multiple groups.

[0156] Obtain the team classification labels corresponding to each of the multiple groups output by the basic analysis model; the model analysis results include multiple team classification labels.

[0157] Please refer to Figure 6 , Figure 6 The method for generating insurance agent team management strategies provided in the embodiments of this application is illustrated in the model architecture diagram of the basic analysis model involved in some embodiments.

[0158] like Figure 6 As shown, after inputting the standardized multidimensional feature matrix into the basic analysis model, the terminal device uses the K-means clustering algorithm to divide the insurance agent team into different groups (such as high-potential group, stable group, and group to be promoted). It then obtains the team classification labels corresponding to each of these groups output by the basic analysis model. This allows for the subsequent generation of different management strategies through table lookups to implement differentiated management strategies for different insurance agent teams. In this case, the multiple team classification labels output by the basic analysis model represent the model analysis results obtained by the terminal device through multi-layer analysis of multidimensional features.

[0159] In some embodiments, when the terminal device analyzes the standardized multidimensional feature matrix using a fundamental analysis model, the fundamental analysis model initializes the multidimensional feature matrix by randomly selecting k data points as initial cluster centers. Then, the fundamental analysis model allocates the multidimensional feature matrix, such as assigning each data point to the nearest cluster center to form k clusters. Next, the fundamental analysis model updates the multidimensional feature matrix by calculating the mean of each cluster as the new cluster center. The fundamental analysis model repeats the above allocation and update steps until the cluster centers no longer change or the maximum number of iterations is reached. For example, the fundamental analysis model can use the following formula 1 to perform the above update steps for the multidimensional feature matrix:

[0160]

[0161] Here, Ci is the i-th cluster, μi is the centroid of the i-th cluster, and k is the preset number of clusters (determined by the elbow rule). Choosing k=4 corresponds to high-performing teams, potential teams, risky teams, and structurally unbalanced teams.

[0162] The centroid update formula can be as follows:

[0163]

[0164] In addition, the basic analysis model outputs team classification labels (1-4 categories), corresponding to 1: high-performing teams, 2: potential teams, 3: risky teams, and 4: structurally unbalanced teams.

[0165] In some embodiments, after step S502 described above, the method for generating an insurance agent team management strategy provided in this application embodiment may further include the following steps:

[0166] The feature combination is subjected to frequent itemset mining processing by the deep mining model to obtain the association rules corresponding to the feature combination.

[0167] The feature combination is sorted by feature importance using the deep mining model to obtain the key influencing factors in the feature combination;

[0168] Obtain the association rules and key influencing factors output by the deep mining model; the model analysis results include the association rules and key influencing factors.

[0169] Please refer to Figure 7 , Figure 7 The method for generating insurance agent team management strategies provided in the embodiments of this application is illustrated in the model architecture diagram of the deep mining model involved in some embodiments.

[0170] like Figure 7 As shown, after the terminal device inputs the discretized feature combination into the deep mining model, the model uses the Apriori algorithm to discover implicit correlation patterns between features (such as "highly educated team + high-frequency training → high renewal rate"). This supports multi-dimensional combination analysis. Furthermore, the deep mining model automatically assesses feature importance based on random forests, identifies core influencing factors, and handles high-dimensional data and non-linear relationships. The combination of association rules and random forests in the deep mining model can meet the multi-dimensional (structure, capability, performance) and multi-interaction feature analysis needs of insurance agent teams.

[0171] In some embodiments, when a terminal device performs frequent itemset mining on feature combinations using a deep mining model, the deep mining model can first generate frequent itemsets, that is, filter high-frequency feature combinations by using a minimum support threshold (e.g., 5%), and then generate association rules: calculate confidence and lift, and filter effective rules. Here, the deep mining model can use Formula 2 to calculate support, Formula 3 to calculate confidence, and Formula 4 to calculate lift.

[0172]

[0173]

[0174]

[0175] In some embodiments, the association rules output by the deep mining model can be as shown in Table 6 below.

[0176]

[0177]

[0178] Table 6

[0179] Among them, the antecedent / consequence refers to: the combination of features and the target event (supporting multiple feature combinations, such as {A,B}→{C}); support refers to: the proportion of teams covered by the rule (such as 11.2% of teams meeting the antecedent); lift refers to: the rule effectiveness index, >1 indicates a positive association.

[0180] In some embodiments, when a terminal device performs feature importance ranking on feature combinations using a deep learning model, it can perform random forest modeling using steps 1 to 3 as shown below. That is:

[0181] Step 1: Define the target variable: such as team renewal rate (categorized as high / low) or churn probability (regression).

[0182] Step 2: Train the model: Generate multiple decision trees by splitting nodes based on Gini impurity.

[0183] Step 3: Calculate feature importance based on the feature's contribution to splitting within the tree.

[0184] Among these, the formula for feature importance can be:

[0185] GiniImportance f =∑ t∈T ΔGini(t,f).

[0186] Where T represents the set of all trees, and ΔGini(t,f) is the reduction in impurity of feature f in tree t. In some embodiments, the output of the random forest model can be shown in Table 7 below.

[0187]

[0188] Table 7

[0189] The importance score is a standardized value of 0-1, with a total of 1. The larger the value, the more significant the impact. An upward trend in the direction of influence indicates that the feature is positively correlated with the target (e.g., the higher the knowledge score, the higher the renewal rate), while a downward trend indicates that the feature is negatively correlated with the target.

[0190] In some embodiments, the terminal device can mine key factors through a deep mining model by using a random forest to find the top 3 features that affect the renewal rate.

[0191] In some embodiments, after step S503 described above, the method for generating an insurance agent team management strategy provided in this application embodiment may further include the following steps:

[0192] The time-series feature matrix is ​​subjected to binary classification prediction processing by the prediction and early warning model to obtain short-term early warning prompts for the insurance agent team.

[0193] The time-series feature matrix is ​​subjected to survival analysis by the prediction and early warning model to obtain the medium- and long-term early warning prompts for the insurance agent team.

[0194] Obtain the short-term and medium-to-long-term early warning prompts output by the prediction and early warning model; the model analysis results include the short-term and medium-to-long-term early warning prompts.

[0195] Please refer to Figure 8 , Figure 8 The method for generating insurance agent team management strategies provided in this application provides a schematic diagram of the model architecture of the prediction and early warning model involved in some embodiments.

[0196] like Figure 8As shown, after the terminal device inputs the time-series feature matrix into the prediction and early warning model, the model uses logistic regression to perform binary classification prediction on the matrix, directly outputting the probability of event occurrence (e.g., the probability of leaving the company in the next 3 months) as a short-term early warning for the insurance agent team. This is suitable for short-term, specific prediction scenarios. Furthermore, the prediction and early warning model uses a Cox proportional hazards model to model survival time, performing survival analysis on the time-series feature matrix to analyze the pattern of employee turnover risk over time (e.g., the peak turnover period is from the 6th to 12th month after joining the company), thus obtaining medium- to long-term early warnings for the insurance agent team. In this way, the terminal device combines short-term and medium- to long-term early warnings through the prediction and early warning model, achieving complementarity in the time dimension. That is, logistic regression focuses on the short term (0-3 months), while the Cox model covers the medium to long term (3-12 months).

[0197] In some embodiments, the steps for a terminal device to generate a short-term early warning based on logistic regression modeling using a predictive early warning model can be as follows:

[0198] Target variable definition: y = 1 (leaving the company within the next 3 months), y = 0 (not leaving the company).

[0199] Feature standardization: Z-score standardization is applied to continuous variables.

[0200] Regularization: L1 regularization (Lasso) is used to prevent overfitting.

[0201] Model training: Maximize the log-likelihood function.

[0202] Among these, the algorithm formulas used in the prediction and early warning model include: Formula 5 for the output probability and Formula 6 for the cross-entropy loss function.

[0203]

[0204] Where β0: intercept term (bias term), representing the baseline log-odds when all features are zero. βj: regression coefficient of the j-th feature, representing the marginal contribution of that feature to the log-odds.

[0205]

[0206] Where N is the number of samples, and yi is the true label of the i-th sample (yi = 1 indicates resignation, yi = 0 indicates not resignation).

[0207] In some embodiments, the steps for a terminal device to perform medium- to long-term early warning based on the Cox proportional hazards model using a predictive early warning model can be as follows:

[0208] Verify the proportional hazards hypothesis: use the Schoenfeld residual test (if the p-value is >0.05, the hypothesis is true).

[0209] Model fitting: Maximize the partial likelihood function.

[0210] Baseline risk estimation: Calculate the baseline survival function.

[0211] Among these, the algorithm formulas used in the prediction and early warning model include: risk function formula 7 and partial likelihood function formula 8.

[0212] h(t|x)=h0(t)exp(β1x1+...+β p x p ), Formula 7.

[0213] Where h0(t) is the baseline risk function, representing the baseline risk over time when all features are at zero (or the reference level); βj is the regression coefficient of the j-th feature, reflecting the amplification / reduction effect of the feature on the risk; and xj is the value of the j-th feature.

[0214]

[0215] Where k is the number of time points when the resignation event occurs; t(i) is the time when the i-th event occurs (sorted by time); x(i) is the feature vector of the individual whose event occurs at time t(i); R(t(i)) is the risk set, that is, the set of individuals who are still alive (not resigned) at time t(i); β'x is a linear combination of feature vectors (β' is the transpose of the coefficient vector, and x is the feature vector).

[0216] In some embodiments, sample data of the results output by the logistic regression model can be shown in Table 8 below.

[0217]

[0218] Table 8

[0219] The predicted probability of leaving the company is between 0 and 1, and the threshold can be dynamically adjusted (e.g., >0.7 is considered high risk). The key risk factors are the top 3 features that contribute the most to the prediction.

[0220] In some embodiments, sample data of the results output by the Cox model can be shown in Table 9 below.

[0221]

[0222]

[0223] Table 9

[0224] Among them, the risk ratio (HR) is HR>1, which means that the risk is higher than the average level (e.g., HR=2.3 means that the risk of employee turnover is 2.3 times the average), and the retention probability is the probability of being employed in a specific future period (e.g., 6 months).

[0225] Please see Figure 9 This application also provides an apparatus for generating an insurance agent team management strategy, which can implement the above-mentioned method for generating an insurance agent team management strategy. The apparatus includes:

[0226] The acquisition module is used to obtain team profiles of insurance agent teams;

[0227] The feature construction module is used to construct multi-dimensional features of the insurance agent team based on the multi-source data corresponding to the team profile;

[0228] The model analysis module is used to input the multidimensional features into a preset multi-level analysis model to obtain the model analysis results of the multi-level analysis model on the multidimensional features;

[0229] The strategy generation module is used to generate management strategies for the insurance agent team based on the analysis results of the model.

[0230] In some embodiments, the multi-source data refers to the multi-type data of the insurance agent team collected from multiple data sources to construct the team profile; the feature construction module is further configured to clean the multi-source data according to a preset data cleaning specification to obtain cleaned multi-source data; construct multi-dimensional structured features of the insurance agent team based on the cleaned multi-source data; and construct multi-dimensional derived features of the insurance agent team based on the cleaned multi-source data; the multi-dimensional features include the multi-dimensional structured features and the multi-dimensional derived features.

[0231] In some embodiments, the multi-layer analysis model includes at least a basic analysis model, a deep mining model, and a prediction and early warning model; the model analysis module is further configured to generate a standardized multi-dimensional feature matrix based on a first part of the multi-dimensional features, and input the multi-dimensional feature matrix into the basic analysis model; generate a discretized feature combination based on a second part of the multi-dimensional features, and input the feature combination into the deep mining model; and generate a time-series feature matrix based on a third part of the multi-dimensional features, and input the time-series feature matrix into the prediction and early warning model.

[0232] In some embodiments, the model analysis module is further configured to perform cluster analysis on the multidimensional feature matrix using the basic analysis model to divide the insurance agent team into multiple groups; and to obtain the team classification labels corresponding to each of the multiple groups output by the basic analysis model; the model analysis results include multiple team classification labels.

[0233] In some embodiments, the model analysis module is further configured to perform frequent itemset mining on the feature combination using the deep mining model to obtain the association rules corresponding to the feature combination; perform feature importance ranking on the feature combination using the deep mining model to obtain the key influencing factors in the feature combination; and obtain the association rules and the key influencing factors output by the deep mining model; the model analysis results include the association rules and the key influencing factors.

[0234] In some embodiments, the model analysis module is further configured to perform binary classification prediction processing on the time-series feature matrix using the prediction and early warning model to obtain short-term early warning prompts for the insurance agent team; perform survival analysis processing on the time-series feature matrix using the prediction and early warning model to obtain medium- and long-term early warning prompts for the insurance agent team; and acquire the short-term early warning prompts and the medium- and long-term early warning prompts output by the prediction and early warning model; the model analysis results include the short-term early warning prompts and the medium- and long-term early warning prompts.

[0235] In some embodiments, the acquisition module is further configured to collect multi-type data of the insurance agent team from a preset data source matrix; the data source matrix includes multiple data source systems; and to construct a multi-dimensional team profile of the insurance agent team based on the collected multi-type data; wherein the profile dimensions of the team profile include multiple main dimensions, and the multiple main dimensions include at least two of the following: team structure dimension, capability distribution dimension, performance dimension, stability analysis dimension, and collaboration effectiveness dimension; each main dimension includes at least two sub-dimensions.

[0236] It should be noted that the specific implementation of the insurance agent team management strategy generation device provided in this application embodiment is basically the same as the specific implementation of the above-described insurance agent team management strategy generation method, and will not be repeated here.

[0237] This application also provides a computer device, which includes a memory and a processor. The memory stores a computer program, and the processor executes the computer program to implement the above-mentioned method for generating insurance agent team management strategies. This computer device can be any smart terminal, including tablet computers, in-vehicle computers, etc.

[0238] Please see Figure 10 , Figure 10 This illustration shows the hardware structure of a computer device according to one embodiment. The computer device includes:

[0239] The processor 1001 can be implemented using a general-purpose CPU (Central Processing Unit), microprocessor, application-specific integrated circuit (ASIC), or one or more integrated circuits, and is used to execute relevant programs to implement the technical solutions provided in the embodiments of this application.

[0240] The memory 1002 can be implemented as a read-only memory (ROM), static storage device, dynamic storage device, or random access memory (RAM). The memory 1002 can store the operating system and other applications. When the technical solutions provided in the embodiments of this specification are implemented through software or firmware, the relevant program code is stored in the memory 1002, and the processor 1001 calls and executes the method for generating the insurance agent team management strategy according to the embodiments of this application.

[0241] Input / output interface 1003 is used to implement information input and output;

[0242] The communication interface 1004 is used to enable communication and interaction between this device and other devices. Communication can be achieved through wired means (such as USB, network cable, etc.) or wireless means (such as mobile network, WIFI, Bluetooth, etc.).

[0243] Bus 1005 transmits information between various components of the device (e.g., processor 1001, memory 1002, input / output interface 1003, and communication interface 1004);

[0244] The processor 1001, memory 1002, input / output interface 1003 and communication interface 1004 are connected to each other within the device via bus 1005.

[0245] This application also provides a computer-readable storage medium storing a computer program that, when executed by a processor, implements the above-described method for generating insurance agent team management strategies.

[0246] Memory, as a non-transitory computer-readable storage medium, can be used to store non-transitory software programs and non-transitory computer-executable programs. Furthermore, memory may include high-speed random access memory, and may also include non-transitory memory, such as at least one disk storage device, flash memory device, or other non-transitory solid-state storage device. In some embodiments, memory may optionally include memory remotely located relative to the processor, and these remote memories can be connected to the processor via a network. Examples of such networks include, but are not limited to, the Internet, intranets, local area networks, mobile communication networks, and combinations thereof.

[0247] This application also provides a computer program product that stores a computer program, which, when executed by a processor, implements the above-described method for generating insurance agent team management strategies.

[0248] The embodiments described in this application are for the purpose of more clearly illustrating the technical solutions of this application, and do not constitute a limitation on the technical solutions provided in this application. As those skilled in the art will know, with the evolution of technology and the emergence of new application scenarios, the technical solutions provided in this application are also applicable to similar technical problems.

[0249] Those skilled in the art will understand that the technical solutions shown in the figures do not constitute a limitation on the embodiments of this application, and may include more or fewer steps than shown, or combine certain steps, or different steps.

[0250] The device embodiments described above are merely illustrative. The units described as separate components may or may not be physically separate; that is, they may be located in one place or distributed across multiple network units. Some or all of the modules can be selected to achieve the purpose of this embodiment according to actual needs.

[0251] Those skilled in the art will understand that all or some of the steps in the methods disclosed above, as well as the functional modules / units in the systems and devices, can be implemented as software, firmware, hardware, or suitable combinations thereof.

[0252] The terms “first,” “second,” “third,” “fourth,” etc. (if present) in the specification and accompanying drawings of this application 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 this application 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 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.

[0253] It should be understood that in this application, "at least one (item)" means one or more, and "more than" means two or more. "And / or" is used to describe the relationship between related objects, indicating that three relationships can exist. For example, "A and / or B" can represent three cases: only A exists, only B exists, and both A and B exist simultaneously, where A and B can be singular or plural. The character " / " generally indicates that the preceding and following related objects are in an "or" relationship. "At least one (item) of the following" or similar expressions refer to any combination of these items, including any combination of single or plural items. For example, at least one (item) of a, b, or c can represent: a, b, c, "a and b", "a and c", "b and c", or "a and b and c", where a, b, and c can be single or multiple.

[0254] In the several embodiments provided in this application, it should be understood that the disclosed apparatus and methods can be implemented in other ways. For example, the apparatus embodiments described above are merely illustrative; for instance, the division of the units described above is only a logical functional division, and in actual implementation, there may be other division methods. For example, multiple units or components may be combined or integrated into another system, or some features may be ignored or not executed. Furthermore, the coupling or direct coupling or communication connection shown or discussed may be through some interfaces; the indirect coupling or communication connection between apparatuses or units may be electrical, mechanical, or other forms.

[0255] The units described above as separate components may or may not be physically separate. The components shown as units may or may not be physical units; that is, they may be located in one place or distributed across multiple network units. Some or all of the units can be selected to achieve the purpose of this embodiment according to actual needs.

[0256] Furthermore, the functional units in the various embodiments of this application can be integrated into one processing unit, or each unit can exist physically separately, or two or more units can be integrated into one unit. The integrated unit can be implemented in hardware or as a software functional unit.

[0257] If the integrated unit is implemented as a software functional unit and sold or used as an independent product, it can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of this application, in essence, or the part that contributes to the prior art, or all or part of the technical solution, can be embodied in the form of a software product. This computer software product is stored in a storage medium and includes multiple instructions to cause a computer device (which may be a personal computer, server, or network device, etc.) to execute all or part of the steps of the methods of the various embodiments of this application. The aforementioned storage medium includes various media capable of storing programs, such as USB flash drives, portable hard drives, read-only memory (ROM), random access memory (RAM), magnetic disks, or optical disks.

[0258] The preferred embodiments of the present application have been described above with reference to the accompanying drawings, but this does not limit the scope of the claims of the present application. Any modifications, equivalent substitutions, and improvements made by those skilled in the art without departing from the scope and substance of the embodiments of the present application shall be within the scope of the claims of the present application.

Claims

1. A method for generating an insurance agent team management strategy, characterized in that, The method includes: Obtain a team profile of the insurance agent team; The multidimensional features of the insurance agent team are constructed based on the multi-source data corresponding to the team profile. The multidimensional features are input into a preset multi-level analysis model to obtain the model analysis results of the multi-level analysis model on the multidimensional features; The management strategy for the insurance agent team is generated based on the analysis results of the model.

2. The method according to claim 1, characterized in that, The multi-source data refers to the various types of data collected from multiple data sources to construct the team profile of the insurance agent team; The construction of multi-dimensional features of the insurance agent team based on the multi-source data corresponding to the team profile includes: The multi-source data is cleaned according to the preset data cleaning specifications to obtain cleaned multi-source data. The insurance agent team is constructed with multidimensional structured features based on the cleaned multi-source data; and multidimensional derived features are constructed with the insurance agent team based on the cleaned multi-source data; the multidimensional features include the multidimensional structured features and the multidimensional derived features.

3. The method according to claim 1, characterized in that, The multi-layered analysis model includes at least a basic analysis model, a deep mining model, and a prediction and early warning model; The step of inputting the multidimensional features into a preset multi-layer analysis model includes: A standardized multidimensional feature matrix is ​​generated based on the first part of the multidimensional features, and the multidimensional feature matrix is ​​input into the basic analysis model. Based on the second part of the multidimensional features, a discretized feature combination is generated, and the feature combination is input into the deep mining model. A time-series feature matrix is ​​generated based on the third part of the multidimensional features, and the time-series feature matrix is ​​input into the prediction and early warning model.

4. The method according to claim 3, characterized in that, After inputting the multidimensional feature matrix into the basic analysis model, the method further includes: The multidimensional feature matrix is ​​subjected to cluster analysis using the basic analysis model to divide the insurance agent team into multiple groups. Obtain the team classification labels corresponding to each of the multiple groups output by the basic analysis model; the model analysis results include multiple team classification labels.

5. The method according to claim 3, characterized in that, After inputting the feature combination into the deep mining model, the method further includes: The feature combination is subjected to frequent itemset mining processing by the deep mining model to obtain the association rules corresponding to the feature combination. The feature combination is sorted by feature importance using the deep mining model to obtain the key influencing factors in the feature combination; Obtain the association rules and key influencing factors output by the deep mining model; the model analysis results include the association rules and key influencing factors.

6. The method according to claim 3, characterized in that, After inputting the time-series feature matrix into the prediction and early warning model, the method further includes: The time-series feature matrix is ​​subjected to binary classification prediction processing by the prediction and early warning model to obtain short-term early warning prompts for the insurance agent team. The time-series feature matrix is ​​subjected to survival analysis by the prediction and early warning model to obtain the medium- and long-term early warning prompts for the insurance agent team. Obtain the short-term and medium-to-long-term early warning prompts output by the prediction and early warning model; the model analysis results include the short-term and medium-to-long-term early warning prompts.

7. The method according to any one of claims 1 to 6, characterized in that, The process of obtaining the team profile of the insurance agent team includes: The insurance agent team collects various types of data from a pre-defined data source matrix; the data source matrix includes multiple data source systems. Based on the collected multi-type data, a multi-dimensional team profile of the insurance agent team is constructed; The team profile includes multiple main dimensions, which include at least two of the following: team structure dimension, capability distribution dimension, performance dimension, stability analysis dimension, and collaboration effectiveness dimension; each main dimension includes at least two sub-dimensions.

8. A device for generating an insurance agent team management strategy, characterized in that, The device includes: The acquisition module is used to obtain team profiles of insurance agent teams; The feature construction module is used to construct multi-dimensional features of the insurance agent team based on the multi-source data corresponding to the team profile; The model analysis module is used to input the multidimensional features into a preset multi-level analysis model to obtain the model analysis results of the multi-level analysis model on the multidimensional features; The strategy generation module is used to generate management strategies for the insurance agent team based on the analysis results of the model.

9. A computer device, characterized in that, The computer device includes a memory and a processor, the memory storing a computer program, and the processor executing the computer program to implement the method for generating an insurance agent team management strategy as described in any one of claims 1 to 7.

10. A computer-readable storage medium storing a computer program, characterized in that, When the computer program is executed by a processor, it implements the method for generating an insurance agent team management strategy as described in any one of claims 1 to 7.

Images