A method, device, electronic device and medium for managing accident vehicle information

By monitoring or polling, the accident vehicle information is obtained, and the sorting and labeling process is carried out, the problem of high information dispersion and maintenance costs in traditional methods is solved, and the rapid and accurate access and automated management of accident vehicle information is achieved, thereby reducing costs.

CN118690265BActive Publication Date: 2024-11-15ZHEJIANG GEELY HLDG GRP CO LTD +2
View PDF 0 Cites 0 Cited by

Patent Information

Application Number
CN202411193871.X
Authority / Receiving Office
CN · China
Patent Type
Patents(China)
Current Assignee / Owner
Filing Date
2024-08-28
Publication Date
2024-11-15
Estimated Expiration
2044-08-28

AI Technical Summary

Technical Problem

The traditional accident car information management method between insurance companies and 4S stores has problems such as dispersed information, huge data volume, high maintenance costs and human errors, which makes it difficult to quickly and accurately push accident car information.

Method used

Accident car messages from multiple sources are obtained through monitoring or polling, sorting, cleaning and labeling, and generating clue data, and allocating the labels of clue data to the corresponding processing process based on the clue data to achieve automated management.

Benefits of technology

It reduces the cost of reaching messages from accident vehicles, improves the accuracy and timeliness of messages, avoids artificial errors, and simplifies the information management process.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN118690265B_ABST
    Figure CN118690265B_ABST
Patent Text Reader

Abstract

The present application provides a management method, device, electronic device and medium for accident vehicle messages. According to an example of the present application, the method may include: obtaining accident vehicle messages from multiple sources by means of monitoring or polling, sorting and cleaning the accident vehicle messages, labeling the accident vehicle messages after sorting and cleaning, generating clue data, and assigning the clue data to the processing flow corresponding to the label based on the label of the clue data, thereby avoiding information errors and omissions caused by manual identification of clue data, automatically obtaining, cleaning and labeling messy clue data, reducing the cost of reaching accident vehicle messages, and improving the accuracy of reaching accident vehicle messages while ensuring the timeliness of reaching accident vehicle messages.
Need to check novelty before this filing date? Find Prior Art

Description

Technical Field

[0001] The present application relates to the field of communication technology, and in particular to a method, device, electronic equipment and medium for managing accident vehicle messages. Background Art

[0002] As the per capita vehicle insurance amount in China continues to increase, the number of accident vehicle information pushed by major insurance companies to vehicle 4S stores is also increasing every day.

[0003] Accident vehicle information between traditional insurance companies and stores is often communicated and interacted through insurance company applications (APP), enterprise WeChat, WeChat, text messages, emails, and other methods. The amount of accident vehicle information data is huge and scattered across different servers, making it difficult for accident vehicle information received by stores to quickly reach store staff within a few minutes.

[0004] At present, the management method of accident vehicle information is usually to achieve the push and notification of accident vehicle information through the interface connection between the store system of the OEM and the insurance company's system. However, the large number of interfaces of many insurance companies connected to the store system may be frequently iterated, and the maintenance cost is high. In addition, the communication mode within the store is complicated. For example, a dedicated person is required to be responsible for distinguishing and forwarding accident vehicle information 24 hours a day, which is costly and prone to human errors, and it is impossible to quickly and accurately push accident vehicle information. Summary of the invention

[0005] In order to overcome the problems existing in the related art, the present application provides a method, device, electronic device and medium for managing accident vehicle messages.

[0006] According to a first aspect of any embodiment of the present application, a method for managing accident vehicle messages is provided, the method comprising:

[0007] Obtain accident vehicle information from multiple sources by monitoring or polling;

[0008] Sorting and cleaning the accident vehicle messages, labeling the sorted and cleaned accident vehicle messages, and generating clue data;

[0009] Based on the label of the clue data, the clue data is allocated to a processing flow corresponding to the label.

[0010] According to a second aspect of any embodiment of the present application, a device for managing accident vehicle information is provided, the device comprising:

[0011] The message monitoring module is used to obtain accident vehicle messages from multiple sources by monitoring or polling;

[0012] A clue generation module is used to sort and clean the accident vehicle messages, label the sorted and cleaned accident vehicle messages, and generate clue data;

[0013] The clue allocation module is used to allocate the clue data to the processing flow corresponding to the tag based on the tag of the clue data.

[0014] According to a third aspect of any embodiment of the present application, there is provided an electronic device, including:

[0015] processor;

[0016] a memory for storing processor-executable instructions;

[0017] The processor implements the method described in any embodiment of the present application by running the executable instructions.

[0018] According to a fourth aspect of any embodiment of the present application, a computer-readable storage medium is provided, on which computer instructions are stored. When the instructions are executed by a processor, the method described in any embodiment of the present application is implemented.

[0019] According to a fifth aspect of any embodiment of the present application, a computer program product is provided, on which a computer program / instruction is stored. When the computer program / instruction is executed by a processor, the method described in any embodiment of the present application is implemented.

[0020] The technical solution provided by this application may have the following beneficial effects:

[0021] According to the above embodiments, accident vehicle messages from multiple sources can be obtained by monitoring or polling, avoiding the need for stores to obtain accident vehicle messages from different service terminals, sorting and cleaning the accident vehicle messages, labeling the sorted and cleaned accident vehicle messages, generating clue data, and allocating the clue data to the processing flow corresponding to the label based on the label of the clue data, thereby avoiding information errors and omissions caused by manual identification of clue data, automatically acquiring, cleaning and marking messy clue data, reducing the cost of reaching accident vehicle messages, and improving the accuracy of reaching accident vehicle messages while ensuring the timeliness of reaching accident vehicle messages.

[0022] It should be understood that the foregoing general description and the following detailed description are exemplary and explanatory only and are not restrictive of the present application. BRIEF DESCRIPTION OF THE DRAWINGS

[0023] The accompanying drawings, which are incorporated in the specification and constitute a part of this application, illustrate embodiments consistent with the application and, together with the description, serve to explain the principles of the application.

[0024] Figure 1is a structural diagram of a message management system according to an exemplary embodiment of the present application;

[0025] Figure 2 is a flow chart of a method for managing accident vehicle information according to an exemplary embodiment of the present application;

[0026] Figure 3 is a schematic diagram of monitoring accident vehicle messages according to an exemplary embodiment of the present application;

[0027] Figure 4 is a schematic diagram of allocating clue data according to an exemplary embodiment of the present application;

[0028] Figure 5 is a flow chart of a clue following method according to an exemplary embodiment of the present application;

[0029] Figure 6 is a flow chart of a clue data generating method according to an exemplary embodiment of the present application;

[0030] Figure 7 is a flow chart of another method for managing accident vehicle information according to an exemplary embodiment of the present application;

[0031] Figure 8 is a schematic diagram of managing accident vehicle messages according to an exemplary embodiment of the present application;

[0032] Fig. 9 is a structural schematic diagram of an electronic device according to an exemplary embodiment of the present application;

[0033] Fig.10 It is a block diagram of a device for managing accident vehicle messages according to an exemplary embodiment of the present application. DETAILED DESCRIPTION

[0034] Exemplary embodiments will be described in detail herein, examples of which are shown in the accompanying drawings. When the following description refers to the drawings, the same numbers in different drawings represent the same or similar elements unless otherwise indicated. The implementations described in the following exemplary embodiments do not represent all implementations consistent with the present application. Instead, they are merely examples of devices and methods consistent with some aspects of the present application as detailed in the appended claims.

[0035] The terms used in this application are for the purpose of describing specific embodiments only and are not intended to limit this application. The singular forms of "a", "said" and "the" used in this application and the appended claims are also intended to include plural forms unless the context clearly indicates other meanings. It should also be understood that the term "and / or" used herein refers to and includes any or all possible combinations of one or more associated listed items.

[0036] It should be understood that although the terms first, second, third, etc. may be used in the present application to describe various information, these information should not be limited to these terms. These terms are only used to distinguish the same type of information from each other. For example, without departing from the scope of the present application, the first information may also be referred to as the second information, and similarly, the second information may also be referred to as the first information. Depending on the context, the word "if" as used herein may be interpreted as "at the time of" or "when" or "in response to determining".

[0037] At present, it is difficult to implement the solution of pushing accident vehicle messages through interface docking, mainly because there are many insurance companies, the number of interfaces is huge and may be frequently iterated, the maintenance cost is high, and manual identification and forwarding of accident vehicle messages are required, which makes it impossible to quickly and accurately push accident vehicle messages.

[0038] In order to solve the above problems, the present application proposes a method for managing accident vehicle messages. To further illustrate the present application, the following embodiments are provided:

[0039] See also Figure 1 , Figure 1 1 is a schematic diagram of a message management system according to an exemplary embodiment of the present application. The message management system may include: terminal devices and a data processing module 104. The terminal devices may include: a first terminal device 101, a second terminal device 102 and a third terminal device 103.

[0040] The terminal device can be a mobile device with the function of receiving accident vehicle messages, such as a smart phone, a tablet computer, etc. The terminal device can be installed with software such as insurance company APP, enterprise WeChat, WeChat, etc. The first terminal device 101, the second terminal device 102 and the third terminal device 103 are terminal devices of different staff members in the store.

[0041] The data processing module 104 is used to manage resources and provide data processing services, and may be a big data processing framework such as Apache Spark, Apache Hadoop, etc. The terminal device and the data processing module 104 may communicate via a network such as a wired or wireless communication link or an optical fiber cable.

[0042] It can be understood that the structure of the message management system shown in the embodiment of the present application is only an example. The number of terminal devices and data processing modules is not limited, and other system structures can also be adopted. As long as the message management system can implement the method described in the embodiment of the present application, the embodiment of the present application does not impose any restrictions on this.

[0043] See also Figure 2 , Figure 2 This is a flowchart of a method for managing accident vehicle messages according to an exemplary embodiment of the present application. The method for managing accident vehicle messages can be applied to a message management system, and the management method may include the following steps:

[0044] Step 202: Obtain accident vehicle information from multiple sources by monitoring or polling.

[0045] In this step, the terminal device in the message management system can obtain accident vehicle messages from various sources such as various insurance company apps, corporate WeChat, text messages, emails, etc. from the underlying message receiving and sending module of the device by setting up a broadcast receiver, listener, or other monitoring methods, or by polling methods such as timed Hypertext Transfer Protocol (HTTP) requests.

[0046] The terminal device can also use the monitoring method to obtain the accident vehicle information. When the monitoring mechanism encounters abnormal conditions such as network problems, the polling method is used as a compensation mechanism to obtain the accident vehicle information through polling.

[0047] The terminal device puts the acquired accident vehicle message into the data stream and pushes it to the data processing module in the message management system to realize real-time data stream transmission and incremental update. This application does not limit the form of data stream, and Kafka data stream is selected as an example.

[0048] Furthermore, the terminal device can identify and extract data from multiple message sources based on the target tag, and obtain accident vehicle messages from multiple sources.

[0049] The message source may include at least one of the following: a device message prompt bar, a broadcast receiver, and a mailbox listener. The device message prompt bar is a message display area on a terminal device.

[0050] The data from multiple message sources can be push messages from external software, system broadcast events related to SMS, email content in emails, etc.

[0051] The target tag is used to sort the accident vehicle messages in the message source. It can be set and adjusted according to the actual business scenario, such as accident, insurance company name, etc. The accident vehicle message is a data stream, which is the data matching the target tag in the data of multiple message sources.

[0052] For accident vehicle messages from external software such as insurance company APP, enterprise WeChat, and WeChat, you can use the device message prompt bar such as the top message bar in the terminal device, listen to the push messages of the external software in the device message prompt bar according to the application package name of the external software, use the target tag for sorting, and obtain the accident vehicle messages.

[0053] For the accident car message in the SMS, the terminal device can create a broadcast receiver (BroadcastReceiver) to listen to the system broadcast events related to SMS, use the target tag for sorting, and obtain the accident car message in the system broadcast events.

[0054] For the accident car message in the email, the terminal device can use the mailbox listener to monitor and intercept the email content in the received email, use the target label to sort it, and obtain the accident car message.

[0055] The terminal device can also identify and obtain target tag data from message sources such as the device message prompt bar, broadcast receiver, mailbox listener, etc. through polling to obtain accident vehicle information.

[0056] See also Figure 3 , Figure 3 This is a schematic diagram of monitoring accident vehicle messages according to an exemplary embodiment of the present application. The first terminal device 101 can obtain accident vehicle messages from multiple sources by monitoring the device message prompt bar, SMS broadcast receiver, and email receiver.

[0057] The first terminal device 101 puts the monitored accident vehicle message into the data stream and pushes it to the data processing module 104 in the message management system.

[0058] As described above, by performing target tag recognition and data screening on data from message sources such as device message prompt bars, broadcast receivers, and mailbox listeners, it is possible to actively obtain accident vehicle messages from multiple sources and types at a low cost, thereby resolving the current complex many-to-many information interaction situation between stores and insurance companies.

[0059] Step 204: sort and clean the accident vehicle messages, label the sorted and cleaned accident vehicle messages, and generate clue data.

[0060] In this step, the data processing module uses the big data processing framework to process the acquired accident vehicle messages. As long as the corresponding data stream is received, it will trigger distributed automatic sorting, cleaning and labeling processing.

[0061] The data processing module can use natural language processing, machine learning, deep learning and other methods to perform semantic analysis and processing on accident vehicle messages, sort and clean the accident vehicle messages, and identify and remove noise, errors, redundancy and irrelevant information in the data.

[0062] The accident vehicle messages after sorting and cleaning are marked with different clue labels to realize label processing of clue data.

[0063] The data processing module stores the clue data corresponding to the generated accident vehicle message in a storage module such as a database, so as to facilitate subsequent operations such as clue allocation and clue analysis.

[0064] Among them, clue data is data with key information and tags in the accident vehicle message.

[0065] Step 206: Based on the tag of the clue data, assign the clue data to the processing flow corresponding to the tag.

[0066] In this step, the store pre-sets the clue data with specific labels according to its own needs and transfers them to the terminal devices of different staff members to establish a processing flow for multiple differentiated clues.

[0067] The data processing module can assign the clue data to the processing flow corresponding to the label based on the clue data, and distribute the clue data to the staff corresponding to the processing flow, to realize a standardized management system and avoid problems such as repeated communication between multiple staff members and car owners and uneven distribution of clue data.

[0068] Among them, the processing flow is a distribution plan personalized according to the actual business scenarios of the store. For example, the lead data marked with labels of self-insurance customer leads are allocated to self-insurance customer follow-up specialists, and the lead data marked with labels of major accidents are allocated to major accident follow-up specialists, etc.

[0069] If the clue data needs to be processed by multiple staff members in sequence, the processing flow may also include: the processing order of the staff members. The data processing module may distribute the clue data to the next staff member in the processing flow according to the processing order of the staff members after the previous staff member completes the processing.

[0070] In one embodiment, each tag may have a corresponding tag level. The corresponding relationship between tags and tag levels can be divided according to the importance of the tags, and the importance of the tags can be set and adjusted according to the actual business scenario.

[0071] See also Figure 4 , Figure 4FIG. 1 is a schematic diagram of allocating clue data according to an exemplary embodiment of the present application. Exemplarily, the tag levels corresponding to the tags may include: primary tags, secondary tags, and tertiary tags.

[0072] The first-level labels can include labels of three dimensions: data source insurance company, whether the user is self-insured, and accident severity. Among them, the label of the data source insurance company dimension includes the name of the insurance company. The label of the self-insured user dimension includes self-insurance and non-self-insurance. The label of the accident severity dimension includes major accidents, scratches, scrapes, deformation, and parts damage.

[0073] Secondary labels can include labels of four dimensions: accident vehicle liability type, accident type, accident location (optional), and willingness to return to the factory (optional). Among them, the labels of the accident vehicle liability type dimension include primary responsibility and three parties. The labels of the accident type dimension include single-party accident and multi-party accident. The labels of the accident location dimension include front side, left side, right side, rear side, and top of the car. The labels of the willingness to return to the factory dimension include high, medium, and low.

[0074] The third-level tags can include tags in three dimensions: other value-added needs (non-essential), customer churn tags (filled in after follow-up of leads), and other manual tags. Among them, the tags in the other value-added needs dimension include accident trailer, replacement vehicle, pickup and delivery vehicle, maintenance, and others. The tags in the customer churn dimension include unable to contact and too high price. The tags in the other manual tag dimension can be customized.

[0075] The personnel roles in the store are set according to the job responsibilities of the staff. The personnel roles may include: ordinary employee of accident vehicle clues, accident external development professional, accident vehicle clues supervisor, store accident vehicle business manager and other store self-built roles.

[0076] When displaying labels of different levels to staff, staff can clearly and quickly identify key information in the first-level label, such as the insurance company from which the data comes and whether the customer is self-insured, which affects the transaction of leads. They can then clarify the specific circumstances of the accident in the second-level label and additional information such as customer requirements in the third-level label, thereby improving the efficiency of reaching lead data.

[0077] The data processing module can allocate the clue data to the processing flow corresponding to the tag according to the tag marked on the clue data.

[0078] For example, the lead data marked with labels of self-insurance customer leads will be assigned to self-insurance customer follow-up specialists first, the lead data marked with labels of major accidents will be assigned to major accident follow-up specialists first, and the lead data marked with labels of value-added will be supplemented with the processing nodes of after-sales value-added service managers, etc.

[0079] When the clue data is assigned to the staff, if the staff does not handle it within a period of time, the message management system can alert the accident vehicle clue supervisor to ensure the timeliness of the clue data.

[0080] After the clue data is generated, the data processing module can also obtain the tag level corresponding to the tag and assign the clue data to the processing flow corresponding to the tag level.

[0081] The tag level can be used to reflect the importance of the tag. For example, the tags in the first-level tags are the most important, the tags in the second-level tags are less important than the tags in the first-level tags, and the tags in the third-level tags are less important than the tags in the second-level tags.

[0082] Each tag level can correspond to different processing procedures such as processing order and staff. For example, the processing procedure corresponding to the first-level tag is to assign the clue data marked as the tag in the first-level tag to the ordinary staff of the accident car clue for processing. After the ordinary staff of the accident car clue has finished processing, the clue data will be assigned to the accident car clue supervisor for review.

[0083] For another example, the processing flow corresponding to the secondary tag is to assign the clue data marked as the label in the secondary tag to the accident outreach specialist for processing. After the accident outreach specialist completes the processing, the clue data will be assigned to the store accident car business manager for review.

[0084] For another example, the processing flow corresponding to the third-level label is to assign the clue data marked with the label in the third-level label to the ordinary employee of the accident vehicle clue for processing. After the ordinary employee of the accident vehicle clue has completed the processing, the processing flow ends.

[0085] If the clue data has been tagged with multiple tags at different tag levels, the data processing module can assign the clue data to the processing flow corresponding to the highest tag level. The tag level can be determined according to the importance of the tag.

[0086] For example, if the clue data is marked with "self-protection" in the first-level label and "main responsibility" in the second-level label, the clue data can be assigned to the processing flow corresponding to the first-level label.

[0087] By obtaining the label level corresponding to the label, allocating the clue data to the processing flow corresponding to the label level, and performing graded processing on the clue data corresponding to different label levels, it can ensure the rational allocation of resources such as staff and time. For example, clue data marked with labels of high importance will be processed first, thereby improving the overall processing efficiency.

[0088] After the clue data is generated, the data processing module may also obtain the tag level corresponding to the tag, and push the clue data to the second terminal based on the personnel role corresponding to the tag level.

[0089] For the second terminal bound to different personnel roles, customized push of clue data corresponding to different tag levels. Among them, the second terminal is the terminal device bound to the personnel role. The terminal device is the device of the staff bound to the personnel role, for example, Figure 1 The first terminal device 101, the second terminal device 102 and the third terminal device 103, etc.

[0090] The correspondence between the label level and the personnel role may be determined according to the responsibilities of the staff member who meets the personnel role.

[0091] For example, if a staff member of an ordinary employee of an accident car clue needs to process clue data marked with a tag in a first-level tag, the clue data marked with a tag in a first-level tag will be pushed to a terminal device bound to the personnel role of an ordinary employee of an accident car clue.

[0092] If the accident outsourcing employee needs to process clue data marked with tags in the secondary tag, the clue data marked with tags in the secondary tag will be pushed to the terminal device bound to the personnel role of the accident outsourcing employee.

[0093] The staff of the after-sales value-added service manager needs to process the clue data marked with the tags in the first-level tag, and push the clue data marked with the tags in the third-level tag to the terminal device bound to the personnel role of the after-sales value-added service manager.

[0094] By pushing clue data to the terminal devices bound to the personnel roles based on the corresponding label levels, it is ensured that the clue data directly related to the staff's duties are pushed to their terminal devices. The staff can directly view and process the clue data matching their personnel roles on the terminal devices, reducing the time cost of label screening and thus improving the work efficiency of the staff.

[0095] Understandably, Figure 4 The labels shown and the specific roles of the staff are set according to the actual business scenarios of the store, and the embodiments of the present application do not limit this.

[0096] Furthermore, the message management system can summarize the lead volume of the store's overall lead data, and display the processing progress and lead volume corresponding to each lead data in the store in real time in the front-end interface of the message management system.

[0097] The message management system can also generate corresponding data analysis of the clue data, displaying weekly and monthly reports on the output value of store accident vehicles, follow-up analysis of accident vehicles, full-cycle process report on clues, business loss of store accident vehicles and other analysis reports, making it convenient for store managers to view and count clue data.

[0098] As mentioned above, by displaying the processing progress and analysis reports corresponding to the clue data, the efficiency of clue processing and management is improved, the clue business of existing accident vehicles in the store is quickly and intuitively displayed and analyzed, the store data visualization and employee efficiency are improved, and the full-process management and statistical difficulties of store accident vehicle clues are solved.

[0099] The accident vehicle message management method of the present embodiment obtains accident vehicle messages from multiple sources by means of monitoring or polling, thereby avoiding the need for stores to obtain accident vehicle messages from different service terminals, sorting and cleaning the accident vehicle messages, labeling the accident vehicle messages after sorting and cleaning, generating clue data, and allocating the clue data to the processing flow corresponding to the label based on the label of the clue data, thereby avoiding information errors and omissions caused by manual identification of clue data, and automatically obtaining, cleaning and labeling messy clue data while supporting the mainstream communication methods of major existing insurance companies, thereby reducing the cost of reaching accident vehicle messages, and improving the accuracy of reaching accident vehicle messages while ensuring the timeliness of reaching accident vehicle messages.

[0100] In addition, by monitoring the accident vehicle information, the timeliness of the accident vehicle information can be further improved, without frequent requests to the source, and the resource utilization efficiency is high. By polling the accident vehicle information, there is no need for complex real-time communication technology, which is simpler to implement and has good compatibility.

[0101] In the above-mentioned embodiments, it is introduced that the accident vehicle messages from various sources are monitored or polled, the accident vehicle messages are sorted, cleaned and labeled, and the generated clue data is assigned to the processing flow corresponding to the label. In the following embodiments, how to further ensure the timeliness of the clue data will be described in detail, and it can be applied to any of the above embodiments.

[0102] In one embodiment, the message management system can call the application programming interface (API) of the Alibaba Cloud robot phone and use voice outbound calls to push the message to the staff corresponding to the assigned processing flow to remind the staff.

[0103] The message management system can also remind staff in real time according to store needs through SMS messages, email messages and other reminders on terminal devices.

[0104] The message management system can also send real-time message reminders in the store’s corporate WeChat group. By connecting to the robot in the corporate WeChat group, message reminders can be distributed within the corporate WeChat group.

[0105] For example, by adding a robot to a group chat and initiating an HTTP or POST request to the robot's unique webhook url, you can send a message reminder to the group.

[0106] For example, you can add a custom robot to a group chat, which supports sending message reminders such as text, markdown, pictures, graphics, files, voice, template cards, etc.

[0107] In addition, for the robot's text / markdown type messages, use the <@userid> extended syntax in the content to specifically remind (@) the relevant staff in the message reminder.

[0108] The message management system can also automatically distribute group chat messages and provide special reminders to relevant personnel at event nodes such as the creation of lead data processing flow and lead follow-up status changes.

[0109] As described above, the staff corresponding to the processing flow are reminded through preset voice calls, SMS messages, email messages, group chat messages and other reminder methods to urge timely follow-up and processing of lead data, thereby further improving the staff's response speed to lead data.

[0110] In one embodiment, the message management system can be connected to the store management system of the store to obtain the scheduling information in the store management system.

[0111] When allocating lead data, the message management system can combine the staff's shift information, match the roles of the personnel who follow up the lead data at each node in the processing flow with the roles of the staff on duty at the store on that day, customize the processing flow corresponding to the label, and allocate the lead data to the on-duty personnel corresponding to the processing flow.

[0112] When the on-duty staff fails to follow up the lead data within the preset time period, the message management system adds personnel to follow up the lead data. For example, the processing flow of lead data is opened up to all staff in the store to follow up, and the processing flow of lead data is opened up to staff corresponding to a certain personnel role.

[0113] The shift information is used to mark the personnel roles of on-duty personnel and off-duty personnel in the store roster. On-duty personnel are staff members who are working or on duty in the store.

[0114] The preset time period is a certain time period after the lead data is assigned, for example: one minute after the lead data is assigned, two minutes after the lead data is assigned, etc. The follow-up staff is the staff in the store who has the authority to process the lead data.

[0115] See also Figure 5 , Figure 5 This is a flow chart of a method for following up a clue according to an exemplary embodiment of the present application. The method for following up a clue may include the following steps:

[0116] Step 502: Allocate clue data to the processing flow corresponding to the tag.

[0117] In this step, the message management system generates a distribution plan corresponding to the clue data and distributes the clue data to the processing flow corresponding to the tag.

[0118] Step 504: Remind the on-duty personnel corresponding to the processing flow.

[0119] In this step, the message management system combines the shift information of the store staff and distributes the clue data to the on-duty personnel corresponding to the processing flow.

[0120] Remind the on-duty personnel corresponding to the processing process through voice calls, text messages, email messages, group chat messages and other reminder methods.

[0121] Step 506: Determine whether to follow up on the clue data within one minute.

[0122] In this step, the message management system determines whether the on-duty personnel follow up the clue data within one minute.

[0123] If the on-duty personnel follows up the clue data within one minute, step 510 is executed;

[0124] If the on-duty personnel do not follow up the clue data within one minute for the first time, step 504 is executed;

[0125] If the on-duty personnel does not follow up the clue data within one minute for the second time, step 508 is executed.

[0126] Step 508: Add follow-up personnel for the clue data.

[0127] In this step, the message management system adds follow-up personnel for the lead data, and opens up the processing flow of the lead data to all relevant employees of the store for follow-up.

[0128] Step 510: The message management system changes the follow-up status of the clue data.

[0129] As mentioned above, by combining the staff's scheduling information, customizing the processing flow corresponding to the label, and allocating the clue data to the on-duty personnel corresponding to the processing flow, it is ensured that the clue data is quickly transferred to the on-duty personnel, greatly improving the efficiency of clue flow and management, and further improving the response speed of clue data; when the on-duty personnel fail to follow up the clue data within the preset time period, increase the number of personnel who follow up the clue data, reduce the waiting time for the clue, and improve the efficiency of the entire processing process.

[0130] In the above-mentioned embodiment, the response speed of reaching clue data is improved by means of message reminder, assignment to on-duty personnel and adding follow-up personnel. In the following embodiment, a more detailed description will be given on how to generate clue data corresponding to the accident vehicle message, which can be applied to any of the above embodiments.

[0131] The message management system can perform word segmentation processing on the accident vehicle message and the first keyword respectively through natural language technologies such as HanLP (Han Language Processing, natural language processing toolkit) and jieba (Chinese word segmentation tool), and obtain the word segmentation results of the accident vehicle message and the first keyword.

[0132] The word segmentation result of the accident vehicle message is matched with the word segmentation result of the first keyword to obtain a first matching result. Similarity analysis and template processing are performed on multiple keywords in the first matching result to obtain initial clue data.

[0133] Among them, the first keyword is a keyword in the accident vehicle message set according to the business scenario, which can be a label under multiple dimensions such as data source insurance company, accident vehicle responsibility type, accident type, etc., for example: main responsibility, third responsibility, single-party accident, multi-party accident, etc.

[0134] The first matching result is a keyword that matches the word segmentation result of the accident vehicle message in the first keyword. The initial clue data is the accident vehicle message after sorting, cleaning and template processing.

[0135] The message management system can match the initial clue data with the second keyword to obtain a second matching result. According to the second matching result, a label corresponding to the matching result is set for the initial clue data to generate labeled clue data.

[0136] The second keyword is a keyword used to label the clue data, which can be a label under multiple dimensions such as the severity of the accident, whether the customer is self-insured, the location of the accident, etc. For example, self-insured, non-self-insured, major accident, top of the car, etc. The second keyword can be the same as the first keyword or different from the first keyword. The second matching result is the keyword that matches the initial clue data in the second keyword.

[0137] See also Figure 6 , Figure 6 This is a flow chart of a clue data generation method according to an exemplary embodiment of the present application. The clue data generation method may include the following steps:

[0138] Step 602: Use natural language processing technology to segment the accident vehicle message and the first keyword.

[0139] In this step, the data processing module uses natural language processing technology to perform word segmentation processing on the accident vehicle message and the first keyword to obtain the word segmentation results of the accident vehicle message and the first keyword.

[0140] Exemplarily, the data processing module uses HanLP to load or train word vector models such as Word2Vec (Word to Vector, a tool for converting words into vector form) and GloVe (Global Vectors for Word Representation, an unsupervised learning algorithm for generating word vectors), and calculates the semantic similarity between texts through word embedding, thereby matching data with similar semantics but different lexical forms.

[0141] Step 604: Determine a first matching result in which the word segmentation result of the accident vehicle message is the same as the word segmentation result of the first keyword.

[0142] In this step, the data processing module takes a union of the word segmentation result of the accident vehicle message and the word segmentation result of the first keyword, and determines a first matching result in which the word segmentation result of the accident vehicle message is the same as the word segmentation result of the first keyword.

[0143] Step 606: Calculate the similarity between the multiple keywords in the first matching result.

[0144] In this step, the data processing module can calculate the word frequency vector according to the word segmentation results of the accident vehicle message and the word segmentation results of the first keyword, and use similarity measurement methods such as cosine similarity and Jaccard (Jaccard similarity coefficient) similarity to calculate the similarity between multiple keywords in the first matching result.

[0145] Step 608: Determine whether the similarity exceeds a similarity threshold.

[0146] In this step, the data processing module can determine whether the similarity exceeds a similarity threshold.

[0147] Among them, the similarity threshold is the offset threshold between the texts in the accident vehicle message, which is used to verify the semantic similarity between the texts in the accident vehicle message, and can determine whether to generate clue data for the accident vehicle message, for example: 40%, 50%, etc.

[0148] If the similarity exceeds the similarity threshold, continue to execute step 610;

[0149] If the similarity does not exceed the similarity threshold, step 616 is executed.

[0150] Step 610: Fill the first matching result into the preset template to obtain initial clue data.

[0151] In this step, the data processing module can fill the first matching result into a preset template to generate initial clue data.

[0152] The preset template is used to assist the staff in understanding the clue data, for example, "{&} occurred in {&}, involving {&}, please handle it as soon as possible!". "{&}" is the matching word segmentation result to be filled.

[0153] Step 612: Match the initial clue data with the second keyword, set a label for the initial clue data according to the second matching result, and generate clue data.

[0154] In this step, the data processing module may match the initial clue data with the second keyword, set a label for the initial clue data according to the second matching result, and generate labeled clue data.

[0155] Step 614: Push the clue data to the staff corresponding to the processing flow and store it in the storage module.

[0156] In this step, the data processing module can push the clue data to the staff corresponding to the processing flow, establish a variety of differentiated accident vehicle clue processing flows, and store the clue data in the storage module.

[0157] Step 616: No clue data is generated.

[0158] In this step, the data processing module does not generate clue data for accident vehicle messages whose similarity does not exceed the similarity threshold, and does not need to allocate them.

[0159] As described above, by performing word segmentation and matching on the accident vehicle message and the first keyword, initial clue data is obtained based on the first matching result, the initial clue data is matched with the second keyword, and a label is set for the initial clue data according to the second matching result, and standardized and labeled clue data is automatically generated to avoid problems such as slow manual processing, high cost, and easy errors.

[0160] Furthermore, after allocating clue data, the message management system can push clue accuracy inquiry information to the first terminal of the staff member corresponding to the processing flow. The staff member uses the first terminal to send the clue accuracy evaluation to the message management system.

[0161] The message management system receives the clue accuracy evaluation sent by the first terminal and adjusts the matching score of the matching result. The matching score is used to determine the matching order of the matching results. The higher the matching score, the higher the priority of using the keywords in the matching result for matching in the subsequent matching process.

[0162] Among them, the clue accuracy evaluation is the feedback information from the staff on the accuracy of the clue data.

[0163] The matching result is a keyword that matches the clue data during the sorting, cleaning and labeling process. The matching result may include: a first matching result and / or a second matching result.

[0164] The clue accuracy evaluation may include: positive feedback, for example, the clue data is accurate. If the clue accuracy evaluation is determined to be positive feedback, the matching score of the matching result may be increased.

[0165] By increasing the matching score of the first matching result, in the subsequent sorting and cleaning process of the accident vehicle messages, the keyword with the highest matching score in the first keyword can be preferentially used to match the accident vehicle messages according to the matching scores of each keyword in the first keyword.

[0166] By increasing the matching score of the second matching result, in the subsequent labeling process, according to the matching scores of the respective keywords in the second keywords, the keyword with the highest matching score in the second keywords can be preferentially used to match the initial clue data.

[0167] The matching score can be adjusted, for example, by adding points to the score field in the field table for the matching result of the pushed clue data, and using the points as the matching score. The matching result with a high score has a higher matching order. In the subsequent sorting and cleaning, there is a greater probability of using the first keyword with a high score to obtain the initial clue data, and in the labeling process, there is a greater probability of using the second keyword with a high score to generate clue data, thereby making the knowledge base more and more accurate.

[0168] Taking the second keyword including self-preservation and non-self-preservation as an example, the integral value of self-preservation in the field table is 2, and the integral value of non-self-preservation is 3. After receiving the clue accuracy evaluation of the clue data marked with the self-preservation label as positive feedback, the integral value of self-preservation in the field table can be increased by 2 points, and the integral value of self-preservation is higher than that of non-self-preservation, and the self-preservation label is used preferentially for processing in the subsequent labeling process.

[0169] The clue accuracy evaluation may also include: negative feedback, for example, the clue data is inaccurate. If the clue accuracy evaluation is determined to be negative feedback, the matching score of the matching result may be reduced.

[0170] The message management system can also adjust the similarity threshold based on the accuracy evaluation of the received clues. As the content of the knowledge base becomes more and more accurate, the message management system can gradually increase the similarity threshold to make the pushed clue data more accurate.

[0171] If the accuracy evaluation of the received clue is negative feedback, the message management system can also feed back the pushed clue data to the administrator of the message management system for improvement.

[0172] As described above, by receiving the clue accuracy evaluation sent by the first terminal, if the clue accuracy evaluation is determined to be positive feedback, the matching score of the matching result is increased, and the matching order of the matching result is dynamically adjusted based on the feedback from the staff, the accuracy of the clue data can be improved while speeding up the sorting, cleaning and labeling of accident vehicle messages.

[0173] To further explain the management process of accident vehicle information, Figure 7 A flowchart showing another method for managing accident vehicle messages is shown. The method for managing accident vehicle messages may include the following steps:

[0174] Step 702: Obtain accident vehicle information from multiple sources by monitoring or polling.

[0175] In this step, the terminal device identifies and obtains target tag data from multiple sources such as the device message prompt bar, broadcast receiver, and mailbox listener by monitoring or polling, and obtains accident vehicle messages from multiple sources and types. The terminal device puts the accident vehicle message into the Kafka data stream and pushes it to the data processing module.

[0176] See also Figure 8 , Figure 8 A schematic diagram of managing accident vehicle information is shown. The store system and the insurance company communicate accident vehicle information through external software, text messages, emails and other communication software. The communication software is installed in the first terminal device 101 of the staff, and the first terminal device 101 pushes the accident vehicle information obtained by monitoring or polling to the data processing module 104.

[0177] Step 704: perform word segmentation and matching on the accident vehicle message and the first keyword, and obtain initial clue data based on the first matching result.

[0178] In this step, the data processing module uses natural language processing technology to perform word segmentation processing and matching on the accident vehicle message and the first keyword, generating the word segmentation result of the accident vehicle message and the word segmentation result of the first keyword, and determining the first matching result in which the word segmentation result of the accident vehicle message is the same as the word segmentation result of the first keyword.

[0179] When the similarity between multiple keywords of the first matching result exceeds a similarity threshold, the data processing module fills the first matching result into a preset template to obtain initial clue data, thereby realizing sorting, cleaning and templating of accident vehicle messages.

[0180] Please continue reading Figure 8 The data processing module 104 sorts, cleans and labels the accident vehicle messages.

[0181] Step 706: Match the initial clue data with the second keyword, set a label for the initial clue data according to the second matching result, and generate clue data.

[0182] In this step, the data processing module matches the initial clue data with the second keyword, sets a label for the initial clue data according to the second matching result, implements labeling processing of the initial clue data, and generates labeled clue data.

[0183] Step 708: Allocate the clue data to the processing flow corresponding to the tag level.

[0184] In this step, the message management system obtains the tag level corresponding to the tag and distributes the clue data to the processing flow corresponding to the tag level.

[0185] The message management system can also combine the staff's shift information, customize the processing flow corresponding to the tags, and create a distribution plan for lead data.

[0186] Please continue reading Figure 8 ,The message management system forms a distribution plan based on the ,tags and scheduling information of the clue data.

[0187] Step 710: Push clue data to the second terminal based on the personnel role corresponding to the tag level.

[0188] In this step, the message management system can customize and push clue data corresponding to different tag levels to the second terminal bound to different personnel roles based on the personnel roles corresponding to the tag levels.

[0189] Step 712: Allocate the clue data to the on-duty personnel corresponding to the processing flow.

[0190] In this step, the message management system distributes the clue data to the on-duty personnel corresponding to the processing flow.

[0191] Step 714: Remind the on-duty personnel based on the preset reminder method.

[0192] In this step, the message management system reminds the on-duty personnel based on reminder methods such as voice calls, text messages, email messages, group chat messages, etc.

[0193] Please continue reading Figure 8 The message management system automatically calls the first terminal device 101 of the on-duty personnel corresponding to the processing flow through a voice outbound call reminder.

[0194] Step 716: If the on-duty personnel do not follow up the clue data within the preset time period, the message management system adds a follow-up personnel for the clue data.

[0195] Step 718: Display the processing progress and analysis report corresponding to the clue data.

[0196] In this step, the message management system displays the processing progress and analysis report corresponding to the clue data.

[0197] Please continue reading Figure 8 , the message management system can form data analysis for lead follow-up cycle management.

[0198] Step 720: Adjust the matching order of the matching results according to the received clue accuracy evaluation.

[0199] In this step, if the clue accuracy evaluation is determined to be positive feedback, the matching score of the first matching result and the matching score of the second matching result can be increased. In the subsequent sorting, cleaning and labeling processing, the keyword with the highest matching score is preferentially used to match the accident vehicle message, so that the knowledge base and the assigned clue data are more and more accurate.

[0200] Fig. 9 1 is a schematic diagram of a structure of an electronic device according to an exemplary embodiment of the present application. The electronic device may be, for example, a mobile phone, a computer, a digital broadcast terminal, a message transceiver device, a game console, a tablet device, a personal digital assistant, a server, a smart home appliance, etc. Fig. 9 At the hardware level, the electronic device includes a processor 902, an internal bus 904, a network interface 906, a memory 908, and a non-volatile memory 910, and may also include hardware required for other services. The processor 902 reads the corresponding computer program from the non-volatile memory 910 into the memory 908 and then runs it, forming a management device for accident vehicle messages at the logical level. Of course, in addition to software implementations, this application does not exclude other implementations, such as logic devices or a combination of software and hardware, etc., that is to say, the execution subject of the following processing flow is not limited to each logic unit, but can also be hardware or logic devices.

[0201] Fig.10 1 is a block diagram of a device for managing accident vehicle information according to an exemplary embodiment of the present application. Fig.10 , the device may include: a message monitoring module 1002, a clue generating module 1004 and a clue allocating module 1006, wherein:

[0202] The message monitoring module 1002 is used to obtain accident vehicle messages from multiple sources by monitoring or polling;

[0203] The clue generation module 1004 is used to sort and clean the accident car messages, label the sorted and cleaned accident car messages, and generate clue data;

[0204] The clue allocation module 1006 is used to allocate the clue data to the processing flow corresponding to the tag based on the tag of the clue data.

[0205] In one example, the message monitoring module 1002, when used to obtain accident vehicle messages from multiple sources by monitoring or polling, includes: identifying and extracting data from multiple message sources based on target tags to obtain the accident vehicle messages; the message sources include at least one of the following: a device message prompt bar, a broadcast receiver, and a mailbox listener.

[0206] In one example, the clue generation module 1004, when used to sort and clean the accident vehicle messages, label the accident vehicle messages after sorting and cleaning, and generate clue data, includes: performing word segmentation and matching on the accident vehicle messages and the first keyword, and obtaining initial clue data based on the first matching result; matching the initial clue data with the second keyword, setting a label for the initial clue data according to the second matching result, and generating the clue data.

[0207] In one example, the clue generation module 1004 is also used to receive a clue accuracy evaluation sent by the first terminal; if the clue accuracy evaluation is determined to be positive feedback, the matching score of the matching result is increased; wherein the matching result includes: the first matching result, and / or, the second matching result, and the matching score is used to determine the matching order of the matching results.

[0208] In one example, the clue allocation module 1006, when used to allocate the clue data to the processing flow corresponding to the tag based on the tag of the clue data, includes: obtaining the tag level corresponding to the tag; allocating the clue data to the processing flow corresponding to the tag level; and / or, based on the personnel role corresponding to the tag level, pushing the clue data to a second terminal; wherein the second terminal is a terminal device bound to the personnel role.

[0209] In one example, the clue allocation module 1006, when used to allocate the clue data to the processing flow corresponding to the label based on the label of the clue data, includes: customizing the processing flow corresponding to the label in combination with the staff's scheduling information; allocating the clue data to the on-duty personnel corresponding to the processing flow; and adding personnel to follow up the clue data if the on-duty personnel fail to follow up the clue data within a preset time period.

[0210] In one example, the clue distribution module 1006 is further used to display the processing progress and analysis report corresponding to the clue data.

[0211] The implementation process of the functions and effects of each unit in the above-mentioned device is specifically described in the implementation process of the corresponding steps in the above-mentioned method, and will not be repeated here.

[0212] For the device embodiment, since it basically corresponds to the method embodiment, the relevant parts can refer to the partial description of the method embodiment. The device embodiment described above is only schematic, wherein the modules described as separate components may or may not be physically separated, and the components displayed as modules may or may not be physical modules, that is, they may be located in one place, or they may be distributed on multiple network modules. Some or all of the modules may be selected according to actual needs to achieve the purpose of the present application scheme. A person of ordinary skill in the art can understand and implement it without paying creative labor.

[0213] In an exemplary embodiment, a non-transitory computer-readable storage medium including instructions is also provided, such as a memory including instructions. The above instructions can be executed by a processor of the accident vehicle message management device to implement any method described in the above embodiments.

[0214] The non-temporary computer-readable storage medium may be a ROM, a random access memory (RAM), a CD-ROM, a magnetic tape, a floppy disk, an optical data storage device, etc., and the present application does not limit this.

[0215] In an exemplary embodiment, a computer program product including a computer program / instruction is also provided. The computer program / instruction can be executed by a processor of a device for managing accident vehicle messages to implement any of the methods described in the above embodiments.

[0216] The above describes specific embodiments of the present application. Other embodiments are within the scope of the appended claims. In some cases, the actions or steps recorded in the claims can be performed in an order different from that in the embodiments and still achieve the desired results. In addition, the processes depicted in the accompanying drawings do not necessarily require the specific order or continuous order shown to achieve the desired results. In some embodiments, multitasking and parallel processing are also possible or may be advantageous.

[0217] Those skilled in the art will readily appreciate other embodiments of the present application after considering the specification and practicing the inventions claimed herein. The present application is not limited to the precise structures described above and shown in the accompanying drawings, and various modifications and changes may be made without departing from the scope thereof. The scope of the present application is limited only by the appended claims.

[0218] The above description is only a preferred embodiment of the present application and is not intended to limit the present application. Any modifications, equivalent substitutions, improvements, etc. made within the spirit and principles of the present application shall be included in the scope of protection of the present application.

Claims

1. A method for managing accident vehicle information, characterized in that: The method comprises: Obtain accident vehicle information from multiple sources by monitoring or polling; Sorting and cleaning the accident vehicle messages, labeling the accident vehicle messages after sorting and cleaning to generate clue data; the clue data is data with key information and labels in the accident vehicle messages; the sorting and cleaning includes identifying and removing noise, errors and irrelevant information in the accident vehicle messages; Based on the label of the clue data, assigning the clue data to a processing flow corresponding to the label; The step of sorting and cleaning the accident vehicle messages, labeling the sorted and cleaned accident vehicle messages, and generating clue data includes: Performing word segmentation and matching on the accident vehicle message and the first keyword, and acquiring initial clue data based on the first matching result; the initial clue data is generated by filling the first matching result into a preset template; The initial clue data is matched with a second keyword, and a label is set for the initial clue data according to a second matching result to generate the clue data.

2. The method according to claim 1, characterized in that The accident vehicle information is obtained from multiple sources by monitoring or polling, including: Identify and extract data from multiple message sources based on target tags to obtain the accident vehicle message; The message source includes at least one of the following: a device message prompt bar, a broadcast receiver, and a mailbox listener.

3. The method according to claim 1, characterized in that The method further comprises: receiving a clue accuracy evaluation sent by the first terminal; If it is determined that the clue accuracy evaluation is positive feedback, the matching score of the matching result is increased; wherein the matching result includes: the first matching result and / or the second matching result, and the matching score is used to determine the matching order of the matching results.

4. The method according to claim 1, characterized in that: The step of allocating the clue data to a processing flow corresponding to the tag based on the clue data includes: Get the tag level corresponding to the tag; Allocating the clue data to the processing flow corresponding to the tag level; and / or, Based on the personnel role corresponding to the tag level, the clue data is pushed to the second terminal; wherein the second terminal is a terminal device bound to the personnel role.

5. The method according to claim 1, characterized in that The step of allocating the clue data to a processing flow corresponding to the tag based on the clue data includes: Customize the processing flow corresponding to the label in combination with the staff's shift information; Allocating the clue data to the on-duty personnel corresponding to the processing flow; In the case that the on-duty personnel do not follow up the clue data within the preset time period, additional personnel are added to follow up the clue data.

6. The method according to claim 1, characterized in that The method further comprises: Display the processing progress and analysis report corresponding to the clue data.

7. A management device for accident vehicle information, characterized in that: The device comprises: The message monitoring module is used to obtain accident vehicle messages from multiple sources by monitoring or polling; A clue generation module is used to sort and clean the accident car messages, label the sorted and cleaned accident car messages, and generate clue data; the clue data is data with key information and labels in the accident car messages; the sorting and cleaning includes identifying and removing noise, errors and irrelevant information in the accident car messages; A clue allocation module, used for allocating the clue data to a processing flow corresponding to the tag based on the tag of the clue data; The clue generation module, when used to sort and clean the accident vehicle messages, label the accident vehicle messages after sorting and cleaning, and generate clue data, includes: segmenting and matching the accident vehicle messages and the first keyword, and obtaining initial clue data based on the first matching result; the initial clue data is generated by filling the first matching result into a preset template; matching the initial clue data with the second keyword, setting a label for the initial clue data according to the second matching result, and generating the clue data.

8. An electronic device, characterized in that: include: processor; a memory for storing processor-executable instructions; The processor implements the method according to any one of claims 1 to 6 by running the executable instructions.

9. A computer-readable storage medium having computer instructions stored thereon, characterized in that: When the instruction is executed by a processor, the method according to any one of claims 1 to 6 is implemented.

10. A computer program product having a computer program / instructions stored thereon, characterized in that: When the computer program / instructions are executed by a processor, the method according to any one of claims 1 to 6 is implemented.