Work order risk identification method and device and electronic equipment

Abstract

The invention discloses a work order risk identification method and apparatus, and an electronic device, belongs to the technical field of computers, and aims to solve the problem of lack of a technical scheme for identifying high-risk work orders in government affair hotline work orders at present. The method comprises the following steps: acquiring item contents of a plurality of registered items in a petition item library, and determining at least one group complaint event according to the item contents; obtaining content information of a plurality of target work orders in a government affair hotline work order library; for each target work order, matching the target work order with all group complaint events according to the content information of the target work order to obtain a matching result; the matching result comprises a corresponding relationship between the target work order and the group complaint event; determining a first number of target work orders corresponding to each group complaint event according to a matching result of all the target work orders; and for each group complaint event, determining that the target work order corresponding to the group complaint event is a high-risk work order under the condition that the first number meets a preset condition.

Description

Technical Field

[0001] This application belongs to the field of computer technology, specifically relating to a risk identification method, device, and electronic device for work orders. Background Technology

[0002] A group complaint incident, also known as a mass appeal incident, refers to an event in which multiple stakeholders (such as customers, employees, and residents) raise the same or a series of related issues regarding the business operations (services, products, management, etc.) of one or more involved entities (organizations, institutions, enterprises, etc.), and thus submit the same or similar demands to a government regulatory platform. Examples include complaints received continuously between 10 PM and midnight about noise pollution from square dancing in a residential area, reports received around noon about food safety at a restaurant, and suggestions received on weekends regarding queuing arrangements at a scenic spot.

[0003] When citizens need to raise questions or requests to government regulatory platforms, the two most common methods are: calling the 12345 government hotline or going through the petition process. When citizens raise issues through the government hotline, they make a phone call. After the call, the operator will assign a work order to the relevant party involved. The issue is generally resolved within 3 to 5 working days, and the relevant government hotline department will reply via SMS afterward. When citizens raise issues through the petition process, they often submit a registration form online. After receiving the registration form, the petition department will forward the issue to other competent authorities. The issue is generally resolved within 75 days, and the petition department will issue a written, officially stamped, handling opinion letter afterward.

[0004] From the public's perspective, making a phone call is inexpensive, indicating that the problems they encounter are not yet very harmful, and they believe their grievances can be easily resolved through the government hotline. In contrast, filling out an online registration form not only consumes time and effort but also requires a long wait for results. This suggests that the problems they face are more serious, and they are willing to expend time and effort to obtain a stamped opinion from a government department for a more reliable resolution. Therefore, the petitions received by the petition department contain a large number of cases that have already caused some harm, while the number of cases involving significant harm dispatched through the government hotline is relatively smaller.

[0005] However, a moderately large city receives tens of thousands of calls to government hotlines every day. It's impossible to resolve all the resulting work orders in a short period. Therefore, some citizens will inevitably resort to petitioning procedures when the harm becomes increasingly severe and phone calls fail. Thus, among the massive number of government hotline calls, there are bound to be some high-risk incidents that are not currently causing significant harm but could escalate dramatically in the future. Identifying these incidents can help prevent problems before they arise. Summary of the Invention

[0006] This application provides a risk identification method, device, and electronic device for work orders, which can solve the problem of the current lack of technical solutions for identifying high-risk work orders in government hotline work orders.

[0007] In a first aspect, embodiments of this application provide a risk identification method for work orders, comprising: acquiring the content of multiple registered matters in a petition matters database; determining at least one group complaint event based on the content of the matters; acquiring the content information of multiple target work orders in a government hotline work order database; for each target work order, matching the target work order with all group complaint events based on the content information of the target work order to obtain a matching result; the matching result includes the correspondence between the target work order and the group complaint event; determining a first number of target work orders corresponding to each group complaint event based on the matching results of all target work orders; and for each group complaint event, determining the target work order corresponding to the group complaint event as a high-risk work order if the first number meets a preset condition. Secondly, embodiments of this application provide a risk identification device for work orders, comprising: an acquisition and determination module, configured to acquire the content of multiple registered matters in a petition matters database, and determine at least one group complaint event based on the content of the matters; an acquisition module, configured to acquire the content information of multiple target work orders in a government hotline work order database; a matching processing module, configured to perform matching processing on each target work order and all group complaint events based on the content information of the target work order, to obtain a matching result; the matching result includes the correspondence between the target work order and the group complaint event; a first determination module, configured to determine a first number of target work orders corresponding to each group complaint event based on the matching result of all target work orders; and a second determination module, configured to determine, for each group complaint event, if the first number meets a preset condition, that the target work order corresponding to the group complaint event is a high-risk work order.

[0008] Thirdly, embodiments of this application provide an electronic device including a processor; and a memory arranged to store computer-executable instructions configured to be executed by the processor to implement the steps of the risk identification method for work orders as described in the first aspect.

[0009] Fourthly, embodiments of this application provide a computer-readable storage medium for storing computer-executable instructions, which, when executed by a processor, implement the steps of the risk identification method for work orders as described in the first aspect.

[0010] Fifthly, embodiments of this application provide a computer program product, which includes a computer program that, when executed by a processor, implements the steps of the risk identification method for work orders as described in the first aspect.

[0011] In a sixth aspect, embodiments of this application provide a chip, the chip including a processor and a communication interface, the communication interface being coupled to the processor, the processor being used to run executable instructions to implement the steps of the risk identification method for work orders as described in the first aspect.

[0012] In this embodiment, the content of multiple registered matters in the petition matters database is obtained, thereby identifying at least one group complaint event based on the content. The content information of multiple target work orders in the government hotline work order database is also obtained. For each target work order, the target work order is matched with all group complaint events based on its content information to obtain a matching result, which includes the correspondence between the target work order and the group complaint event. Furthermore, based on the matching results of all target work orders, a first number of target work orders corresponding to each group complaint event is determined. For each group complaint event, if the first number meets a preset condition, the target work order corresponding to that group complaint event is identified as a high-risk work order. As can be seen, this technical solution identifies group complaint events in the petition database and matches target work orders obtained from the government hotline work order database with each group complaint event to determine the group complaint event to which each target work order belongs. Based on the number of target work orders contained in each group complaint event, high-risk work orders are identified among the target work orders. This achieves the effect of identifying high-risk work orders in the government hotline work order database, which is conducive to promoting the handling of such work orders by staff, not only avoiding greater impact, but also improving the processing efficiency of high-risk work orders. Attached Figure Description

[0013] Figure 1 This is a flowchart illustrating a risk identification method for work orders provided in an embodiment of this application; Figure 2 This is a flowchart illustrating a risk identification method for work orders provided in another embodiment of this application; Figure 3 This is a schematic diagram of the structure of a work order risk identification device provided in an embodiment of this application; Figure 4This is a schematic diagram of the structure of the electronic device provided in the embodiments of this application. Detailed Implementation

[0014] The technical solutions of the embodiments of this application will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of this application, not all embodiments. Based on the embodiments of this application, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of this application.

[0015] The terms "first," "second," etc., used in the specification and claims of this application are used to distinguish similar objects and not to describe a specific order or sequence. It should be understood that such use of data can be interchanged where appropriate so that embodiments of this application can be implemented in orders other than those illustrated or described herein, and the objects distinguished by "first," "second," etc., are generally of the same class and the number of objects is not limited; for example, a first object can be one or more. Furthermore, in the specification and claims, "and / or" indicates at least one of the connected objects, and the character " / " generally indicates that the preceding and following objects are in an "or" relationship.

[0016] The risk identification method, apparatus, and electronic equipment for work orders provided in this application will be described in detail below with reference to the accompanying drawings and through specific embodiments and application scenarios.

[0017] Figure 1 This application illustrates a risk identification method for work orders according to an embodiment. This method can be executed by an electronic device, which may include a server and / or a terminal device, such as a vehicle-mounted terminal or a mobile terminal. In other words, the method can be executed by software or hardware installed on the electronic device, and includes the following steps: Step 102: Obtain the content of multiple registered matters in the petition matters database, and determine at least one group litigation event based on the content of the matters.

[0018] Each registered item corresponds to a public request, and the content of the request can include the entire content of the request as well as the opinions of all relevant departments. It is understandable that the opinions of all relevant departments can be targeted responses to the specific request, or responses regarding the expected response time, the handling department, contact information, etc.

[0019] During the operation of the petition department, petitions submitted by the public are forwarded to the authorized units for processing and simultaneously stored in the petition department's petition database in the form of petition registration forms. Therefore, the petition database contains a large number of registered petitions. To facilitate management, a visible time window for registered petitions can be set, such as one week, two weeks, or one month. Each time the risk identification process for a work order is executed, the content of all registered petitions within the currently visible time window in the petition database is retrieved to identify at least one group complaint incident. This ensures that subsequent group complaint incidents are consistent with recent events and are highly timely.

[0020] In one embodiment, since some group litigation incidents are not risky, in order to efficiently exclude these non-risk incidents, high-risk screening conditions can be set to further filter the group litigation incidents according to actual application needs. These high-risk screening conditions can include the type of claim, whether it includes involved parties, the urgency of the incident, and the magnitude of its social impact. Therefore, after identifying at least one group litigation incident, at least one high-risk group litigation incident can be selected from it based on the preset high-risk screening conditions. This high-risk incident can then replace subsequent group litigation incidents in the process, ultimately determining the high-risk work orders.

[0021] High-risk group litigation incidents refer to those selected from group litigation cases that may pose significant harm to stakeholders and have a major negative impact on the parties involved and even society. These incidents typically require higher attention from relevant departments due to the urgency of the demands, the large number of people involved, and the severity of subsequent problems. For example, incidents involving public safety hazards, serious service quality issues, or frequent product malfunctions during major holidays may all be classified as high-risk group litigation incidents. Therefore, for government regulatory platforms, timely identification of high-risk group litigation incidents within their jurisdiction can help them quickly pinpoint the type and source of the risk and urge relevant departments to resolve related issues, thereby preventing larger-scale negative impacts or more serious consequences.

[0022] Step 104: Obtain the content information of multiple target work orders from the government hotline work order database.

[0023] Each work order in the government hotline work order database corresponds to a public request. The work order can contain the full content of the public request and all the processing records of the public request by the platform.

[0024] Similar to the petition database, the government hotline work order database contains a large number of work orders. To facilitate management, a visible time window for each work order can be set, such as one week, two weeks, or one month. Each time the risk identification process for a work order is executed, the content information of all target work orders within the currently visible time window in the government hotline work order database is retrieved to determine which target work orders are high-risk. This ensures that subsequent high-risk work orders are consistent with recent event characteristics and are highly timely.

[0025] Step 106: For each target work order, match the target work order with all group complaint events based on the content information of the target work order to obtain the matching result.

[0026] The matching results include the correspondence between the target work order and the group complaint event.

[0027] Step 108: Based on the matching results of all target work orders, determine the first number of target work orders corresponding to each group complaint event.

[0028] Step 110: For each group complaint incident, if the first number meets the preset conditions, determine the target work order corresponding to the group complaint incident as a high-risk work order.

[0029] Optionally, the preset condition can be that the first quantity is greater than or equal to a preset threshold. By setting preset conditions, high-risk work orders can be identified more accurately. For example, if there are 10,000 target work orders, 500 can be set as the preset threshold. That is, a group complaint event must contain 500 or more target work orders to be considered as having appeared in the government hotline. In this case, these 500 or more target work orders are all high-risk work orders.

[0030] Understandably, a group complaint indicates that multiple people have raised demands regarding the same or similar issues, suggesting that the issue has caused some harm and affected the relevant people. This embodiment, by setting preset conditions, can further determine the urgency of a group complaint based on the number of target work orders in a group complaint after identifying the group complaint to which the target work order belongs. This allows all target work orders in group complaint events with a number of target work orders exceeding a preset threshold to be marked as high-risk work orders, thereby improving the accuracy of high-risk work order identification.

[0031] In this embodiment, the content of multiple registered matters in the petition matters database is obtained, thereby identifying at least one group complaint event based on the content. The content information of multiple target work orders in the government hotline work order database is also obtained. For each target work order, the target work order is matched with all group complaint events based on its content information to obtain a matching result, which includes the correspondence between the target work order and the group complaint event. Furthermore, based on the matching results of all target work orders, a first number of target work orders corresponding to each group complaint event is determined. For each group complaint event, if the first number meets a preset condition, the target work order corresponding to that group complaint event is identified as a high-risk work order. As can be seen, this technical solution identifies group complaint events in the petition database and matches target work orders obtained from the government hotline work order database with each group complaint event to determine the group complaint event to which each target work order belongs. Based on the number of target work orders contained in each group complaint event, high-risk work orders are identified among the target work orders. This achieves the effect of identifying high-risk work orders in the government hotline work order database, which is conducive to promoting the handling of such work orders by staff, not only avoiding greater impact, but also improving the processing efficiency of high-risk work orders.

[0032] In one implementation, determining at least one group litigation event based on the content of multiple registered items (i.e., step 102) can be performed as follows: Steps A1-A2: Step A1: Input the content of multiple registered items into the feature extraction network to extract text features and obtain multiple first feature vectors of the first dimension.

[0033] Optionally, the feature extraction network can adopt the doc2vec (text to vector) model, which can encode the text content of registered items into vectors. The vector dimension can range from tens to hundreds, and the vector dimension can be customized according to the clustering requirements.

[0034] Step A2: Based on the first feature vectors of multiple first dimensions, cluster the registered items to obtain at least one group litigation event.

[0035] In this embodiment, by clearly defining the specific method for determining at least one group complaint event based on the content of multiple registered items, it is beneficial to the accurate execution of the risk identification process for work orders.

[0036] In one implementation, clustering of registered items based on multiple first feature vectors of the first dimension yields at least one group complaint event (i.e., step A2), which can be executed as follows: steps A21-A26: A21. Among the multiple first feature vectors of the first dimension, determine K first feature vectors as initial cluster centers.

[0037] K first feature vectors can be randomly selected as initial cluster centers. K is a positive integer.

[0038] A22. Based on the K initial cluster centers, cluster the first feature vectors of multiple first dimensions to obtain K initial clusters.

[0039] In implementation, for each first feature vector of the first dimension, the first feature vector can be assigned to the cluster represented by the nearest initial cluster center by calculating the Euclidean distance from the first feature vector to each initial cluster center, so as to obtain K initial clusters.

[0040] A23. For each initial cluster, determine the first cluster center corresponding to the initial cluster based on the mean of all first feature vectors in the initial cluster.

[0041] The first cluster center is the new cluster center of the cluster.

[0042] A24. Based on the K first cluster centers, perform clustering on the first feature vectors of multiple first dimensions to obtain K first clusters.

[0043] The specific implementation method of clustering can be referred to step A22 above, and will not be repeated here.

[0044] A25. If the K first clusters satisfy the iteration termination condition, then the iteration stops, and the K group litigation events are determined based on the K first clusters. If the K first clusters do not satisfy the iteration termination condition, then the first cluster centers and first clusters continue to be updated until the iteration termination condition is met, at which point the K group litigation events are determined based on the updated K second clusters.

[0045] Optionally, the iteration termination condition may include reaching a preset number of iterations, the cluster centers no longer changing, or the first feature vector contained in the cluster no longer changing.

[0046] A26. For each group complaint event, determine the centroid vector corresponding to the group complaint event based on the cluster center at the time of stopping iteration.

[0047] Among them, the centroid vector is used to represent group litigation events.

[0048] In this embodiment, K-means clustering is used to identify at least one group complaint event based on the content of multiple registered items. This clarifies the process for identifying group complaint events and facilitates the accurate execution of the work order risk identification process. Furthermore, since this technical solution involves a large number of events, all of which are text-based, K-means clustering can minimize the classification error rate for text vector features.

[0049] In one implementation, before determining K first feature vectors as initial cluster centers from multiple first feature vectors of the first dimension (i.e., step A21), the following steps B1-B3 can be performed: Step B1 involves sampling multiple registered items from the petition database to obtain multiple sets of sample registered items, with each set containing multiple sample registered items.

[0050] Optionally, all registered matters can be retrieved from the petition matters database with a pre-set visible time window, assuming there are M registered matters. These M registered matters are then sampled N times, requiring that the number of samplings N and the amount of data collected in each sampling be sufficiently large to ensure broad coverage. Here, M and N are both positive integers.

[0051] Step B2: For each group of sample registration items, input the content of multiple sample registration items into the classification model for text classification to obtain multiple classification results. Based on the multiple classification results, determine the number of categories corresponding to the group of sample registration items.

[0052] Alternatively, a large model can be used for classification.

[0053] Step B3: Determine the number K of cluster centers corresponding to multiple registered items based on the number of categories corresponding to the registered items in the multiple sample registration items.

[0054] Alternatively, the maximum value among the number of categories can be determined as the number of cluster centers K, or the minimum value among the number of categories can be determined as the number of cluster centers K, or the average value among the number of categories can be determined as the number of cluster centers K, and so on.

[0055] In applications, the maximum value among the number of categories can be determined as the number of cluster centers K. The purpose of taking the maximum value is to find the number of data categories corresponding to the sampling with the highest data diversity, thereby approximating the true value of K.

[0056] In this embodiment, when using K-means clustering to cluster petitions received by the petitioning department from the public, in order to determine the value of the number of cluster centers K, it is necessary to sample multiple registered items in the petition database N times. After each sampling, the content of the item is input into the large model to determine the number of cluster centers. Finally, the maximum value obtained after N sampling is selected as the value of the number of cluster centers K, without the need for manual determination. This not only makes the value of K more consistent with the actual situation of the application scenario, but also avoids the inaccuracy caused by human factors.

[0057] In one implementation, before obtaining the content information of multiple target work orders in the government hotline work order database (i.e., step 104), multiple target work orders can be determined from all work orders in the government hotline work order database according to preset data retrieval conditions.

[0058] The preset data retrieval conditions may include: the work order type conforms to a first preset type, which may include at least one of complaint, report, suggestion, consultation, or request for help; and / or, the work order item type conforms to a second preset type, which is determined based on the government services that require risk identification.

[0059] Government services that require risk identification include processing ID cards, processing business licenses, and urging rectification, etc. These are all types of matters set by the departments responsible for performing government services.

[0060] In this embodiment, since the number and types of work orders in the government hotline work order database are large, multiple target work orders can be identified from all work orders in the government hotline work order database by setting data retrieval conditions. This allows for more targeted risk identification of work orders that are likely to cause harm.

[0061] In one implementation, for each target work order, the target work order is matched with all group complaint events based on its content information to obtain the matching result (i.e., step 106), which can be executed as follows: Steps C1-C3: Step C1: Input the content information of the target work order into the feature extraction network to extract text features and obtain the second feature vector of the first dimension.

[0062] Optionally, the feature extraction network can adopt the doc2vec (text to vector) model. The doc2vec model can encode the content information of the target work order into vectors. The vector dimension ranges from tens to hundreds, and the vector dimension can be customized according to the clustering requirements.

[0063] Step C2: Calculate the matching degree between the second feature vector and the centroid vector corresponding to each group lawsuit event, and determine the maximum matching degree among all matching degrees.

[0064] The matching degree can also be called the similarity degree. The specific calculation method of the matching degree can be selected according to the specific application scenario. This application embodiment does not limit this.

[0065] Step C3: If the maximum matching degree is greater than the preset matching threshold, determine that the target work order matches the group complaint event corresponding to the maximum matching degree.

[0066] Optionally, a preset matching threshold can be set according to business needs. For example, if a lower threshold is required, the preset matching threshold can be set to 0.6; if a higher threshold is required, the preset matching threshold can be set to 0.8.

[0067] Furthermore, if the maximum matching degree of the target work order is less than the preset matching threshold, then the target work order does not match any of the identified group complaint events. In this case, the target work order can be classified as a new group complaint event.

[0068] In this embodiment, by specifying the specific method for matching each target work order with all group complaint events based on the content information of the target work order, the risk identification process of the work order is facilitated and the matching results are obtained.

[0069] Figure 2 This is a flowchart illustrating a risk identification method for work orders according to another embodiment of this application, as shown below. Figure 2 As shown, the risk identification method for work orders may include the following steps: Step 201: Obtain the content of multiple registered matters in the petition matters database, and determine at least one group litigation event based on the content of the matters.

[0070] Step 202: Based on preset data retrieval conditions, identify multiple target work orders from all work orders in the government hotline work order database.

[0071] The preset data retrieval conditions may include: the work order type conforms to a first preset type, which includes at least one of complaint, report, suggestion, consultation, and request for help; and / or, the work order item type conforms to a second preset type, which is determined based on the government services that require risk identification.

[0072] Step 203: Obtain the content information of multiple target work orders from the government hotline work order database.

[0073] Step 204: For each target work order, input the content information of the target work order into the feature extraction network to extract text features and obtain the second feature vector of the first dimension.

[0074] Step 205: Calculate the matching degree between the second feature vector and the centroid vector corresponding to each group lawsuit event, and determine the maximum matching degree among all matching degrees.

[0075] Step 206: If the maximum matching degree is greater than the preset matching threshold, determine that the target work order matches the group complaint event corresponding to the maximum matching degree.

[0076] Step 207: Based on the matching results of all target work orders, determine the first number of target work orders corresponding to each group complaint event.

[0077] The matching results include the correspondence between the target work order and the group complaint event.

[0078] Step 208: For each group complaint incident, if the first number meets the preset conditions, determine the target work order corresponding to the group complaint incident as a high-risk work order.

[0079] The specific processes of steps 201 to 208 have been described in detail in the above embodiments and will not be repeated here.

[0080] In this embodiment, the content of multiple registered matters in the petition matters database is obtained, thereby identifying at least one group complaint event based on the content. The content information of multiple target work orders in the government hotline work order database is also obtained. For each target work order, the target work order is matched with all group complaint events based on its content information to obtain a matching result, which includes the correspondence between the target work order and the group complaint event. Furthermore, based on the matching results of all target work orders, a first number of target work orders corresponding to each group complaint event is determined. For each group complaint event, if the first number meets a preset condition, the target work order corresponding to that group complaint event is identified as a high-risk work order. As can be seen, this technical solution identifies group complaint events in the petition database and matches target work orders obtained from the government hotline work order database with each group complaint event to determine the group complaint event to which each target work order belongs. Based on the number of target work orders contained in each group complaint event, high-risk work orders are identified among the target work orders. This achieves the effect of identifying high-risk work orders in the government hotline work order database, which is conducive to promoting the handling of such work orders by staff, not only avoiding greater impact, but also improving the processing efficiency of high-risk work orders.

[0081] It should be noted that the execution entity of the work order risk identification method provided in this application embodiment can be a work order risk identification device, or a control module in the work order risk identification device for executing the work order risk identification method. This application embodiment uses the example of a work order risk identification device executing a work order risk identification method to illustrate the work order risk identification device provided in this application embodiment.

[0082] Figure 3 This is a schematic diagram of the structure of a work order risk identification device provided in an embodiment of this application. Figure 3 As shown, the risk identification device for work orders includes: an acquisition and determination module 310, an acquisition module 320, a matching processing module 330, a first determination module 340, and a second determination module 350.

[0083] The acquisition and determination module 310 is used to acquire the content of multiple registered matters in the petition matters database and determine at least one group complaint event based on the content of the matters; the acquisition module 320 is used to acquire the content information of multiple target work orders in the government hotline work order database; the matching processing module 330 is used to match each target work order with all group complaint events based on the content information of the target work order to obtain the matching result; the matching result includes the correspondence between the target work order and the group complaint event; the first determination module 340 is used to determine the first number of target work orders corresponding to each group complaint event based on the matching result of all target work orders; the second determination module 350 is used to determine the target work order corresponding to each group complaint event as a high-risk work order if the first number meets the preset conditions.

[0084] In one implementation, the acquisition and determination module 310 includes: a first text feature extraction unit and a clustering processing unit.

[0085] The first text feature extraction unit is used to input the content of multiple registered items into the feature extraction network for text feature extraction to obtain multiple first feature vectors of the first dimension; the clustering processing unit is used to perform clustering processing on the registered items based on the multiple first feature vectors of the first dimension to obtain at least one group lawsuit event.

[0086] In one implementation, the clustering processing unit is specifically used for: In a dataset containing multiple first-dimensional first feature vectors, K first feature vectors are selected as initial cluster centers. Based on these K initial cluster centers, the multiple first-dimensional first feature vectors are clustered to obtain K initial clusters. For each initial cluster, the first cluster center is determined based on the mean of all first feature vectors within that initial cluster. Based on these K first cluster centers, the multiple first-dimensional first feature vectors are clustered again to obtain K first clusters. If the K first clusters satisfy the iteration termination condition, the iteration stops, and K group litigation events are determined based on these K first clusters. If the K first clusters do not satisfy the iteration termination condition, the first cluster centers and first clusters are updated until the iteration termination condition is met. Then, K group litigation events are determined based on the updated K second clusters. For each group litigation event, the centroid vector corresponding to the event is determined based on the cluster centers at the time of iteration termination. The centroid vector is used to represent the group litigation event.

[0087] In one implementation, the risk identification device for work orders further includes: a sampling module, a classification and determination module, and a third determination module.

[0088] The sampling module is used to perform multiple sampling processes on multiple registered items in the petition database before determining K first feature vectors from multiple first feature vectors as initial cluster centers, resulting in multiple sets of sample registered items; each set of sample registered items includes multiple sample registered items; the classification and determination module is used to input the content of multiple sample registered items into the classification model for text classification for each set of sample registered items, resulting in multiple classification results; based on the multiple classification results, the number of categories corresponding to the set of sample registered items is determined; the third determination module is used to determine the number K of cluster centers corresponding to the multiple registered items based on the number of categories corresponding to the multiple sets of sample registered items.

[0089] In one implementation, the matching processing module 330 includes: a second text feature extraction unit, a calculation and determination unit, and a determination unit.

[0090] The second text feature extraction unit is used to input the content information of the target work order into the feature extraction network for text feature extraction to obtain the second feature vector of the first dimension; the calculation and determination unit is used to calculate the matching degree between the second feature vector and the centroid vector corresponding to each group complaint event, and determine the maximum matching degree among all matching degrees; the determination unit is used to determine that the target work order matches the group complaint event corresponding to the maximum matching degree when the maximum matching degree is greater than the preset matching threshold.

[0091] In one implementation, the risk identification device for work orders also includes a fourth determination module.

[0092] The fourth determination module is used to determine multiple target work orders from all work orders in the government hotline work order database before obtaining the content information of multiple target work orders in the database, based on preset data retrieval conditions. The preset data retrieval conditions include: the work order type conforms to a first preset type, which includes at least one of complaint, report, suggestion, consultation, and assistance; and / or, the matter type of the work order conforms to a second preset type, which is determined based on government services that require risk identification.

[0093] In this embodiment, the content of multiple registered matters in the petition matters database is obtained, thereby identifying at least one group complaint event based on the content. The content information of multiple target work orders in the government hotline work order database is also obtained. For each target work order, the target work order is matched with all group complaint events based on its content information to obtain a matching result, which includes the correspondence between the target work order and the group complaint event. Furthermore, based on the matching results of all target work orders, a first number of target work orders corresponding to each group complaint event is determined. For each group complaint event, if the first number meets a preset condition, the target work order corresponding to that group complaint event is identified as a high-risk work order. As can be seen, this technical solution identifies group complaint events in the petition database and matches target work orders obtained from the government hotline work order database with each group complaint event to determine the group complaint event to which each target work order belongs. Based on the number of target work orders contained in each group complaint event, high-risk work orders are identified among the target work orders. This achieves the effect of identifying high-risk work orders in the government hotline work order database, which is conducive to promoting the handling of such work orders by staff, not only avoiding greater impact, but also improving the processing efficiency of high-risk work orders.

[0094] The risk identification device for work orders in this application embodiment can be a device, or a component, integrated circuit, or chip in a terminal. The device can be a mobile electronic device or a non-mobile electronic device. For example, mobile electronic devices can be mobile phones, tablets, laptops, PDAs, in-vehicle electronic devices, wearable devices, ultra-mobile personal computers (UMPCs), netbooks, or personal digital assistants (PDAs), etc., while non-mobile electronic devices can be servers, network-attached storage (NAS), personal computers (PCs), televisions (TVs), ATMs, or self-service machines, etc. This application embodiment does not impose specific limitations.

[0095] The risk identification device for work orders in this application embodiment can be a device with an operating system. This operating system can be Android, iOS, or other possible operating systems; this application embodiment does not specifically limit it.

[0096] The risk identification device for work orders provided in this application embodiment can achieve... Figures 1 to 2 The various processes implemented in the method embodiments are not described in detail here to avoid repetition.

[0097] Based on the same technical concept, embodiments of this application also provide an electronic device for executing the aforementioned risk identification method for work orders. Figure 4 This is a schematic diagram of the structure of an electronic device to implement various embodiments of this application. The electronic device can vary significantly due to differences in configuration or performance, and may include a processor 410, a communications interface 420, a memory 430, and a communication bus 440. The processor 410, communications interface 420, and memory 430 communicate with each other via the communication bus 440. The processor 410 can call a computer program stored in the memory 430 and executable on the processor 410 to perform the following steps: The system retrieves the content of multiple registered petitions from the petition database and identifies at least one group complaint incident based on the content. It also retrieves the content information of multiple target work orders from the government hotline work order database. For each target work order, it matches it with all group complaint incidents based on its content information, obtaining matching results. The matching results include the correspondence between target work orders and group complaint incidents. Based on the matching results of all target work orders, it determines the first number of target work orders corresponding to each group complaint incident. For each group complaint incident, if the first number meets preset conditions, the target work order corresponding to the group complaint incident is identified as a high-risk work order.

[0098] In this embodiment, the content of multiple registered matters in the petition matters database is obtained, thereby identifying at least one group complaint event based on the content. The content information of multiple target work orders in the government hotline work order database is also obtained. For each target work order, the target work order is matched with all group complaint events based on its content information to obtain a matching result, which includes the correspondence between the target work order and the group complaint event. Furthermore, based on the matching results of all target work orders, a first number of target work orders corresponding to each group complaint event is determined. For each group complaint event, if the first number meets a preset condition, the target work order corresponding to that group complaint event is identified as a high-risk work order. As can be seen, this technical solution identifies group complaint events in the petition database and matches target work orders obtained from the government hotline work order database with each group complaint event to determine the group complaint event to which each target work order belongs. Based on the number of target work orders contained in each group complaint event, high-risk work orders are identified among the target work orders. This achieves the effect of identifying high-risk work orders in the government hotline work order database, which is conducive to promoting the handling of such work orders by staff, not only avoiding greater impact, but also improving the processing efficiency of high-risk work orders.

[0099] The specific execution steps can be found in the various steps of the above-mentioned work order risk identification method embodiment, and can achieve the same technical effect. To avoid repetition, they will not be repeated here.

[0100] It should be noted that the electronic devices in the embodiments of this application include: servers, terminals, or other devices besides terminals.

[0101] The above electronic device structure does not constitute a limitation on the electronic device. An electronic device may include more or fewer components than illustrated, or combine certain components, or arrange them differently. For example, an input unit may include a Graphics Processing Unit (GPU) and a microphone, and a display unit may use a liquid crystal display (LCD), organic light-emitting diode (OLED), or other similar display panels. User input units include at least one of a touch panel and other input devices. A touch panel is also called a touchscreen. Other input devices may include, but are not limited to, physical keyboards, function keys (such as volume control buttons, power buttons, etc.), trackballs, mice, and joysticks, which will not be elaborated further here.

[0102] Memory can be used to store software programs and various data. Memory can primarily include a first storage area for storing programs or instructions and a second storage area for storing data. The first storage area can store the operating system, application programs or instructions required for at least one function (such as sound playback, image playback, etc.). Furthermore, memory can include volatile memory or non-volatile memory, or both. Non-volatile memory can be read-only memory (ROM), programmable read-only memory (PROM), erasable programmable read-only memory (EPROM), electrically erasable programmable read-only memory (EEPROM), or flash memory. Volatile memory can be random access memory (RAM), static random access memory (SRAM), dynamic random access memory (DRAM), synchronous dynamic random access memory (SDRAM), double data rate synchronous dynamic random access memory (DDRSDRAM), enhanced synchronous dynamic random access memory (ESDRAM), synchronous linked dynamic random access memory (Synchlink DRAM, SLDRAM), and direct memory bus RAM (DRRAM).

[0103] The processor may include one or more processing units; optionally, the processor integrates an application processor and a modem processor, wherein the application processor mainly handles operations related to the operating system, user interface, and applications, while the modem processor mainly handles wireless communication signals, such as a baseband processor. It is understood that the aforementioned modem processor may also not be integrated into the processor.

[0104] This application also provides a computer-readable storage medium for storing computer-executable instructions. When the computer-executable instructions are executed by a processor, they implement the various processes of the above-described work order risk identification method embodiment and achieve the same technical effect. To avoid repetition, they will not be described again here.

[0105] The processor is the processor in the electronic device described in the above embodiments. The readable storage medium includes computer-readable storage media, such as computer read-only memory (ROM), random access memory (RAM), magnetic disk, or optical disk.

[0106] This application also provides a computer program product, which includes a computer program. When the computer program is executed by a processor, it implements the various processes of the above-described work order risk identification method embodiment and can achieve the same technical effect. To avoid repetition, it will not be described again here.

[0107] This application embodiment also provides a chip, which includes a processor and a communication interface. The communication interface is coupled to the processor. The processor is used to run programs or instructions to implement the various processes of the above-described work order risk identification method embodiment, and can achieve the same technical effect. To avoid repetition, it will not be described again here.

[0108] It should be understood that the chip mentioned in the embodiments of this application may also be referred to as a system-on-a-chip, system chip, chip system, or system-on-a-chip, etc.

[0109] It should be noted that, in this document, the terms "comprising," "including," or any other variations thereof are intended to cover non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements includes not only those elements but also other elements not expressly listed, or elements inherent to such a process, method, article, or apparatus. Without further limitations, an element defined by the phrase "comprising one..." does not exclude the presence of other identical elements in the process, method, article, or apparatus that includes that element. Furthermore, it should be noted that the scope of the methods and apparatuses in the embodiments of this application is not limited to performing functions in the order shown or discussed, but may also include performing functions substantially simultaneously or in the reverse order, depending on the functions involved. For example, the described methods may be performed in a different order than described, and various steps may be added, omitted, or combined. Additionally, features described with reference to certain examples may be combined in other examples.

[0110] Through the above description of the embodiments, those skilled in the art can clearly understand that the methods of the above embodiments can be implemented by means of software plus necessary general-purpose hardware platforms. Of course, they can also be implemented by hardware, but in many cases the former is a better implementation method. Based on this understanding, the technical solution of this application, in essence, or the part that contributes to the prior art, can be embodied in the form of a software product. This computer software product is stored in a storage medium (such as ROM / RAM, magnetic disk, optical disk) and includes several instructions to cause a terminal (which may be a mobile phone, computer, server, air conditioner, or network device, etc.) to execute the methods described in the various embodiments of this application.

[0111] The embodiments of this application have been described above with reference to the accompanying drawings. However, this application is not limited to the specific embodiments described above. The specific embodiments described above are merely illustrative and not restrictive. Those skilled in the art can make many other forms under the guidance of this application without departing from the spirit and scope of the claims, and all of these forms are within the protection scope of this application.

Claims

1. A risk identification method for work orders, characterized in that, include: Obtain the details of multiple registered matters in the petition database, and determine at least one group litigation incident based on the details of the matters; Retrieve the content information of multiple target work orders from the government hotline work order database; For each target work order, based on the content information of the target work order, the target work order is matched with all group complaint events to obtain a matching result; the matching result includes the correspondence between the target work order and the group complaint event. Based on the matching results of all target work orders, determine the first number of target work orders corresponding to each group complaint event; For each of the group complaint events, if the first number meets the preset conditions, the target work order corresponding to the group complaint event is determined to be a high-risk work order.

2. The method according to claim 1, characterized in that, The determination of at least one group litigation incident based on the content of the matter includes: The content of each of the registered items is input into a feature extraction network for text feature extraction, resulting in multiple first feature vectors of the first dimension. Based on the first feature vectors of the multiple first dimensions, the registered items are clustered to obtain the at least one group lawsuit event.

3. The method according to claim 2, characterized in that, The step of clustering the registered items based on the first feature vectors of the multiple first dimensions to obtain the at least one group complaint event includes: Among the multiple first feature vectors of the first dimension, K first feature vectors are determined as initial cluster centers; Based on K initial cluster centers, the first feature vectors of the multiple first dimensions are clustered to obtain K initial clusters; For each initial cluster, the first cluster center corresponding to the initial cluster is determined based on the mean of all first feature vectors in the initial cluster. Based on the K first cluster centers, the first feature vectors of the multiple first dimensions are clustered to obtain K first clusters; If the K first clusters satisfy the iteration termination condition, the iteration stops, and K group litigation events are determined based on the K first clusters; if the K first clusters do not satisfy the iteration termination condition, the first cluster centers and the first clusters continue to be updated until the iteration termination condition is satisfied, and K group litigation events are determined based on the updated K second clusters. For each of the group lawsuit events, the centroid vector corresponding to the group lawsuit event is determined based on the cluster center at the time of stopping iteration; the centroid vector is used to characterize the group lawsuit event.

4. The method according to claim 3, characterized in that, Before determining K first feature vectors as initial cluster centers from the plurality of first feature vectors of the first dimension, the method further includes: Multiple sampling processes are performed on the multiple registered items in the petition database to obtain multiple sets of sample registered items; each set of sample registered items includes multiple sample registered items; For each group of sample registration items, the content of the multiple sample registration items is input into a classification model for text classification to obtain multiple classification results; based on the multiple classification results, the number of categories corresponding to the group of sample registration items is determined. Based on the number of categories corresponding to the multiple sets of sample registration items, the number of cluster centers K corresponding to the multiple registered items is determined.

5. The method according to claim 3, characterized in that, For each target work order, based on its content information, the target work order is matched with all group complaint events to obtain matching results, including: The content information of the target work order is input into the feature extraction network for text feature extraction to obtain the second feature vector of the first dimension. Calculate the matching degree between the second feature vector and the centroid vector corresponding to each group lawsuit event, and determine the maximum matching degree among all matching degrees; If the maximum matching degree is greater than a preset matching threshold, the target work order is determined to match the group complaint event corresponding to the maximum matching degree.

6. The method according to claim 1, characterized in that, Before obtaining the content information of multiple target work orders in the government hotline work order database, the method further includes: Based on preset data retrieval conditions, the multiple target work orders are determined from all work orders in the government hotline work order database; The preset data retrieval conditions include: the work order type conforms to a first preset type, the first preset type includes at least one of complaint, report, suggestion, consultation, and assistance; and / or, the work order item type conforms to a second preset type, the second preset type is determined based on government services that require risk identification.

7. A risk identification device for work orders, characterized in that, include: The acquisition and determination module is used to acquire the content of multiple registered matters in the petition matters database, and determine at least one group litigation event based on the content of the matters; The acquisition module is used to acquire the content information of multiple target work orders in the government hotline work order database; The matching processing module is used to match each target work order with all group complaint events based on the content information of the target work order, and obtain a matching result; the matching result includes the correspondence between the target work order and the group complaint events. The first determination module is used to determine the first number of target work orders corresponding to each group complaint event based on the matching results of all target work orders. The second determining module is used to determine, for each of the group complaint events, the target work order corresponding to the group complaint event as a high-risk work order if the first quantity meets the preset conditions.

8. An electronic device, characterized in that, include: processor; as well as A memory configured to store computer-executable instructions configured to be executed by the processor to implement the risk identification method for work orders as described in any one of claims 1-6.

9. A computer-readable storage medium, characterized in that, The computer-readable storage medium is used to store computer-executable instructions, which, when executed by a processor, implement the risk identification method for work orders as described in any one of claims 1-6.

10. A computer program product, characterized in that, It includes a computer program that, when executed by a processor, implements the risk identification method for work orders as described in any one of claims 1-6.

Images