Survey Task Allocation Method, Device, Electronic Device and Storage Medium

Abstract

An embodiment of the present application provides a survey task allocation method, device, electronic device, and storage medium, belonging to the field of artificial intelligence technology. The method includes: obtaining original case data and business data; performing case simulation analysis based on the original case data and business data to obtain preliminary survey data; performing identification processing on target cases in a preset area through the preliminary survey data and a preset case identification model to obtain target survey data of the target cases and case processing data of the target cases; performing data analysis on the target survey data and the case processing data through a pre-trained data analysis model to obtain predicted survey indicators; performing measurement processing on the target survey data according to the predicted survey indicators and reference survey indicators to obtain survey manpower data; and allocating survey tasks according to the survey manpower data. The embodiment of the present application can improve the accuracy of survey task dispatching.

Description

Technical Field

[0001] This application relates to the field of artificial intelligence technology, and particularly to a survey task allocation method, apparatus, electronic device, and storage medium. Background Art

[0002] Currently, in various survey tasks, the calculation of the manpower requirements for surveys usually relies on the experience of the business department. For example, when there are more survey cases, additional survey manpower is added, and when there are fewer survey cases, the survey manpower is reduced. This approach often results in uneven distribution of survey tasks and situations such as redundant survey personnel. In addition, the manpower requirements of surveyors are often affected by various random interference factors. Therefore, using this survey task allocation method based on the experience of the business department for survey manpower scheduling often leads to low accuracy in dispatching survey tasks. Therefore, how to provide a survey task allocation method that can improve the accuracy of dispatching survey tasks has become an urgent technical problem to be solved. Summary of the Invention

[0003] The main purpose of the embodiments of this application is to propose a survey task allocation method, apparatus, electronic device, and storage medium, aiming to improve the accuracy of dispatching survey tasks.

[0004] To achieve the above object, the first aspect of the embodiments of this application proposes a survey task allocation method, and the method includes:

[0005] Obtain original case data and business data;

[0006] Conduct case simulation analysis based on the original case data and the business data to obtain preliminary survey data;

[0007] Perform identification processing on target cases in a preset area through the preliminary survey data and a preset case identification model to obtain the target survey data of the target cases and the case processing data of the target cases;

[0008] Perform data analysis on the target survey data and the case processing data through a pre-trained data analysis model to obtain predicted survey indicators;

[0009] Perform measurement processing on the target survey data according to the predicted survey indicators and reference survey indicators to obtain survey manpower data;

[0010] Allocate survey tasks according to the survey manpower data.

[0011] In some embodiments, the step of obtaining original case data and business data includes:

[0012] Obtain target map data;

[0013] Divide the target map into multiple grid maps according to a preset ratio and the target map data;

[0014] Collect the case situations that occurred in each of the grid maps during a first preset time period to obtain original case data and business data.

[0015] In some embodiments, the step of performing case simulation analysis based on the original case data and the business data to obtain preliminary survey data includes:

[0016] Perform data preprocessing on the original case data and the business data through Poisson distribution to obtain target case situation data corresponding to the target case;

[0017] Use the Monte Carlo algorithm and the target case situation data to perform simulation analysis on the target case to obtain surveyor scheduling data;

[0018] Output a first scheduling instruction according to the surveyor scheduling data; the first scheduling instruction is used to instruct a surveyor to perform simulation patrol on the target case in a preset area to obtain simulation patrol data;

[0019] Perform fusion processing on the surveyor scheduling data and the simulation patrol data to obtain the preliminary survey data.

[0020] In some embodiments, the step of performing identification processing on the target case in a preset area through the preliminary survey data and a preset case identification model to obtain the target survey data of the target case and the case processing data of the target case includes:

[0021] Obtain the number of cases of the target case existing in a preset area during a second preset time period through a preset case identification model;

[0022] If the number of cases is greater than 0, perform screening processing on the preliminary survey data according to a preset priority order to obtain the target survey data of the target case;

[0023] Dispatch a corresponding target surveyor according to the target survey data and instruct the target surveyor to process the target case to obtain the case processing data of the target case.

[0024] In some embodiments, the step of performing data analysis on the target survey data and the case processing data through a pre-trained data analysis model to obtain predicted survey indicators includes:

[0025] Perform feature extraction on the target survey data and the case processing data respectively through the convolutional layer of the data analysis model to obtain a target survey vector and a case processing vector;

[0026] The pooling layer of the data analysis model respectively performs pooling processing on the target survey vector and the case processing vector to obtain a survey feature vector and a case processing feature vector;

[0027] The fully connected layer of the data analysis model splices the survey feature vector and the case processing feature vector to obtain a comprehensive survey vector;

[0028] The preset function of the data analysis model and the survey category label are used to activate the comprehensive survey vector to obtain a predicted survey index corresponding to each survey category label.

[0029] In some embodiments, the calculating and processing the target survey data according to the predicted survey index and the reference survey index to obtain survey manpower data includes:

[0030] Generating fine-tuning data according to the magnitude relationship between the predicted survey index and the reference survey index;

[0031] Inputting the fine-tuning data and the target survey data into a preset deep learning model;

[0032] Generating a fine-tuning index of the deep learning model according to the fine-tuning data;

[0033] Performing fine-tuning processing on the target survey data through the deep learning model and the fine-tuning index to obtain survey data to be calculated;

[0034] Performing manpower calculation processing on the survey data to be calculated through the calculation function of the deep learning model and the environmental impact factor to obtain the survey manpower data.

[0035] In some embodiments, before the step of performing data analysis on the target survey data and the case processing data through a pre-trained data analysis model to obtain a predicted survey index, the method further includes pre-training the data analysis model, specifically including:

[0036] Obtaining sample case data;

[0037] Performing simulation analysis on the sample case data to obtain sample survey data and sample case processing data;

[0038] Inputting the sample survey data and the sample case processing data into the data analysis model;

[0039] Calculating sample survey indexes through the loss function of the data analysis model;

[0040] Optimizing the loss function of the data analysis model according to the sample survey indexes to update the data analysis model.

[0041] To achieve the above object, a second aspect of the embodiments of the present application provides a survey task allocation device, which includes:

[0042] A data acquisition module, configured to acquire original case data and service data;

[0043] A simulation analysis module, configured to perform case simulation analysis based on the original case data and the service data to obtain preliminary survey data;

[0044] An identification module, configured to perform identification processing on target cases in a preset area through the preliminary survey data and a preset case identification model to obtain target survey data of the target cases and case processing data of the target cases;

[0045] A data analysis module, configured to perform data analysis on the target survey data and the case processing data through a pre-trained data analysis model to obtain predicted survey indicators;

[0046] A human resource calculation module, configured to perform calculation processing on the target survey data according to the predicted survey indicators and reference survey indicators to obtain survey human resource data;

[0047] A task allocation module, configured to allocate survey tasks according to the survey human resource data.

[0048] To achieve the above object, a third aspect of the embodiments of the present application provides an electronic device, which includes a memory, a processor, a program stored on the memory and executable on the processor, and a data bus for realizing connection communication between the processor and the memory. When the program is executed by the processor, the method described in the first aspect above is implemented.

[0049] To achieve the above object, a fourth aspect of the embodiments of the present application provides a storage medium, which is a computer-readable storage medium for computer-readable storage. The storage medium stores one or more programs, and the one or more programs can be executed by one or more processors to implement the method described in the first aspect above.

[0050] The survey task allocation method, device, electronic device and storage medium proposed in this application obtain original case data and business data; perform case simulation analysis based on the original case data and business data to obtain preliminary survey data, which can more conveniently simulate and estimate the manpower requirements of surveyors corresponding to the case; furthermore, identify and process target cases in a preset area through the preliminary survey data and a preset case identification model to obtain the target survey data of the target case and the case processing data of the target case. At the same time, perform data analysis on the target survey data and case processing data through a pre-trained data analysis model to obtain predicted survey indicators, which can quickly simulate the working conditions of surveyors. Finally, based on the predicted survey indicators and reference survey indicators, calculate and process the target survey data to obtain survey manpower data, and allocate survey tasks according to the survey manpower data, which can improve the speed and accuracy of survey manpower calculation, thereby improving the accuracy of survey task dispatching. Brief Description of the Drawings

[0051] Figure 1 is a flowchart of the survey task allocation method provided by an embodiment of this application;

[0052] Figure 2 is Figure 1 a flowchart of step S101 in

[0053] Figure 3 is Figure 1 a flowchart of step S102 in

[0054] Figure 4 is Figure 1 a flowchart of step S103 in

[0055] Figure 5 is another flowchart of the survey task allocation method provided by an embodiment of this application;

[0056] Figure 6 is Figure 1 a flowchart of step S104 in

[0057] Figure 7 is Figure 1 a flowchart of step S105 in

[0058] Figure 8 is a schematic structural diagram of the survey task allocation device provided by an embodiment of this application;

[0059] Figure 9 is a schematic hardware structure diagram of the electronic device provided by an embodiment of this application. Detailed Embodiments

[0060] In order to make the objectives, technical solutions and advantages of the present application more clear and understandable, the present application will be further described in detail below with reference to the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are only used to explain the present application and are not used to limit the present application.

[0061] It should be noted that although the functional modules are divided in the device schematic diagram and the logical sequence is shown in the flowchart, in some cases, the steps shown or described may be executed in a different module division in the device or a different sequence in the flowchart. The terms "first", "second", etc. in the description, claims and the above-mentioned drawings are used to distinguish similar objects and do not necessarily need to describe a specific order or sequence.

[0062] Unless otherwise defined, all technical and scientific terms used herein have the same meaning as commonly understood by those skilled in the technical field to which this application belongs. The terms used herein are only for the purpose of describing the embodiments of this application and are not intended to limit this application.

[0063] First, several terms involved in the present application are analyzed:

[0064] Artificial intelligence (AI): It is a new technical science that studies, develops theories, methods, technologies and application systems for simulating, extending and expanding human intelligence; artificial intelligence is a branch of computer science. Artificial intelligence attempts to understand the essence of intelligence and produce a new intelligent machine that can respond in a way similar to human intelligence. The research in this field includes robots, speech recognition, image recognition, natural language processing and expert systems, etc. Artificial intelligence can simulate the information process of human consciousness and thinking. Artificial intelligence also uses digital computers or machines controlled by digital computers to simulate, extend and expand human intelligence, perceive the environment, acquire knowledge and use knowledge to obtain the best results in terms of theories, methods, technologies and application systems.

[0065] Natural language processing (NLP): NLP uses computers to process, understand and apply human languages (such as Chinese, English, etc.). NLP is a branch of artificial intelligence and an interdisciplinary subject of computer science and linguistics, and is often referred to as computational linguistics. Natural language processing includes syntactic analysis, semantic analysis, discourse understanding, etc. Natural language processing is commonly used in technical fields such as machine translation, handwritten and printed character recognition, speech recognition and text-to-speech conversion, information intention recognition, information extraction and filtering, text classification and clustering, public opinion analysis and opinion mining, etc. It involves data mining, machine learning, knowledge acquisition, knowledge engineering, artificial intelligence research related to language processing and linguistic research related to language computing, etc.

[0066] Information Extraction (NER): A text processing technology that extracts factual information such as entities, relationships, and events of a specified type from natural language text and forms structured data output. Information extraction is a technology for extracting specific information from text data. Text data is composed of some specific units, such as sentences, paragraphs, and passages. Text information is precisely composed of some small specific units, such as characters, words, phrases, sentences, paragraphs, or combinations of these specific units. Extracting noun phrases, personal names, place names, etc. from text data are all text information extraction. Of course, the information extracted by text information extraction technology can be various types of information.

[0067] Poisson distribution: A discrete probability distribution commonly seen in statistics and probability theory. The Poisson distribution is one of the most important discrete distributions and often appears when X represents the number of events occurring within a certain period of time or space. In practical cases, when a random event, such as the calls received by a telephone exchange, the passengers arriving at a bus stop, the particles emitted by a radioactive substance, the white blood cells in a certain area under a microscope, etc., occur randomly and independently at a fixed average instantaneous rate λ (or density), then the number of times or the number of occurrences of this event within a unit time (area or volume) approximately follows the Poisson distribution P(λ).

[0068] The Monte Carlo method, also known as the random sampling or statistical test method, belongs to a branch of computational mathematics. It was developed in the mid-1940s to adapt to the development of the atomic energy industry at that time. Traditional empirical methods were difficult to obtain satisfactory results because they could not approximate the real physical process, while the Monte Carlo method could realistically simulate the actual physical process, so the solutions to problems were very consistent with the reality and could obtain very satisfactory results. This is also a computational method based on probability and statistical theory, which is a method of using random numbers (or more commonly, pseudo-random numbers) to solve many computational problems. The problem to be solved is associated with a certain probability model, and statistical simulation or sampling is implemented using an electronic computer to obtain an approximate solution to the problem.

[0069] The three main steps in the problem-solving process of the Monte Carlo method:

[0070] (1) Construct or describe the probability process

[0071] For problems that are inherently random, such as particle transport problems, the main task is to correctly describe and simulate this probability process. For deterministic problems that are not inherently random, such as calculating definite integrals, an artificial probability process must be constructed in advance, and some of its parameters are exactly the solutions to the problems we want. That is, we need to transform problems without random properties into problems with random properties.

[0072] (2) Realize sampling from known probability distributions

[0073] After constructing a probability model, since various probability models can be regarded as composed of various probability distributions, generating random variables (or random vectors) with known probability distributions becomes the basic means to implement Monte Carlo method simulation experiments. This is also the reason why the Monte Carlo method is called random sampling. The simplest, most basic, and most important probability distribution is the uniform distribution (or rectangular distribution) on (0, 1). Random numbers are random variables with this uniform distribution. A random number sequence is a simple subsample of the population with this distribution, that is, a sequence of independent random variables with this distribution. The problem of generating random numbers is the problem of sampling from this distribution. On a computer, random numbers can be generated physically, but it is expensive, non-repeatable, and inconvenient to use. Another method is to generate them using mathematical recurrence formulas. The sequences generated in this way are different from true random number sequences, so they are called pseudo-random numbers, or pseudo-random number sequences. However, after various statistical tests, it is shown that they have similar properties to true random numbers or random number sequences, so they can be used as true random numbers. There are various methods for random sampling from known distributions. Different from sampling from the uniform distribution on (0, 1), these methods are all realized by means of random sequences, that is, they are all based on generating random numbers. Thus, random numbers are the basic tools for us to implement Monte Carlo simulation.

[0074] (3) Establish various estimators

[0075] Generally speaking, after constructing a probability model and being able to sample from it, that is, after implementing the simulation experiment, we need to determine a random variable as the solution to the problem we want, and we call it an unbiased estimator. Establishing various estimators is equivalent to examining and registering the results of the simulation experiment to obtain the solution to the problem.

[0076] Deep Learning (DL): It is a new research direction in the field of Machine Learning (ML). Deep learning is a type of machine learning, and machine learning is the necessary path to achieve artificial intelligence. The concept of deep learning stems from the research of artificial neural networks. A multi-layer perceptron with multiple hidden layers is a deep learning structure. Deep learning forms more abstract high-level representations (such as attribute categories or features) by combining low-level features to discover the distributed feature representations of data. The motivation for studying deep learning is to build a neural network that simulates the human brain for analysis and learning. It mimics the mechanism of the human brain to interpret data, such as images, sounds, and text, etc.

[0077] Deep learning is to learn the internal laws and representation levels of sample data. The information obtained during these learning processes is very helpful for the interpretation of data such as text, images, and sounds. Its ultimate goal is to enable machines to have the ability of analysis and learning like humans, and be able to recognize data such as text, images, and sounds. Deep learning is a complex machine learning algorithm, and the effects achieved in speech and image recognition far exceed the previous related technologies.

[0078] Currently, in various survey tasks, the measurement of the manpower requirement for surveys usually relies on the experience of the business department. For example, when there are more survey cases, the survey manpower is increased, and when there are fewer survey cases, the survey manpower is reduced. This method often causes uneven distribution of survey tasks and situations such as redundant survey personnel. In addition, the manpower requirement of surveyors is often affected by various random interference factors. Therefore, using this method of survey task allocation based on the experience of the business department for survey manpower scheduling often results in low accuracy of survey task dispatching. Therefore, how to provide a survey task allocation method that can improve the accuracy of survey task dispatching has become a technical problem to be solved urgently.

[0079] Based on this, the embodiments of the present application provide a survey task allocation method, device, electronic device, and storage medium, aiming to improve the accuracy of survey task dispatching.

[0080] The survey task allocation method, device, electronic device, and storage medium provided by the embodiments of the present application are specifically described through the following embodiments. First, the survey task allocation method in the embodiments of the present application is described.

[0081] The embodiments of the present application can acquire and process relevant data based on artificial intelligence technology. Among them, Artificial Intelligence (AI) is to use a digital computer or a machine controlled by a digital computer to simulate, extend, and expand human intelligence, sense the environment, acquire knowledge, and use knowledge to obtain the best results of theory, method, technology, and application system.

[0082] The basic technologies of artificial intelligence generally include technologies such as sensors, dedicated artificial intelligence chips, cloud computing, distributed storage, big data processing technologies, operation / interaction systems, and mechatronics. The software technologies of artificial intelligence mainly include several major directions such as computer vision technology, robotics, biometric technology, speech processing technology, natural language processing technology, and machine learning / deep learning.

[0083] The survey task allocation method provided by the embodiments of this application relates to the field of artificial intelligence technology. The survey task allocation method provided by the embodiments of this application can be applied to a terminal, or can be applied to a server side, or can also be software running on a terminal or a server side. In some embodiments, the terminal can be a smart phone, a tablet computer, a laptop computer, a desktop computer, etc.; the server side can be configured as an independent physical server, or can be configured as a server cluster or a distributed system composed of multiple physical servers, or can also be configured as a cloud server providing basic cloud computing services such as cloud services, cloud databases, cloud computing, cloud functions, cloud storage, network services, cloud communications, middleware services, domain name services, security services, CDN, and big data and artificial intelligence platforms; the software can be an application that implements the survey task allocation method, etc., but is not limited to the above forms.

[0084] This application can be used in many general or special computer system environments or configurations. For example: personal computers, server computers, handheld or portable devices, tablet devices, multi-processor systems, microprocessor-based systems, set-top boxes, programmable consumer electronic devices, network PCs, small computers, large computers, distributed computing environments including any of the above systems or devices, and so on. This application can be described in the general context of computer-executable instructions executed by a computer, such as program modules. Generally, program modules include routines, programs, objects, components, data structures, etc. that perform specific tasks or implement specific abstract data types. This application can also be practiced in a distributed computing environment, where tasks are performed by remote processing devices connected through a communication network. In a distributed computing environment, program modules can be located in local and remote computer storage media including storage devices.

[0085] Figure 1 is an optional flowchart of the survey task allocation method provided by the embodiments of this application, Figure 1 The method in may include but is not limited to steps S101 to S106.

[0086] Step S101, obtain the original case data and business data;

[0087] Step S102, perform case simulation analysis according to the original case data and business data to obtain preliminary survey data;

[0088] Step S103, identify and process target cases in a preset area through preliminary survey data and a preset case identification model to obtain target survey data of the target cases and case processing data of the target cases;

[0089] Step S104, perform data analysis on the target survey data and case processing data through a pre-trained data analysis model to obtain predicted survey indicators;

[0090] Step S105, perform measurement processing on the target survey data according to the predicted survey indicators and reference survey indicators to obtain survey manpower data;

[0091] Step S106, allocate survey tasks according to the survey manpower data.

[0092] In steps S101 to S106 shown in the embodiments of the present application, by obtaining original case data and business data; performing case simulation analysis according to the original case data and business data to obtain preliminary survey data, it is possible to conveniently simulate and estimate the manpower requirements of the surveyors corresponding to the cases. By identifying and processing target cases in a preset area through preliminary survey data and a preset case identification model to obtain target survey data of the target cases and case processing data of the target cases, and by performing data analysis on the target survey data and case processing data through a pre-trained data analysis model to obtain predicted survey indicators, the working conditions of the surveyors can be simulated relatively quickly. Finally, according to the predicted survey indicators and reference survey indicators, performing measurement processing on the target survey data to obtain survey manpower data, and allocating survey tasks according to the survey manpower data can improve the speed and accuracy of survey manpower measurement, thereby improving the accuracy of survey task dispatching.

[0093] Please refer to Figure 2 In some embodiments, step S101 may include but is not limited to steps S201 to S203:

[0094] Step S201, obtain target map data;

[0095] Step S202, divide the target map into multiple grid maps according to a preset ratio and the target map data;

[0096] Step S203, collect the case situations occurring in each grid map within a first preset time period to obtain original case data and business data.

[0097] In step S201 of some embodiments, data can be purposefully crawled by writing a web crawler after setting the data source to obtain target map data; alternatively, the target map data can be retrieved from a preset map database, or the target map data can be obtained by other means, which is not limited thereto. It should be noted that the target map data is national map data or the map data of a certain area where a survey task needs to be assigned.

[0098] In step S202 of some embodiments, the target map can be a national map, and the preset ratio can be set according to actual needs. For example, first obtain the actual geographical data of the national map, and then determine that the preset ratio is 1:100. According to the ratio of 1:100, the target map is scaled into a corresponding small map, and the target map is cut into multiple 1km*1km grid maps, so as to realize the gridification of the target map.

[0099] In step S203 of some embodiments, data can be crawled by means of a web crawler, etc., and the case data of traffic cases in each grid map within a preset first preset time period is collected from multiple preset data sources to obtain corresponding original case data and business data. Among them, the original case data includes the number of vehicle insurance cases, the case handling records of traffic cases, the beginning and end of the case, etc., and the business data includes the basic data of surveyors, such as the number of surveyors, gender, age, etc. Through this method, the spatio-temporal distribution data of vehicle insurance cases can be conveniently counted.

[0100] Please refer to Figure 3 , in some embodiments, step S102 may include but is not limited to steps S301 to S304:

[0101] Step S301, perform data preprocessing on the original case data and business data through Poisson distribution to obtain target case data corresponding to the target case;

[0102] Step S302, use the Monte Carlo algorithm and the target case data to perform simulation analysis on the target case to obtain surveyor scheduling data;

[0103] Step S303, output a first scheduling instruction according to the surveyor scheduling data; the first scheduling instruction is used to instruct the surveyor to perform simulated patrols on the target cases in the preset area to obtain simulated patrol data;

[0104] Step S304, perform fusion processing on the surveyor scheduling data and the simulated patrol data to obtain preliminary survey data.

[0105] In step S301 of some embodiments, first, the original case data is screened according to preset keywords such as car crash, traffic accident, drunk driving, car accident, etc., and the target cases involving car insurance accidents are extracted. At the same time, data such as the number and location of surveyors in the business data are extracted, and the average occurrence times λ of the target cases in each grid map are calculated using the probability function of the Poisson distribution and the car insurance case data. The probability function of the Poisson distribution can be expressed as shown in formula (1):

[0106]

[0107] where X represents the occurrence times of the target cases in each grid map, k = 0, 1,..., and P(X = k) is the probability function.

[0108] In step S302 of some embodiments, the target case data is simulated and analyzed through the Monte Carlo algorithm, and sampling is performed on the probability distribution of the target cases. For example, a probability space (W, A, P) is constructed through the Monte Carlo algorithm, where W is a set of target cases, A is a subset of the set W, and P is the average occurrence probability of the target cases established on A. In this probability space, a random variable q(w) is selected, and through the Monte Carlo algorithm, the expected value of the random variable q(w) is made equal to the occurrence probability P(X = k) of the target cases in each grid map. The arithmetic mean of the simple subsample of the random variable q(w) is used as an approximation of the occurrence probability. Then, through comprehensive analysis of this approximation value and the data such as the number and location of the extracted surveyors, the surveyor dispatch data is obtained, where the surveyor dispatch data includes the number of dispatched surveyors, the dispatch path, etc.

[0109] In step S303 of some embodiments, the surveyor dispatch data can be uploaded to the central control system, which can be a data center such as a traffic control platform. A first dispatch instruction is sent to the surveyor through the central control system, and the surveyor is instructed to conduct inspections on the target cases in the preset area to generate inspection data, which includes the basic details of the target cases, the processing duration of the target cases, etc.

[0110] In step S304 of some embodiments, the surveyor dispatch data and the inspection data can be optimized, for example, abnormal data is eliminated, missing data is supplemented, etc. Finally, the surveyor dispatch data and the inspection data after optimization are summarized to obtain preliminary survey data.

[0111] Please refer to Figure 4 , in some embodiments, step S103 may include but is not limited to steps S401 to S403:

[0112] Step S401: Obtain the number of target cases existing in a preset area during a second preset time period through a preset case recognition model;

[0113] Step S402: If the number of cases is greater than 0, screen and process the preliminary survey data according to a preset priority order to obtain the target survey data of the target case;

[0114] Step S403: Dispatch the corresponding target surveyor according to the target survey data, and instruct the target surveyor to process the target case to obtain the case processing data of the target case.

[0115] In step S401 of some embodiments, a case recognition model is used to detect whether a target case appears within different time periods. It should be noted that the case recognition model is a semantic recognition model based on keywords. Through preset keywords such as vehicle insurance, traffic accidents, car accidents, etc., the case recognition model can obtain whether a target case appears in a preset area during a second preset time period and the number of target cases existing.

[0116] In step S402 of some embodiments, if the number of cases is greater than 0, the case recognition model can extract features of the identified target case through a convolutional layer to obtain target case features, which include the urgency, processing difficulty, etc. of the target case. Among them, the urgency and processing difficulty of the target case can be set according to the actual situation. For example, the urgency includes general and urgent, and the processing difficulty includes simple, medium, difficult, etc. Furthermore, the initial survey data is screened and processed according to the target case features and a preset priority order. For example, the most suitable surveyor is determined according to the matching degree between the surveyor and the target case and the distance between the surveyor and the target case, as well as the path and time for the surveyor to reach the scene, etc., so as to improve the rationality of the target survey data. For example, a surveyor closest to the target case with a higher urgency is assigned to handle it, and a surveyor with the highest matching degree is assigned to handle a target case with a greater processing difficulty, etc.

[0117] In step S403 of some embodiments, the target survey data can be uploaded to a central control system, and a second dispatch instruction is sent to the surveyor through the central control system. The second dispatch instruction is used to instruct the target surveyor to go to the location where the target case occurred. When the target surveyor arrives at the place where the target case occurred and processes the target case, calculate the required time for the target surveyor to process the target case, and generate a case processing log according to the processing situation to obtain the case processing data.

[0118] Please refer to Figure 5 , in some embodiments, before step S104, the method further includes pre-training a data analysis model, specifically including:

[0119] Step S501: Obtain sample case data;

[0120] Step S502: Conduct simulation analysis on the sample case data to obtain sample survey data and sample case processing data;

[0121] Step S503: Input the sample survey data and the sample case processing data into the data analysis model;

[0122] Step S504: Calculate the sample survey metrics through the loss function of the data analysis model;

[0123] Step S505: Optimize the loss function of the data analysis model according to the sample survey metrics to update the data analysis model.

[0124] In step S501 of some embodiments, the sample case data can be obtained by writing a web crawler to crawl data purposefully after setting the data source, or by retrieving the sample case data from the historical case database, or by other means, which is not limited thereto.

[0125] In step S502 of some embodiments, the sample case data is simulated and analyzed through the Poisson distribution and the Monte Carlo algorithm to obtain the sample survey data and the sample case processing data. This process is basically the same as the processing process from step S102 to step S103 above, and will not be elaborated here.

[0126] In steps S503 and S504 of some embodiments, the sample survey data and the sample case processing data are input into the data analysis model. The data analysis model is a convolutional neural network model, including a convolutional layer, a pooling layer, and a fully connected layer.

[0127] In step S504 of some embodiments, the Transformer algorithm is used to encode the sample survey data and the sample case processing data to obtain a sample survey vector and a sample case processing vector. Max-pooling processing is respectively performed on the sample survey vector and the sample case processing vector to obtain the sample survey max-pooling feature and the sample case processing max-pooling feature. Average-pooling processing is respectively performed on the sample survey vector and the sample case processing vector to obtain the sample survey average-pooling feature and the sample case processing average-pooling feature. The sample survey average-pooling feature and the sample survey max-pooling feature are vector-concatenated to obtain a sample survey feature vector, and the sample case processing max-pooling feature and the sample case processing average-pooling feature are vector-concatenated to obtain a sample case processing feature vector. The sample survey feature vector and the sample case processing feature vector are activated through a preset function to obtain the sample survey metrics.

[0128] In step S505 of some embodiments, the sample survey index is compared with the reference survey index. According to the magnitude relationship between the sample survey index and the reference survey index, backpropagation is performed on the model loss of the loss function of the data analysis model to fine-tune the model parameters, so that the sample survey index is less than the reference survey index, and the update of the data analysis model is stopped.

[0129] Please refer to Figure 6 , in some embodiments, step S104 may further include but is not limited to steps S601 to S604:

[0130] Step S601, feature extraction is respectively performed on the target survey data and the case processing data through the convolutional layer of the data analysis model to obtain a target survey vector and a case processing vector;

[0131] Step S602, pooling processing is respectively performed on the target survey vector and the case processing vector through the pooling layer of the data analysis model to obtain a survey feature vector and a case processing feature vector;

[0132] Step S603, concatenation processing is performed on the survey feature vector and the case processing feature vector through the fully connected layer of the data analysis model to obtain a comprehensive survey vector;

[0133] Step S604, activation processing is performed on the comprehensive survey vector through the preset function of the data analysis model and the survey category label to obtain a predicted survey index corresponding to each survey category label.

[0134] In step S601 of some embodiments, feature extraction is performed on the case processing data and the target survey data through the convolutional layer of the data analysis model, and the Transformer algorithm is used to encode the case processing data and the target survey data to obtain an initial survey vector and an initial case processing vector.

[0135] In step S602 of some embodiments, max-pooling processing is respectively performed on the initial survey vector and the initial case processing vector by using the pooling layer in the data analysis model to obtain a survey max-pooling feature and a case processing max-pooling feature, and average-pooling processing is respectively performed on the initial survey vector and the initial case processing vector to obtain a survey average-pooling feature and a case processing average-pooling feature.

[0136] Step S603, vector concatenation processing is performed on the survey average-pooling feature and the survey max-pooling feature through the fully connected layer of the data analysis model to obtain a target survey feature vector, and vector concatenation processing is performed on the case processing max-pooling feature and the case processing average-pooling feature through the fully connected layer of the data analysis model to obtain a target case processing feature vector.

[0137] Step S604, activate the target survey feature vector and the target case processing feature vector through the softmax function in the fully connected layer, or other activation functions (such as the tanh function) can also be used, not limited to this. For example, when using the softmax function to activate the target survey feature vector and the target case processing feature vector, a probability distribution is created on the preset survey category labels through the softmax function, so as to mark and classify the target survey feature vector and the target case processing feature vector according to the probability distribution, and obtain the predicted survey indicators corresponding to each survey category label. Among them, the survey category labels include the surveyor's moving distance, the surveyor's moving speed, the proportion of invalid moving distance, the case waiting time, and so on.

[0138] Please refer to Figure 7 , in some embodiments, step S105 may further include but is not limited to steps S701 to S705:

[0139] Step S701, generate fine-tuning data according to the size relationship between the predicted survey indicators and the reference survey indicators;

[0140] Step S702, input the fine-tuning data and the target survey data into a preset deep learning model;

[0141] Step S703, generate fine-tuning indicators of the deep learning model according to the fine-tuning data;

[0142] Step S704, perform fine-tuning processing on the target survey data through the deep learning model and the fine-tuning indicators to obtain the survey data to be measured;

[0143] Step S705, perform human resource measurement processing on the survey data to be measured through the measurement function of the deep learning model and the environmental impact factor to obtain the survey human resource data.

[0144] In step S701 of some embodiments, compare the predicted survey indicators with the reference survey indicators to check whether the current number of surveyors meets the preset constraint conditions, that is, by comparing the predicted survey indicators with the reference survey indicators, identify whether the predicted survey indicators meet the requirements of the reference survey indicators, so as to generate survey deviation data according to the comparison results, that is, fine-tuning data. For example, compare the predicted waiting time in the predicted survey indicators with the reference waiting time, and according to the comparison results, check whether the current human resources can meet the human resource requirements.

[0145] In step S702 of some embodiments, input the fine-tuning data and the target survey data into a preset deep learning model, where the preset deep learning model is a BP neural network model, and the deep learning model includes an input layer, a hidden layer, and an output layer.

[0146] In steps S703 and S704 of some embodiments, the fine-tuning data is parsed and processed through an input layer to obtain a survey index deviation value, and the survey index deviation value is used as a fine-tuning index. Further, the survey index deviation value is encoded through an LSTM algorithm to obtain a survey index deviation vector, and the target survey data is encoded to obtain a target survey vector. Furthermore, the target survey vector and the survey index deviation vector are subjected to vector splicing processing to obtain a survey vector to be measured. Finally, the survey vector to be measured is decoded to obtain the survey data to be measured.

[0147] In step S705 of some embodiments, the survey data to be measured is processed for human survey through the sigmoid function of the hidden layer and environmental impact factors. Among them, the environmental impact factors include preset traffic flow parameters, weather parameters, etc. The sigmoid function is shown in formula (2):

[0148]

[0149] Through the sigmoid function, the predicted number of surveyors can be simulated and calculated based on the number of benchmark surveyors, the number of target cases, and environmental factors such as traffic flow parameters and weather parameters in the survey data to be measured. At the same time, by fine-tuning the input data of the deep learning model (i.e., the preset environmental impact factors and the data to be measured), the actual number of surveyors required in different scenarios can be easily measured, making the accuracy of the output survey manpower data relatively high.

[0150] In step S106 of some embodiments, the survey manpower data can be uploaded to the central control system. The central control system distributes survey tasks to surveyors according to the survey manpower data, that is, issues a third scheduling instruction to the surveyors. The third scheduling instruction is used to instruct the surveyors to go to the corresponding target case occurrence location and process the target case, improving the accuracy of survey task dispatching.

[0151] In the embodiments of the present application, by obtaining original case data and business data, and performing case simulation analysis on the original case data and business data using the Poisson distribution and the Monte Carlo algorithm to obtain preliminary survey data, it is possible to conveniently simulate and estimate the manpower requirements of surveyors corresponding to the cases. Furthermore, by using the preliminary survey data and a preset case recognition model to identify and process target cases in a preset area, the target survey data of the target cases and the case processing data of the target cases are obtained. At the same time, by performing data analysis on the target survey data and the case processing data through a pre-trained data analysis model, predicted survey indicators are obtained, and the working conditions of surveyors can be simulated relatively quickly. Finally, by comparing the predicted survey indicators with reference survey indicators using a deep learning model, the target survey data is measured and processed to obtain survey manpower data, and survey tasks are allocated according to the survey manpower data. The embodiments of the present application combine the Monte Carlo simulation algorithm and the deep learning algorithm to perform survey manpower measurement, which can improve the speed and accuracy of survey manpower measurement, thereby improving the accuracy of survey task dispatch.

[0152] Please refer to Figure 8 , the embodiments of the present application further provide a survey task allocation device, which can implement the above survey task allocation method. The device includes:

[0153] A data acquisition module 801, configured to acquire original case data and business data;

[0154] A simulation analysis module 802, configured to perform case simulation analysis according to the original case data and business data to obtain preliminary survey data;

[0155] An identification module 803, configured to identify and process target cases in a preset area through the preliminary survey data and a preset case recognition model to obtain the target survey data of the target cases and the case processing data of the target cases;

[0156] A data analysis module 804, configured to perform data analysis on the target survey data and the case processing data through a pre-trained data analysis model to obtain predicted survey indicators;

[0157] A manpower measurement module 805, configured to measure and process the target survey data according to the predicted survey indicators and reference survey indicators to obtain survey manpower data;

[0158] A task allocation module 806, configured to allocate survey tasks according to the survey manpower data.

[0159] The specific implementation manner of this survey task allocation device is basically the same as the specific embodiments of the above survey task allocation method, and will not be elaborated herein.

[0160] The embodiments of the present application further provide an electronic device, which includes: a memory, a processor, a program stored on the memory and executable on the processor, and a data bus for realizing the connection and communication between the processor and the memory. When the program is executed by the processor, the above-mentioned survey task allocation method is realized. The electronic device can be any intelligent terminal including a tablet computer, a vehicle-mounted computer, etc.

[0161] Please refer to Figure 9 , Figure 9 which schematically shows the hardware structure of an electronic device in another embodiment. The electronic device includes:

[0162] A processor 901, which can be implemented in ways such as a general-purpose CPU (Central Processing Unit), a microprocessor, an application-specific integrated circuit (ASIC), or one or more integrated circuits, and is used to execute relevant programs to implement the technical solutions provided by the embodiments of the present application;

[0163] A memory 902, which can be implemented in forms such as a read-only memory (ROM), a static storage device, a dynamic storage device, or a random access memory (RAM). The memory 902 can store an operating system and other application programs. When implementing the technical solutions provided by the embodiments of this specification through software or firmware, the relevant program codes are stored in the memory 902 and are called by the processor 901 to execute the survey task allocation method of the embodiments of the present application;

[0164] An input / output interface 903, which is used to realize information input and output;

[0165] A communication interface 904, which is used to realize the communication interaction between this device and other devices. It can communicate through wired means (such as USB, network cable, etc.) or wireless means (such as mobile network, WIFI, Bluetooth, etc.);

[0166] A bus 905, which transmits information between various components of the device (such as the processor 901, the memory 902, the input / output interface 903, and the communication interface 904);

[0167] Among them, the processor 901, the memory 902, the input / output interface 903, and the communication interface 904 realize the communication connection among themselves inside the device through the bus 905.

[0168] The embodiments of the present application also provide a storage medium, which is a computer-readable storage medium for computer-readable storage. The storage medium stores one or more programs, and the one or more programs can be executed by one or more processors to implement the above-mentioned survey task allocation method.

[0169] As a non-transitory computer-readable storage medium, the memory can be used to store non-transitory software programs and non-transitory computer-executable programs. In addition, the memory may include high-speed random access memory, and may also include non-transitory memory, such as at least one magnetic disk storage device, a flash memory device, or other non-transitory solid-state storage devices. In some embodiments, the memory may optionally include a memory remotely disposed relative to the processor, and these remote memories can be connected to the processor through a network. Examples of the above networks include, but are not limited to, the Internet, an enterprise intranet, a local area network, a mobile communication network, and combinations thereof.

[0170] The embodiments described in the embodiments of the present application are for more clearly illustrating the technical solutions of the embodiments of the present application, and do not constitute a limitation on the technical solutions provided by the embodiments of the present application. Those skilled in the art will know that with the evolution of technology and the emergence of new application scenarios, the technical solutions provided by the embodiments of the present application are equally applicable to similar technical problems.

[0171] Those skilled in the art can understand that Figure 1-7 the technical solutions shown in do not constitute a limitation on the embodiments of the present application, and may include more or fewer steps than those shown in the figures, or combine certain steps, or different steps.

[0172] The device embodiments described above are merely illustrative. The units described as separate components may or may not be physically separated, that is, they may be located in one place, or may be distributed to multiple network units. Some or all of the modules can be selected according to actual needs to achieve the purpose of the solution of this embodiment.

[0173] Those of ordinary skill in the art can understand that all or some of the steps in the methods disclosed above, and the functional modules / units in the systems and devices, can be implemented as software, firmware, hardware, and appropriate combinations thereof.

[0174] In the description of this application and the above-mentioned drawings, terms such as "first", "second", "third", "fourth", etc. (if any) are used to distinguish similar objects and do not necessarily describe a specific order or sequence. It should be understood that the data used in this way can be interchanged under appropriate circumstances so that the embodiments of this application described here can be implemented in an order different from those illustrated or described here. In addition, the terms "comprising" and "having" and any variations thereof are intended to cover non-exclusive inclusion. For example, a process, method, system, product or device comprising a series of steps or units does not necessarily have to be limited to those steps or units clearly listed, but may include other steps or units not clearly listed or inherent to these processes, methods, products or devices.

[0175] It should be understood that in this application, "at least one (item)" means one or more, and "a plurality" means two or more. "And / or" is used to describe the association relationship of associated objects and indicates that there can be three relationships. For example, "A and / or B" can mean: only A exists, only B exists, and both A and B exist at the same time. Here, A and B can be singular or plural. The character " / " generally indicates that the associated objects before and after are in an "or" relationship. "At least one (one) of the following" or its similar expressions refer to any combination of these items, including any combination of single item (one) or plural items (ones). For example, at least one (one) of a, b or c can mean: a, b, c, "a and b", "a and c", "b and c", or "a and b and c", where a, b, c can be single or multiple.

[0176] In several embodiments provided in this application, it should be understood that the disclosed devices and methods can be implemented in other ways. For example, the device embodiments described above are only illustrative. For example, the above-mentioned unit division is only a logical function division. In actual implementation, there can be other division methods. For example, multiple units or components can be combined or integrated into another system, or some features can be ignored or not executed. Another point is that the displayed or discussed coupling or direct coupling or communication connection to each other can be through some interfaces. The indirect coupling or communication connection of devices or units can be in electrical, mechanical or other forms.

[0177] The units described above as separate components may or may not be physically separated. The components displayed as units may or may not be physical units, that is, they can be located in one place or distributed to multiple network units. Some or all of the units can be selected according to actual needs to achieve the purpose of the solution of this embodiment.

[0178] In addition, in each embodiment of the present application, the functional units can be integrated into one processing unit, or each unit can exist physically alone, or two or more units can be integrated into one unit. The above-mentioned integrated unit can be implemented in the form of hardware or in the form of a software functional unit.

[0179] If the integrated unit is implemented in the form of a software functional unit and sold or used as an independent product, it can be stored in a computer-readable storage medium. Based on such an understanding, the technical solution of the present application, in essence, or the part that contributes to the prior art, or all or part of the technical solution, can be embodied in the form of a software product. The computer software product is stored in a storage medium and includes multiple instructions for causing a computer device (which can be a personal computer, a server, or a network device, etc.) to execute all or part of the steps of the methods in the various embodiments of the present application. The foregoing storage medium includes: various media that can store programs, such as USB flash drives, mobile hard disks, read-only memories (ROM), random access memories (RAM), magnetic disks, or optical discs.

[0180] The preferred embodiments of the embodiments of the present application have been described above with reference to the accompanying drawings. However, this does not limit the scope of the rights of the embodiments of the present application. Any modifications, equivalent replacements, and improvements made by those skilled in the art without departing from the scope and essence of the embodiments of the present application shall fall within the scope of the rights of the embodiments of the present application.

Claims

1. A survey task allocation method, characterized in that, The method includes: Obtaining original case data and business data; Performing case simulation analysis based on the original case data and the business data to obtain preliminary survey data; Identifying and processing target cases in a preset area through the preliminary survey data and a preset case identification model to obtain target survey data of the target cases and case processing data of the target cases; Performing data analysis on the target survey data and the case processing data through a pre-trained data analysis model to obtain predicted survey indicators; Performing measurement processing on the target survey data according to the predicted survey indicators and reference survey indicators to obtain survey manpower data; Allocating survey tasks according to the survey manpower data; The performing case simulation analysis based on the original case data and the business data to obtain preliminary survey data includes: Performing data preprocessing on the original case data and the business data through Poisson distribution to obtain target case data corresponding to the target cases; Performing simulation analysis on the target cases by using the Monte Carlo algorithm and the target case data to obtain surveyor scheduling data; Outputting a first scheduling instruction according to the surveyor scheduling data; the first scheduling instruction is used to instruct a surveyor to perform simulated patrol on the target cases in a preset area to obtain simulated patrol data; Performing fusion processing on the surveyor scheduling data and the simulated patrol data to obtain the preliminary survey data.

2. The survey task allocation method according to claim 1, characterized in that, The step of obtaining original case data and business data includes: Obtaining target map data; Dividing the target map into multiple grid maps according to a preset ratio and the target map data; Collecting the case situations occurring in each grid map within a first preset time period to obtain the original case data and the business data.

3. The survey task allocation method according to claim 1, characterized in that, The step of identifying and processing target cases in a preset area through the preliminary survey data and a preset case identification model to obtain target survey data of the target cases and case processing data of the target cases includes: Obtaining the number of target cases existing in a preset area within a second preset time period through a preset case identification model; If the number of cases is greater than 0, screening and processing the preliminary survey data according to a preset priority order to obtain target survey data of the target cases; Scheduling corresponding target surveyors according to the target survey data and instructing the target surveyors to process the target cases to obtain case processing data of the target cases.

4. The survey task allocation method according to claim 1, characterized in that, The step of performing data analysis on the target survey data and the case processing data through a pre-trained data analysis model to obtain predicted survey indicators includes: Performing feature extraction on the target survey data and the case processing data respectively through the convolutional layer of the data analysis model to obtain a target survey vector and a case processing vector; Performing pooling processing on the target survey vector and the case processing vector respectively through the pooling layer of the data analysis model to obtain a survey feature vector and a case processing feature vector; Concatenate the survey feature vector and the case processing feature vector through the fully connected layer of the data analysis model to obtain a comprehensive survey vector; Perform activation processing on the comprehensive survey vector through the preset function of the data analysis model and the survey category label to obtain the predicted survey indicators corresponding to each survey category label.

5. The survey task allocation method according to claim 1, characterized in that, Based on the predicted survey indicators and the reference survey indicators, perform measurement processing on the target survey data to obtain survey manpower data, including: Generate fine-tuning data according to the magnitude relationship between the predicted survey indicators and the reference survey indicators; Input the fine-tuning data and the target survey data into a preset deep learning model; Generate fine-tuning indicators for the deep learning model according to the fine-tuning data; Perform fine-tuning processing on the target survey data through the deep learning model and the fine-tuning indicators to obtain the survey data to be measured; Perform manpower measurement processing on the survey data to be measured through the measurement function of the deep learning model and the environmental impact factor to obtain the survey manpower data.

6. The survey task allocation method according to any one of claims 1 to 5, characterized in that, Before the step of performing data analysis on the target survey data and the case processing data through a pre-trained data analysis model to obtain predicted survey indicators, the method further includes pre-training the data analysis model, specifically including: Obtain sample case data; Perform simulation analysis on the sample case data to obtain sample survey data and sample case processing data; Input the sample survey data and the sample case processing data into the data analysis model; Calculate sample survey indicators through the loss function of the data analysis model; Optimize the loss function of the data analysis model according to the sample survey indicators to update the data analysis model.

7. A survey task allocation device, characterized in that, The device includes: A data acquisition module for acquiring original case data and business data; A simulation analysis module for performing case simulation analysis according to the original case data and the business data to obtain preliminary survey data; An identification module for identifying the target case in the preset area through the preliminary survey data and a preset case identification model to obtain the target survey data of the target case and the case processing data of the target case; A data analysis module for performing data analysis on the target survey data and the case processing data through a pre-trained data analysis model to obtain predicted survey indicators; A manpower measurement module for performing measurement processing on the target survey data according to the predicted survey indicators and the reference survey indicators to obtain survey manpower data; A task allocation module for allocating survey tasks according to the survey manpower data; Performing case simulation analysis according to the original case data and the business data to obtain preliminary survey data, including: Perform data preprocessing on the original case data and the business data through Poisson distribution to obtain target case data corresponding to the target case; Perform simulation analysis on the target case by using the Monte Carlo algorithm and the target case data to obtain surveyor scheduling data; Output a first dispatch instruction according to the surveyor dispatch data; the first dispatch instruction is used to instruct the surveyor to perform a simulated patrol on the target cases in the preset area to obtain simulated patrol data; Perform a fusion process on the surveyor dispatch data and the simulated patrol data to obtain the preliminary survey data.

8. An electronic device, characterized in that, The electronic device includes a memory, a processor, a program stored on the memory and executable on the processor, and a data bus for realizing connection communication between the processor and the memory. When the program is executed by the processor, the steps of the survey task allocation method according to any one of claims 1 to 6 are realized.

9. A storage medium, the storage medium being a computer-readable storage medium for computer-readable storage, characterized in that, The storage medium stores one or more programs, and the one or more programs can be executed by one or more processors to realize the steps of the survey task allocation method according to any one of claims 1 to 6.

Images