Real estate risk assessment method and device and storage medium
The risk assessment of real estate data is carried out through the target risk detection model, which solves the problems of high evaluation costs, low efficiency and inaccurate results in the existing technology, and realizes automated and objective risk assessment, which improves the evaluation efficiency and accuracy.
Patent Information
- Application Number
- CN202410957264.X
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2024-07-16
- Publication Date
- 2024-11-12
AI Technical Summary
The existing real estate evaluation methods rely on manual on site inspection, resulting in high cost, low efficiency, inaccurate results, and lack of objectivity and consistency.
The target risk detection model is used to conduct risk assessment of real estate data, and by obtaining the original data for processing and training, risk probability distribution results are generated, risk assessment is automatically completed, and human intervention is reduced.
It improves the evaluation efficiency, reduces costs, and the results are more objective and accurate, avoiding the subjective impact of manual evaluation.
Smart Images

Figure CN118941074A_ABST
Abstract
Description
Technical Field
[0001] The present invention relates to the technical field of risk assessment, and in particular to a real estate risk assessment method, device and storage medium. Background Art
[0002] At present, in the field of real estate appraisal, it mainly relies on manual on-site inspection and experience judgment. Among them, the appraiser needs to personally go to the location of the real estate for on-site inspection, observe the surrounding environment, building structure and other factors of the real estate, and combine professional knowledge and experience to conduct risk assessment of the real estate. Among them, the above method can take into account the actual situation of the real estate, but the appraisal cost is high and the appraisal efficiency is low, and the appraisal results are easily affected by the appraiser's personal experience and subjective judgment, lack of objectivity and consistency, and make the appraisal results inaccurate. Summary of the invention
[0003] The present invention provides a real estate risk assessment method and device to solve the technical problems of high assessment cost, low assessment efficiency and inaccurate assessment results in the above-mentioned prior art.
[0004] To this end, the present invention proposes a real estate risk assessment method, which can conduct risk assessment on real estate data through a target risk detection model, without manual intervention, and can automatically complete risk assessment of a large number of real estates, thereby improving assessment efficiency and reducing assessment costs. At the same time, the risk probability distribution results obtained through the target risk detection model are more objective, not affected by human factors, and improve the accuracy of risk assessment results.
[0005] Another object of the present invention is to provide a real estate risk assessment device.
[0006] To achieve the above object, the present invention provides a real estate risk assessment method, which includes:
[0007] Acquiring original data of real estate, and performing data processing on the original data to obtain a first data set;
[0008] Training the initial risk detection model based on the first data set to obtain a target risk detection model;
[0009] Acquire real estate data to be detected, input the real estate data to be detected into the target risk detection model, and obtain corresponding risk probability distribution results;
[0010] Based on the risk probability distribution result, the corresponding target risk result is determined and output.
[0011] The real estate risk assessment method of the embodiment of the present invention may also have the following additional technical features:
[0012] In one embodiment of the present invention, the obtaining of original data of real estate includes:
[0013] Acquire original multimodal data of real estate, wherein the original multimodal data includes original images and / or original videos;
[0014] If the original multimodal data includes original video data, performing image extraction on the original video data to obtain a corresponding original image;
[0015] The original image is risk-labeled, and the labeled image data is determined as the original data of the real estate.
[0016] In one embodiment of the present invention, the step of processing the original data to obtain the first data set includes:
[0017] Performing image format processing on the original data to obtain a corresponding second data set;
[0018] Performing image size processing on the data in the second data set to obtain a corresponding third data set;
[0019] Performing data enhancement processing on the data in the third data set to obtain a corresponding fourth data set;
[0020] The data in the fourth data set is normalized or standardized to obtain the corresponding first data set.
[0021] In one embodiment of the present invention, the training of the initial risk detection model based on the first data set to obtain the target risk detection model includes:
[0022] Based on a preset ratio, the first data set is divided into a training data set and a test data set;
[0023] Using the training data set to train the initial risk detection model to obtain a trained risk detection model;
[0024] The trained risk detection model is tested and optimized using the test data set to obtain a target risk detection model.
[0025] In one embodiment of the present invention, the using the training data set to train the initial risk detection model to obtain a trained risk detection model includes:
[0026] Inputting the training data set into the initial risk detection model to obtain a predicted risk probability distribution result;
[0027] Based on the predicted risk probability distribution result and the actual risk distribution result, a corresponding loss value is obtained through a loss function;
[0028] The network parameters in the initial risk detection model are updated based on the loss value until the initial risk detection model converges, or when the number of iterations of the network parameters reaches a preset number, a trained risk detection model is obtained.
[0029] In one embodiment of the present invention, the risk probability distribution result includes probability values and location information of various risk types corresponding to different areas in the real estate to be inspected.
[0030] In one embodiment of the present invention, determining and outputting a corresponding target risk result based on the risk probability distribution result includes:
[0031] Compare the probability values of various risk types in each area in the risk probability distribution result with the corresponding preset thresholds, and determine the target risk type in each area and the target location information corresponding to the target risk type based on the comparison result;
[0032] In the real estate data to be detected, the target location information corresponding to each target risk type is graphically annotated, and the annotated data is output.
[0033] To achieve the above object, the present invention provides a real estate risk assessment device on the other hand, the device comprising:
[0034] A data processing module, used for acquiring original data of real estate, and performing data processing on the original data to obtain a first data set;
[0035] A training module, used to train the initial risk detection model based on the first data set to obtain a target risk detection model;
[0036] An evaluation module, used for acquiring real estate data to be detected, inputting the real estate data to be detected into the target risk detection model, and obtaining a corresponding risk probability distribution result;
[0037] The output module is used to determine and output the corresponding target risk result based on the risk probability distribution result.
[0038] Another object of the present invention is to provide an electronic device, comprising:
[0039] at least one processor; and
[0040] a memory communicatively connected to the at least one processor; wherein,
[0041] The memory stores instructions that can be executed by the at least one processor, and the instructions are executed by the at least one processor to enable the at least one processor to perform any one of the methods described in the above aspect.
[0042] Another object of the present invention is to provide a computer storage medium, wherein the computer storage medium stores computer executable instructions; after the computer executable instructions are executed by a processor, the computer executes any one of the methods described in the above aspect.
[0043] The real estate risk assessment method and device of the embodiment of the present invention obtains the original data of the real estate, and processes the original data to obtain a first data set; trains the initial risk detection model based on the first data set to obtain a target risk detection model; obtains the real estate data to be detected, inputs the real estate data to be detected into the target risk detection model to obtain the corresponding risk probability distribution result; based on the risk probability distribution result, determines and outputs the corresponding target risk result. Therefore, the present invention performs risk assessment on real estate data through the target risk detection model without human intervention, and can automatically complete the risk assessment of a large number of real estates, thereby improving the assessment efficiency and reducing the assessment cost. At the same time, the risk probability distribution result obtained by the target risk detection model is more objective, not affected by human factors, and improves the accuracy of the risk assessment result.
[0044] Additional aspects and advantages of the present invention will be given in part in the following description and in part will be obvious from the following description, or will be learned through practice of the present invention. BRIEF DESCRIPTION OF THE DRAWINGS
[0045] The above and / or additional aspects and advantages of the present invention will become apparent and easily understood from the following description of the embodiments in conjunction with the accompanying drawings, in which:
[0046] Figure 1 is a flow chart of a real estate risk assessment method according to an embodiment of the present invention;
[0047] Figure 2 4 is a structural diagram of a real estate risk assessment device according to an embodiment of the present invention. DETAILED DESCRIPTION
[0048] It should be noted that, in the absence of conflict, the embodiments of the present invention and the features in the embodiments can be combined with each other. The present invention will be described in detail below with reference to the accompanying drawings and in combination with the embodiments.
[0049] In order to enable those skilled in the art to better understand the scheme of the present invention, the technical scheme in the embodiments of the present invention will be clearly and completely described below in conjunction with the drawings in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work should fall within the scope of protection of the present invention.
[0050] The following describes a real estate risk assessment method and device according to an embodiment of the present invention with reference to the accompanying drawings.
[0051] Figure 1 It is a flow chart of a real estate risk assessment method according to an embodiment of the present invention.
[0052] like Figure 1 As shown, the method includes:
[0053] S1, obtaining original data of real estate, and performing data processing on the original data to obtain a first data set;
[0054] In one embodiment of the present invention, the above-mentioned real estate may include: houses and land.
[0055] Furthermore, in one embodiment of the present invention, the method for obtaining the original data of real estate may include the following steps:
[0056] Step 1: Acquire original multimodal data of real estate, wherein the original multimodal data includes original images and / or original videos;
[0057] Step 2: If the original multimodal data includes original video data, image extraction is performed on the original video data to obtain a corresponding original image;
[0058] Step 3: Label the original image for risks, and determine the labeled image data as the original data of the real estate.
[0059] In one embodiment of the present invention, the original multimodal data of the real estate may be obtained through public authorized channels (such as the Internet) or through field surveys and photography.
[0060] Furthermore, in one embodiment of the present invention, if the acquired original multimodal data includes original video data, images in the original video data may be extracted frame by frame to filter out duplicate images and obtain corresponding original images.
[0061] Furthermore, in one embodiment of the present invention, after the original image is obtained through the above steps, the risks in the original image can be manually annotated. In one embodiment of the present invention, the surface risks and potential risks of real estate can be annotated. And, in one embodiment of the present invention, the surface risks can include cracks and / or wall skins and / or load-bearing marks; the potential risks can be the location of the real estate and / or the unfavorable factors inside the real estate. For example, in one embodiment of the present invention, the unfavorable factors can be protruding corners in the room.
[0062] Furthermore, in one embodiment of the present invention, after the original data of the real estate is obtained, the original data may be processed to obtain a first data set.
[0063] Specifically, in one embodiment of the present invention, the method of processing the original data to obtain the first data set may include the following steps:
[0064] S11, performing image format processing on the original data to obtain a corresponding second data set;
[0065] S12, performing image size processing on the data in the second data set to obtain a corresponding third data set;
[0066] S13, performing data enhancement processing on the data in the third data set to obtain a corresponding fourth data set;
[0067] S14, normalize or standardize the data in the fourth data set to obtain the corresponding first data set.
[0068] In one embodiment of the present invention, the original data may be processed in image format so that the format of the image data in the second data set is unified (such as JPEG or PNG). Also, in one embodiment of the present invention, the data in the second data set may be processed in image size so that the image data in the third data set is adjusted to a uniform size or resolution.
[0069] Furthermore, in one embodiment of the present invention, data in the third data set may be subjected to data enhancement processing (such as rotation or scaling or cropping or flipping) to increase the generalization ability of the subsequent model. Also, in one embodiment of the present invention, image data in the fourth data set may be subjected to normalization or standardization processing to eliminate differences between different images due to factors such as illumination or contrast.
[0070] S2, training the initial risk detection model based on the first data set to obtain a target risk detection model;
[0071] In one embodiment of the present invention, after the first data set is obtained through the above steps, the initial risk detection model can be trained based on the first data set to obtain a target risk detection model.
[0072] In one embodiment of the present invention, before training the initial risk detection model based on the first data set, the types of data annotations in the first data set can be compared to determine whether there is a certain type of image with a smaller number or significantly less than other types. If so, oversampling (such as the SMOTE algorithm) or undersampling technology can be used to balance the data set so that the analysis results of the subsequent model trained using the first data set are more accurate.
[0073] Furthermore, in one embodiment of the present invention, the method of training the initial risk detection model based on the first data set to obtain the target risk detection model may include the following steps:
[0074] S21, dividing the first data set into a training data set and a test data set based on a preset ratio;
[0075] S22, training the initial risk detection model using the training data set to obtain a trained risk detection model;
[0076] S23, using the test data set to test and tune the trained risk detection model to obtain a target risk detection model.
[0077] In one embodiment of the present invention, the first data set can be divided into a training data set and a test data set based on a preset ratio. For example, assuming that the preset ratio is 8:2, that is, 80% of the data in the first data set is randomly divided into a training data set and 20% of the data is divided into a test data set.
[0078] Furthermore, in one embodiment of the present invention, the method of training the initial risk detection model using the training data set to obtain the trained risk detection model may include the following steps:
[0079] Step a: input the training data set into the initial risk detection model to obtain the predicted risk probability distribution result;
[0080] Step b: Based on the predicted risk probability distribution result and the actual risk distribution result, the corresponding loss value is obtained through the loss function;
[0081] Step c: updating the network parameters in the initial risk detection model based on the loss value until the initial risk detection model converges, or when the number of iterations of the network parameters reaches a preset number, a trained risk detection model is obtained.
[0082] Among them, in one embodiment of the present invention, the above-mentioned initial risk detection model can be a deep residual network (such as ResNet). Specifically, in one embodiment of the present invention, deep residual networks of different depths can be selected according to the complexity and data size of the first data set. For example, in one embodiment of the present invention, if the amount of data in the first data set does not exceed the quantity threshold, the initial risk detection model can be ResNet-50; if the amount of data in the first data set exceeds the quantity threshold, the initial risk detection model can be ResNet-101.
[0083] And, in one embodiment of the present invention, the above-mentioned deep residual network may include a region proposal network (RPN), so as to generate at least one region where there is risk using the region proposal network, and then the predicted risk probability distribution result obtained by the ResNet network may include the probability value and location information of each risk type in each region. The location information may be represented by a bounding box in the image data corresponding to the first data set.
[0084] Further, in one embodiment of the present invention, after obtaining the predicted risk probability distribution result through the above steps, the above method of obtaining the corresponding loss value through the loss function based on the predicted risk probability distribution result and the actual risk distribution result may include: comparing the probability values of various risk types in each area in the predicted risk probability distribution result with the corresponding preset threshold value, and determining the risk type and location information whose probability value exceeds the corresponding preset threshold value as the predicted risk distribution result; based on the predicted risk distribution result and the actual risk distribution result, obtaining the corresponding loss value through the loss function. Wherein, in one embodiment of the present invention, the above loss function may be a cross entropy loss function.
[0085] Furthermore, in one embodiment of the present invention, after the trained risk detection model is obtained through the above steps, the trained risk detection model can be tested using the test data in the test set to obtain the test output result of the trained risk detection model. Also, in one embodiment of the present invention, the test output result is compared with the corresponding real risk distribution result. If the accuracy of the test output result is within the threshold range, it means that the trained risk detection model meets the requirements, and the trained risk detection model is determined as the target risk detection model; otherwise, the above steps are repeated for the trained risk detection model, and the network parameters of the trained risk detection model are adjusted until the trained risk detection model meets the requirements.
[0086] S3, obtaining real estate data to be detected, inputting the real estate data to be detected into the target risk detection model, and obtaining the corresponding risk probability distribution result;
[0087] In one embodiment of the present invention, after obtaining the real estate data to be detected, before inputting the real estate data to be detected into the target risk detection model, the real estate data to be detected can be processed, and the real estate data after data processing can be input into the target risk detection model to obtain the corresponding risk probability distribution result. In one embodiment of the present invention, the details of data processing can refer to the relevant introduction in the above step S1, and the present embodiment will not be repeated here.
[0088] In one embodiment of the present invention, the above-mentioned risk probability distribution result may include probability values and location information of various risk types corresponding to different areas in the real estate to be detected.
[0089] S4: Based on the risk probability distribution results, determine the corresponding target risk results and output them.
[0090] In one embodiment of the present invention, after obtaining the risk probability distribution result corresponding to the real estate data to be detected through the above steps, the corresponding target risk result can be determined and output based on the risk probability distribution result.
[0091] Specifically, in one embodiment of the present invention, a method for determining and outputting corresponding target risk results based on risk probability distribution results may include: comparing the probability values of various risk types in each area in the risk probability distribution results with corresponding preset thresholds, and determining the target risk type in each area and the target location information corresponding to the target risk type based on the comparison results; graphically marking the target location information corresponding to each target risk type in the real estate data to be detected, and outputting the marked data.
[0092] Among them, in one embodiment of the present invention, the above-mentioned method of comparing the probability values of various risk types in each area in the risk probability distribution results with the corresponding preset thresholds, and determining the target risk type in each area and the target location information corresponding to the target risk type based on the comparison results may include: comparing the probability values of various risk types in each area in the risk probability distribution results with the corresponding preset thresholds, and determining the risk type and location information in each area whose probability value exceeds the corresponding preset threshold as the target risk type in each area and the target location information corresponding to the target risk type.
[0093] Furthermore, in one embodiment of the present invention, the target location information corresponding to each target risk type may be graphically annotated, wherein different risk types may be annotated by boundary boxes of different colors to distinguish between different types.
[0094] Furthermore, in one embodiment of the present invention, the above method can also generate a corresponding risk assessment report based on the real estate data to be detected and the risk detection results, listing all detected risk types, location information and possible impacts, so that users can intuitively understand the risks of the real estate.
[0095] In the real estate risk assessment method proposed in the present invention, the original data of the real estate is obtained, and the original data is processed to obtain a first data set; the initial risk detection model is trained based on the first data set to obtain a target risk detection model; the real estate data to be detected is obtained, and the real estate data to be detected is input into the target risk detection model to obtain the corresponding risk probability distribution result; based on the risk probability distribution result, the corresponding target risk result is determined and output. Therefore, the present invention performs risk assessment on real estate data through the target risk detection model without human intervention, and can automatically complete the risk assessment of a large number of real estates, thereby improving the assessment efficiency and reducing the assessment cost. At the same time, the risk probability distribution result obtained by the target risk detection model is more objective, not affected by human factors, and improves the accuracy of the risk assessment result.
[0096] In order to implement the above embodiment, Figure 2 As shown, this embodiment also provides a real estate risk assessment device 10, which includes a data processing module 201, a training module 202, an assessment module 203, and an output module 204;
[0097] The data processing module 201 is used to obtain the original data of the real estate and process the original data to obtain a first data set;
[0098] A training module 202 is used to train the initial risk detection model based on the first data set to obtain a target risk detection model;
[0099] The evaluation module 203 is used to obtain the real estate data to be detected, input the real estate data to be detected into the target risk detection model, and obtain the corresponding risk probability distribution result;
[0100] The output module 204 is used to determine and output the corresponding target risk result based on the risk probability distribution result.
[0101] In one embodiment of the present invention, the above data processing module is specifically used for:
[0102] Acquire original multimodal data of the real estate, wherein the original multimodal data includes original images and / or original videos;
[0103] If the original multimodal data includes original video data, performing image extraction on the original video data to obtain a corresponding original image;
[0104] The original image is annotated for risks, and the annotated image data is determined as the original data of the real estate.
[0105] Furthermore, in one embodiment of the present invention, the data processing module is specifically used to:
[0106] Performing image format processing on the original data to obtain a corresponding second data set;
[0107] Performing image size processing on the data in the second data set to obtain a corresponding third data set;
[0108] Performing data enhancement processing on the data in the third data set to obtain a corresponding fourth data set;
[0109] The data in the fourth data set is normalized or standardized to obtain the corresponding first data set.
[0110] Furthermore, in one embodiment of the present invention, the training module is specifically used for:
[0111] Based on a preset ratio, the first data set is divided into a training data set and a test data set;
[0112] The initial risk detection model is trained using the training data set to obtain a trained risk detection model;
[0113] The trained risk detection model is tested and optimized using the test data set to obtain the target risk detection model.
[0114] Furthermore, in one embodiment of the present invention, the training module is specifically used for:
[0115] Inputting the training data set into the initial risk detection model to obtain a predicted risk probability distribution result;
[0116] Based on the predicted risk probability distribution result and the actual risk distribution result, a corresponding loss value is obtained through a loss function;
[0117] The network parameters in the initial risk detection model are updated based on the loss value until the initial risk detection model converges, or when the number of iterations of the network parameters reaches a preset number, a trained risk detection model is obtained.
[0118] Furthermore, in one embodiment of the present invention, the risk probability distribution result includes probability values and location information of various risk types corresponding to different areas in the real estate to be inspected.
[0119] Furthermore, in one embodiment of the present invention, the output module is further used to:
[0120] Compare the probability values of various risk types in each area in the risk probability distribution result with the corresponding preset thresholds, and determine the target risk type in each area and the target location information corresponding to the target risk type based on the comparison result;
[0121] In the real estate data to be detected, the target location information corresponding to each target risk type is graphically annotated, and the annotated data is output.
[0122] In the real estate risk assessment device proposed by the present invention, the original data of the real estate is obtained, and the original data is processed to obtain a first data set; the initial risk detection model is trained based on the first data set to obtain a target risk detection model; the real estate data to be detected is obtained, and the real estate data to be detected is input into the target risk detection model to obtain the corresponding risk probability distribution result; based on the risk probability distribution result, the corresponding target risk result is determined and output. Therefore, the present invention performs risk assessment on real estate data through the target risk detection model without human intervention, and can automatically complete the risk assessment of a large number of real estates, thereby improving the assessment efficiency and reducing the assessment cost. At the same time, the risk probability distribution result obtained by the target risk detection model is more objective, not affected by human factors, and improves the accuracy of the risk assessment result.
[0123] In the description of this specification, the description with reference to the terms "one embodiment", "some embodiments", "example", "specific example", or "some examples" etc. means that the specific features, structures, materials or characteristics described in conjunction with the embodiment or example are included in at least one embodiment or example of the present invention. In this specification, the schematic representations of the above terms do not necessarily refer to the same embodiment or example. Moreover, the specific features, structures, materials or characteristics described may be combined in any one or more embodiments or examples in a suitable manner. In addition, those skilled in the art may combine and combine the different embodiments or examples described in this specification and the features of the different embodiments or examples, without contradiction.
[0124] In addition, the terms "first" and "second" are used for descriptive purposes only and should not be understood as indicating or implying relative importance or implicitly indicating the number of the indicated technical features. Therefore, the features defined as "first" and "second" may explicitly or implicitly include at least one of the features. In the description of the present invention, the meaning of "plurality" is at least two, such as two, three, etc., unless otherwise clearly and specifically defined.
Claims
1. A real estate risk assessment method, It is characterized in that The method comprises: Acquiring original data of real estate, and performing data processing on the original data to obtain a first data set; Training the initial risk detection model based on the first data set to obtain a target risk detection model; Acquire real estate data to be detected, input the real estate data to be detected into the target risk detection model, and obtain corresponding risk probability distribution results; Based on the risk probability distribution result, the corresponding target risk result is determined and output.
2. The method according to claim 1, It is characterized in that The obtaining of original data of real estate includes: Acquire original multimodal data of real estate, wherein the original multimodal data includes original images and / or original videos; If the original multimodal data includes original video data, performing image extraction on the original video data to obtain a corresponding original image; The original image is risk-labeled, and the labeled image data is determined as the original data of the real estate.
3. The method according to claim 1, It is characterized in that The step of processing the original data to obtain a first data set includes: Performing image format processing on the original data to obtain a corresponding second data set; Performing image size processing on the data in the second data set to obtain a corresponding third data set; Performing data enhancement processing on the data in the third data set to obtain a corresponding fourth data set; The data in the fourth data set is normalized or standardized to obtain the corresponding first data set.
4. The method according to claim 1, It is characterized in that The training of the initial risk detection model based on the first data set to obtain a target risk detection model includes: Based on a preset ratio, dividing the first data set into a training data set and a test data set; Using the training data set to train the initial risk detection model to obtain a trained risk detection model; The trained risk detection model is tested and optimized using the test data set to obtain a target risk detection model.
5. The method according to claim 4, It is characterized in that The step of training the initial risk detection model using the training data set to obtain a trained risk detection model includes: Inputting the training data set into the initial risk detection model to obtain a predicted risk probability distribution result; Based on the predicted risk probability distribution result and the actual risk distribution result, a corresponding loss value is obtained through a loss function; The network parameters in the initial risk detection model are updated based on the loss value until the initial risk detection model converges, or when the number of iterations of the network parameters reaches a preset number, a trained risk detection model is obtained.
6. The method according to claim 1, It is characterized in that The risk probability distribution result includes probability values and location information of various risk types corresponding to different areas in the real estate to be detected.
7. The method according to claim 6, It is characterized in that The determining and outputting a corresponding target risk result based on the risk probability distribution result includes: Compare the probability values of various risk types in each area in the risk probability distribution result with the corresponding preset thresholds, and determine the target risk type in each area and the target location information corresponding to the target risk type based on the comparison result; In the real estate data to be detected, the target location information corresponding to each target risk type is graphically annotated, and the annotated data is output.
8. A real estate risk assessment device, It is characterized in that The device comprises: A data processing module, used for acquiring original data of real estate, and performing data processing on the original data to obtain a first data set; A training module, used to train the initial risk detection model based on the first data set to obtain a target risk detection model; An evaluation module, used for acquiring real estate data to be detected, inputting the real estate data to be detected into the target risk detection model, and obtaining a corresponding risk probability distribution result; The output module is used to determine and output the corresponding target risk result based on the risk probability distribution result.
9. An electronic device comprising: at least one processor; and a memory communicatively coupled to the at least one processor; in, The memory stores instructions that can be executed by the at least one processor, and the instructions are executed by the at least one processor to enable the at least one processor to perform the method according to any one of claims 1 to 7.
10. A computer storage medium, in, The computer storage medium stores computer executable instructions; after the computer executable instructions are executed by the processor, the method according to any one of claims 1 to 7 can be implemented.
