Check of predicted data

Abstract

Disclosed in embodiments of the present disclosure are a predicted data check method and apparatus, an electronic device, and a storage medium. The predicted data check method comprises determining calculation factors corresponding to a target calculation process, and respectively enumerating all factor values corresponding to each calculation factor; then permuting and combining the factor values on the basis of a combination algorithm to obtain a multi-dimensional matrix; for any prediction scenario, on the basis of first value ranges of factor values in a calculation factor combination, determining second value ranges of the calculation factor combination; and finally, on the basis of the second value ranges, checking a scenario predicted value corresponding to each prediction scenario.

Description

Verification of forecast data Technical Field

[0001] The embodiments of the present disclosure belong to the field of data processing technology, and particularly relate to a method, device, electronic device, and storage medium for verifying prediction data. Background Art

[0002] In fields that require complex calculations, such as social science research, market research, medical biostatistics, engineering quality control, and insurance pricing, a set of specific variables is often used to represent the various factors that affect the calculation results, and these variables are used as calculation factors to predict the calculation results in a specific scenario.

[0003] However, due to the potential discrepancy between actual data distribution and ideal conditions, these discrepancies can lead to algorithmic errors in real-world scenarios, ultimately resulting in irrational predictions. Currently, due to the complexity of these prediction calculations, irrational predictions are difficult to detect in a timely manner, leading to potential losses. Summary of the Invention

[0004] The embodiments of the present disclosure provide a method, device, electronic device, and storage medium for verifying prediction data, and the technical solutions thereof are as follows.

[0005] In a first aspect, an embodiment of the present disclosure provides a method for verifying prediction data, comprising: determining each calculation factor corresponding to a target calculation process, and enumerating all factor values ​​corresponding to each of the calculation factors; arranging and combining each of the factor values ​​based on a combination algorithm to obtain a multidimensional matrix, wherein the multidimensional matrix is ​​used to represent a mapping relationship between prediction scenarios and calculation factor combinations; for any of the prediction scenarios, determining a second value range of the calculation factor combination based on a first value range of each of the factor values ​​in the calculation factor combination; and verifying the scenario prediction value corresponding to each of the prediction scenarios based on each of the second value ranges.

[0006] In a second aspect, an embodiment of the present disclosure provides a method for verifying predicted data, including: determining each pricing factor corresponding to the insurance pricing process, and enumerating all factor values ​​corresponding to each of the pricing factors; arranging and combining each of the factor values ​​based on a combination algorithm to obtain a multidimensional matrix, wherein the multidimensional matrix is ​​used to represent the mapping relationship between pricing scenarios and pricing factor combinations; for any of the pricing scenarios, determining the second value range of the pricing factor combination based on the first value range of each of the factor values ​​in the pricing factor combination; and verifying the actual pricing corresponding to each of the pricing scenarios based on the second value ranges.

[0007] In a third aspect, an embodiment of the present disclosure provides a device for verifying prediction data, comprising: a first determination module for determining the calculation factors corresponding to the target calculation process, and enumerating all factor values ​​corresponding to each of the calculation factors; a first permutation and combination module for permuting and combining the factor values ​​based on a combination algorithm to obtain a multidimensional matrix, wherein the multidimensional matrix is ​​used to represent the mapping relationship between the prediction scenario and the calculation factor combination; a second determination module for determining, for any of the prediction scenarios, the second value range of the calculation factor combination based on the first value range of each factor value in the calculation factor combination; and a first verification module for verifying the scenario prediction value corresponding to each of the prediction scenarios based on the second value range.

[0008] In a fourth aspect, an embodiment of the present disclosure provides a device for verifying predicted data, comprising: a third determination module, for determining each pricing factor corresponding to the insurance pricing process, and enumerating all factor values ​​corresponding to each of the pricing factors; a second permutation and combination module, for permuting and combining each of the factor values ​​based on a combination algorithm to obtain a multidimensional matrix, wherein the multidimensional matrix is ​​used to represent the mapping relationship between pricing scenarios and pricing factor combinations; a fourth determination module, for determining, for any of the pricing scenarios, the second value range of the pricing factor combination based on the first value range of each of the factor values ​​in the pricing factor combination; and a second verification module, for verifying the actual pricing corresponding to each of the pricing scenarios based on the second value ranges.

[0009] In a fifth aspect, an embodiment of the present disclosure further provides an electronic device, which may include: a processor and a memory; wherein the memory stores a computer program, and the computer program is suitable for being loaded by the processor and executing the steps of the prediction data verification method of the first aspect mentioned above.

[0010] In a sixth aspect, an embodiment of the present disclosure further provides an electronic device, which may include: a processor and a memory; wherein the memory stores a computer program, and the computer program is suitable for being loaded by the processor and executing the steps of the method for verifying the prediction data of the second aspect mentioned above.

[0011] In a seventh aspect, an embodiment of the present disclosure provides a computer storage medium, which stores a plurality of instructions, and the instructions are suitable for being loaded by a processor and executing the steps of the method for checking the prediction data of the first aspect mentioned above.

[0012] In an eighth aspect, an embodiment of the present disclosure provides a computer storage medium storing a plurality of instructions, wherein the instructions are suitable for being loaded by a processor and executing the steps of the method for verifying the prediction data of the second aspect described above.

[0013] The beneficial effects brought about by the technical solutions provided by some embodiments of the present disclosure include at least the following: In one or more embodiments of the present disclosure, the various calculation factors in the target calculation process can be determined, and the factor values ​​of each calculation factor can be enumerated to obtain all the factor values ​​of each calculation factor. Then, by permuting and combining the factor values, the calculation factor combinations under different prediction scenarios are determined, and the second value range of the calculation factor combination is calculated. When the scenario prediction value exceeds the second value range, it means that an error occurs in the calculation process of the scenario prediction value, and an unreasonable prediction result is generated. In this way, unreasonable prediction results can be discovered in a timely manner based on the second value range, avoiding actual losses caused by users making corresponding decisions based on the prediction results. BRIEF DESCRIPTION OF THE DRAWINGS

[0014] FIG1 is a schematic diagram of the system architecture of a prediction data verification method provided by an embodiment of the present disclosure.

[0015] FIG2 is an overall flow chart of a method for verifying prediction data provided by an embodiment of the present disclosure.

[0016] FIG3 is a schematic diagram showing the relationship between calculation factors and factor values ​​provided by an embodiment of the present disclosure.

[0017] FIG4 is a schematic diagram showing the relationship between prediction scenarios and calculation factor combinations in a multidimensional matrix provided by an embodiment of the present disclosure.

[0018] FIG5 is an overall flow chart of another method for verifying prediction data provided by an embodiment of the present disclosure.

[0019] FIG6 is an overall flow chart of another method for verifying prediction data provided by an embodiment of the present disclosure.

[0020] FIG7 is an overall flow chart of another method for verifying prediction data provided by an embodiment of the present disclosure.

[0021] FIG8 is an overall flow chart of another method for verifying prediction data provided by an embodiment of the present disclosure.

[0022] FIG9 is an overall flow chart of another method for verifying prediction data provided by an embodiment of the present disclosure.

[0023] FIG10 is a schematic diagram of the structure of a prediction data verification device provided by an embodiment of the present disclosure.

[0024] FIG11 is a schematic structural diagram of another prediction data verification device provided by an embodiment of the present disclosure.

[0025] FIG12 is a schematic structural diagram of an electronic device provided in an embodiment of the present disclosure.

[0026] FIG13 is a schematic structural diagram of another electronic device provided by an embodiment of the present disclosure. DETAILED DESCRIPTION

[0027] The technical solutions in the embodiments of the present disclosure will be described clearly and completely below with reference to the accompanying drawings in the embodiments of the present disclosure.

[0028] The terms "first," "second," "third," and so on, in the specification and claims of this disclosure and the accompanying drawings are used to distinguish between different objects, not to describe a specific order. Furthermore, the terms "including," "having," and any variations thereof, are intended to cover non-exclusive inclusions. For example, a process, method, system, product, or apparatus comprising a series of steps or elements is not limited to the listed steps or elements, but may optionally include steps or elements not listed, or may optionally include other steps or elements inherent to the process, method, product, or apparatus.

[0029] The following description provides examples and does not limit the scope, applicability or examples set forth in the claims. Changes may be made to the function and arrangement of the elements described without departing from the scope of this disclosure. Various examples may appropriately omit, replace or add various processes or components. For example, the described method may be performed in an order different from the described order, and various steps may be added, omitted or combined. In addition, features described in some examples may be combined in other examples.

[0030] First, some of the embodiments of the present application will be explained to facilitate understanding by those skilled in the art.

[0031] Calculation factors: A specific set of variables that can explain and predict data results. Depending on the calculation field, it will consider different factors such as market demand, cost structure, product characteristics, etc. to determine a reasonable value range.

[0032] Pricing factor: Pricing factor is a calculation factor used in the insurance pricing forecasting process to adjust the price of a product or service to determine a reasonable price level.

[0033] Logical factor: In statistics and data analysis, a logical factor is a variable or factor that represents a binary logical value (True or False). Logical factors are often used to describe whether an event has occurred or whether a condition has been met.

[0034] Multidimensional matrix: A method and system for dealing with multi-index problems, which is a rectangular array formed by arranging elements of m rows and n columns.

[0035] Please refer to FIG1 , which shows a schematic diagram of the system architecture of a prediction data verification method provided by an embodiment of the present disclosure.

[0036] As shown in FIG1 , the system architecture of the prediction data verification method may include at least a terminal 100 , a server 200 and a network 300 .

[0037] Terminal 100 includes, but is not limited to, electronic devices such as smartphones, desktop computers, tablet computers, laptops, smart speakers, digital assistants, augmented reality (AR) / virtual reality (VR) devices, and smart wearable devices. It may also be software running on these electronic devices, such as applications. Optionally, the operating system running on the electronic device may include, but is not limited to, Android, iOS, Linux, Windows, etc. Optionally, terminal 100 provides a scene prediction value verification service to the user. Terminal 100 may obtain scene prediction value verification information through an application program interface.

[0038] The server 200 can provide background services for the terminal 100, select a corresponding second value range to verify the scenario prediction value based on the prediction data verification information sent by the terminal 100, and send the verification result back to the terminal 100. The server 200 can be an independent physical server, or a server cluster or distributed system composed of multiple physical servers, or a cloud server that provides 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 (Content Delivery Network), and big data and artificial intelligence platforms.

[0039] The network 300 is used to provide a medium for a communication link between the terminal 100 and the server 200. The network 300 may include various connection types, such as wired or wireless communication links or fiber optic cables.

[0040] In addition, it should be noted that what is shown in FIG1 is only a system provided by the present disclosure. In practical applications, other systems may also be included, for example, more terminals may be included.

[0041] In the embodiment of the present disclosure, the terminal 100 and the server 200 may be directly or indirectly connected via wired or wireless communication, which is not limited in the present disclosure.

[0042] Next, please refer to FIG. 2 , which shows an overall flow chart of a method for verifying prediction data provided by an embodiment of the present disclosure. The method for verifying prediction data can be used in a server.

[0043] As shown in FIG2 , the method for verifying the prediction data may include at least the following steps.

[0044] Step 201: Determine each calculation factor corresponding to the target calculation process, and enumerate all factor values ​​corresponding to each calculation factor.

[0045] Different computing fields (such as social science research, market research, medical biostatistics, engineering quality control, insurance pricing, etc.) can be set up with different target computing processes. Each target computing process can be set up with at least one computing factor that affects the prediction results according to the characteristics of its own field category, so the server can determine each computing factor through the target computing process. Each computing factor can be a key-value pair, containing two elements: factor key and factor value. The factor key can be used to indicate the type of factor that specifically affects the prediction results represented by the computing factor, and the factor value can be used to indicate the value of the computing factor under different conditions. Among them, the factor key and the factor value can be a 1:n relationship, that is, each factor key corresponds to one or more factor values. As shown in Figure 3, the factor values ​​corresponding to factor key A can be A1, A2, A3, and the factor values ​​corresponding to factor key B can be B1, B2, B3, and so on.

[0046] In the target calculation process, several factors that affect the predicted value can be manually set in advance for the target calculation process based on the specific category of the calculation field. The individual factors separated from these factors can be regarded as a calculation factor determined from the target calculation process. In other embodiments, the corresponding calculation factor can also be manually pre-set for each target calculation process. The specific method for determining the calculation factor is not limited here.

[0047] After determining the calculation factors of the target calculation process, the server can enumerate each calculation factor in the historical data through the offline stored historical data to enumerate all enumeration fields corresponding to each calculation factor, that is, enumerate all factor values ​​corresponding to each calculation factor.

[0048] As an example, taking mass travel research in the social sciences as an example, the calculation factors corresponding to the target calculation process may include travel destination, travel distance, travel origin, travel vehicle, number of travelers, travel duration, etc. For example, for travel distance, different distance intervals can be mapped to different factor values. For travel destinations, different destinations can be mapped to different factor values.

[0049] As another example, taking the catering research in the field of market research as an example, the calculation factors corresponding to the target calculation process may include the type of restaurant (Chinese food, Western food, hot pot, etc.), the average price of the restaurant, the business district where the restaurant is located, the daily traffic volume of the restaurant, the monthly cost of the restaurant, etc. For example, for the restaurant type, different restaurant types can be corresponding to different factor values.

[0050] As another example, taking disease statistics in medical biostatistics as an example, the calculation factors corresponding to the target calculation process may include disease type, disease onset temperature, disease onset pressure, age of the patient population, and weight of the patient population. For example, taking the age of the patient population as an example, different age ranges can correspond to different factor values.

[0051] As another example, taking building quality control in the field of engineering quality control as an example, the calculation factors corresponding to the target calculation process may include building height, building floor area, load-bearing pressure of the building's load-bearing walls, building structural strength, indoor air quality, building materials, building moisture-proofing technology, etc. For example, different building moisture-proofing technologies can correspond to different factor values. Different building height ranges can also correspond to different factor values.

[0052] As another example, taking auto insurance in the insurance pricing field as an example, the calculation factors corresponding to the target calculation process may include the type of vehicle (for example, motorcycles, private cars, tractors, commercial trucks, commercial buses, non-commercial trucks, non-commercial buses, etc.), the price of the vehicle, the cumulative mileage of the vehicle, the weight range of the vehicle, the nature of the vehicle's use (for example, family use, use by institutions, use by enterprises, rental use, freight use, teaching use, examination use, etc.), the age of the vehicle, the gender of the driver, the driving experience of the driver, etc. For example, for the type of vehicle, motorcycles, private cars, tractors and other specific types can correspond to different factor values. For the price of the vehicle, different price ranges of the price correspond to different factor values. For the age of the vehicle, different years correspond to different factor values, etc.

[0053] Step 203: Arrange and combine the factor values ​​based on a combination algorithm to obtain a multi-dimensional matrix.

[0054] Among them, the multidimensional matrix is ​​used to represent the mapping relationship between the prediction scenario and the calculation factor combination.

[0055] In practice, to improve prediction accuracy, multiple calculation factors are generally considered comprehensively. Different calculation factors are selected for different prediction scenarios and arranged and combined to create a corresponding calculation factor combination for each prediction scenario. Based on the calculation factor combination corresponding to the prediction scenario, the scenario prediction value corresponding to that prediction scenario can be quickly determined. As shown in Figure 4, after the factor values ​​of each calculation factor are arranged and combined using a combination algorithm to determine the mapping relationship between each prediction scenario and each calculation factor combination, this can be represented in a multidimensional matrix, allowing the server to quickly find the corresponding calculation factor combination through the multidimensional matrix to calculate the prediction value.

[0056] The combination algorithm can use machine learning algorithms, such as cluster analysis, decision trees, neural networks, etc., to discover hidden patterns and correlations from multiple calculation factors, and then combine these calculation factors into a prediction model, each of which can correspond to a prediction scenario. In other embodiments, the combination algorithm can also be other algorithms, such as correlation analysis methods, which analyze the correlation between the calculation factors and the prediction scenario and select the calculation factors with the greater correlation for combination.

[0057] Step 205: For any of the prediction scenarios, determine a second value range of the calculation factor combination based on the first value range of each factor value in the calculation factor combination.

[0058] The calculation factors are factors that are artificially set and determined according to the target calculation process and have a high correlation with the calculation field and affect the prediction results. However, in addition to these, there may be some factors that have relatively small effects and are not used as calculation factors. These factors can be used as derivative features of certain factor values. Derivative features can be regarded as different calculation dimensions of factor values, which are used to affect the specific values ​​of factor values. Therefore, in each calculation factor, the specific numerical values ​​corresponding to at least some factor values ​​are not fixed values, but there is a value range. Therefore, the server can calculate the values ​​of the calculation factor combination through the first value range of each factor value to obtain the second value range. The second value range can represent the numerical range of the calculation factor combination under the influence of different derivative features.

[0059] As an example, for the aforementioned social science research on mass travel, assuming the calculation factor is trip duration, the corresponding factor value can be 5 hours, 8 hours, 10 hours, and so on. For the same 5 hours, different travel time periods will have different impacts on travel outcomes. Therefore, the travel time period can be a derivative feature of the factor value.

[0060] As another example, for the aforementioned market research on catering, assuming the calculation factor is the monthly cost of a restaurant, the corresponding factor value can be 50,000, 100,000, 150,000, etc., depending on the specific type of ingredients. For the same ingredient cost, the freshness of the ingredients will affect the actual loss rate of the ingredients, resulting in changes in the final estimated revenue value for the same cost. Therefore, the freshness of the ingredients can be a derivative feature of the factor value.

[0061] As another example, for the aforementioned medical biostatistics-based disease statistics, assuming the calculation factor is the patient's weight, the calculation factor could be 40 kg, 50 kg, 60 kg, and so on. However, for patients with the same weight, their body fat percentages could vary, which would also affect the final disease outcome prediction. Therefore, the patient's body fat percentage could be a derivative feature of the factor value.

[0062] As another example, in the aforementioned field of engineering quality control, for example, consider the calculation factor for house quality control. For example, if the calculation factor is the house moisture-proofing process, different factor values ​​can be assigned to different processes. However, for the same house moisture-proofing process, differences in the house's environmental climate will affect the actual effectiveness of the process and, in turn, the estimated house quality. Therefore, the house's environmental climate can be a derivative feature of this factor value.

[0063] As another example, for the aforementioned auto insurance in the insurance pricing field, assuming that the calculation factor is the age of the vehicle, the factor value corresponding to the calculation factor can be 1 year, 2 years, and so on. For the same car that has been used for 1 year, the degree of depreciation may also be different. Therefore, the valuation of a one-year vehicle with high depreciation may be 200,000 yuan, while the valuation of a one-year vehicle with low depreciation may be 210,000 yuan. The degree of depreciation may be a derivative feature of the factor value.

[0064] Step 207: Check the scene prediction value corresponding to each of the prediction scenes based on each of the second value ranges.

[0065] Since the second value range is the value range of the calculation factor combination calculated after considering the derivative characteristics of the calculation factors, the maximum and minimum values ​​of the second value range can be expressed as the upper and lower limits of the calculation value corresponding to the calculation factor combination. If the server's scenario prediction value for the calculation factor combination exceeds this range, it means that the current calculation result may be an unreasonable result due to a parameter calculation error. The server can use this to check the scenario prediction value according to the corresponding second value range when generating a new scenario prediction value for each prediction scenario to determine whether the current prediction result is reasonable.

[0066] As an optional embodiment of the present disclosure, the step 207 of checking the scene prediction value corresponding to each of the prediction scenarios based on each of the second value ranges includes: for any of the prediction scenarios, obtaining the scene prediction value corresponding to the prediction scenario, and comparing the scene prediction value with the second value range corresponding to the prediction scenario; when the scene prediction value is within the second value range, generating first verification information, the first verification information is used to indicate that the scene prediction value is normal; when the scene prediction value is not within the second value range, generating second verification information, the second verification information is used to indicate that the scene prediction value is abnormal.

[0067] The server generates different verification information based on the comparison between the scenario prediction value and the second value range. If the scenario prediction value is within the second value range, the server considers the prediction result normal and generates first verification information. If the scenario prediction value is not within the second value range, the server considers the prediction result abnormal and generates second verification information. After the verification information is generated, it can be sent to the terminal of the relevant staff so that the staff can promptly identify the abnormal prediction result. The verification information can be a text message containing a text description of "This prediction result is abnormal / normal" or a voice broadcast information, etc.

[0068] In the embodiment of the present disclosure, each calculation factor in the target calculation process is determined, and the factor value of each calculation factor is enumerated to obtain all the factor values ​​of each calculation factor. Then, by permuting and combining the factor values, the calculation factor combination under different prediction scenarios is determined, and the second value range of the calculation factor combination is determined. When the scenario prediction value exceeds the second value range, it means that an error has occurred in the calculation process of the scenario prediction value, resulting in an unreasonable prediction result. In this way, unreasonable prediction results can be discovered in a timely manner based on the second value range, avoiding actual losses caused by decisions based on the prediction results.

[0069] The foregoing description describes specific embodiments of the present disclosure. Other embodiments are within the scope of the appended claims. In some cases, the actions or steps recited in the claims can be performed in an order different from that described in the embodiments and still achieve the desired results. Furthermore, the processes depicted in the accompanying drawings do not necessarily require the specific order shown or the sequential order to achieve the desired results. In certain embodiments, multitasking and parallel processing are also possible or may be advantageous.

[0070] As another optional embodiment of the present disclosure, please refer to FIG5 , which shows an overall flow chart of another method for verifying prediction data provided by an embodiment of the present disclosure. The method for verifying prediction data can be used in a server.

[0071] As shown in FIG5 , the method for verifying the prediction data may include at least the following steps.

[0072] Step 501: Obtain a logical factor corresponding to a target calculation process, and determine various calculation factors based on the logical factor.

[0073] Logical factors can be pre-configured for the target calculation process. These factors describe whether a specific event or condition occurs at each step of the target calculation process. Examples include whether the transportation is a car, whether the restaurant is Chinese food, whether the patient is over 50 years old, whether the building is taller than 30 meters, and whether the driver has more than five years of driving experience. The server will retrieve these logical factors and use them to determine the various calculation factors. For example, if the restaurant is Chinese food, the calculation factor could be restaurant type; if the vehicle is a commercial vehicle, the calculation factor could be vehicle usage type; if the vehicle price is greater than 200,000 yuan, the calculation factor could be vehicle price, and so on. Determining the calculation factors based on logical factors can be done by extracting field keywords, or by pre-configuring the corresponding calculation factors for each logical factor. Other methods are also possible and are not limited here.

[0074] Step 503: Obtain historical calculation data corresponding to each calculation factor.

[0075] The server will retrieve the historical calculation data for each calculation factor from the database based on the specific data category corresponding to that category, so that it can traverse and determine all the factor values ​​corresponding to the calculation factor based on the historical calculation data. For example, for the calculation factor of travel destination, the server will obtain historical calculation data related to the travel destination, that is, the calculation process took the travel destination into account; for the calculation factor of building materials, the server will obtain historical calculation data related to building materials, that is, the calculation process took building materials into account; for the calculation factor of vehicle age, the server will obtain historical calculation data related to vehicle age, that is, the calculation process took vehicle age into account, and so on.

[0076] Step 505: De-duplicate features of the historical calculation data, and enumerate to obtain all factor values ​​corresponding to the calculation factors.

[0077] The calculation factor characteristics of different factor values ​​are different.

[0078] There will be a lot of similar data in the historical calculation data, and the server can perform feature deduplication on the historical calculation data based on the calculation factor characteristics. Taking the historical calculation data corresponding to the vehicle age as an example, each historical calculation data has a vehicle age feature (for example, 1 year vehicle age, 2 years vehicle age, etc.). The server will traverse the historical calculation data to perform feature deduplication on the historical calculation data based on the calculation factor feature (i.e., vehicle age feature), and then deduplication multiple historical calculation data with the same vehicle age, and only retain one historical calculation data with the same vehicle age. Through feature deduplication, the server can enumerate all the calculation factor features of the calculation factor in the historical calculation data, and these calculation factor features can be used to represent the factor value corresponding to the calculation factor. For example, if after feature deduplication, there are a total of 20 historical pricing data corresponding to 1 year vehicle age to 20 years vehicle age, then it can be considered that the factor values ​​enumerated for the calculation factor are 20 in total, namely 1 year vehicle age to 20 years vehicle age.

[0079] Step 507: traverse the calculation factors based on the combination algorithm, arrange and combine the calculation factors into calculation factor combinations, and construct a multidimensional matrix based on the calculation factor combinations.

[0080] Each calculation factor has at most one factor value in each calculation factor combination.

[0081] The server traverses and calculates each calculation factor through a combination algorithm, and can determine multiple calculation factors with high correlation. By combining these calculation factors, a calculation factor combination can be obtained. Each calculation factor combination can correspond to a combination under an actual prediction scenario. By summarizing the calculation factor combinations, a multidimensional matrix can be constructed. The multidimensional matrix can represent the mapping relationship between each prediction scenario and each calculation factor combination. During the actual verification process, the server can obtain the prediction scenario contained in the prediction information corresponding to the scenario prediction value, and query it in the multidimensional matrix to obtain the corresponding calculation factor combination, and then judge whether the actual scenario prediction value is abnormal based on the second value range of the calculation factor combination.

[0082] As an example, a calculation factor combination is a combination of calculation factors, where each calculation factor may select at most one factor value for the combination. For example, the calculation factor combination may be for a family vehicle that is one year old and has a driver with three years of driving experience. In other embodiments, more than one factor value may be selected for the combination from at least some of the calculation factors, and this is not limited here.

[0083] Step 509: For any of the prediction scenarios, determine a second value range of the calculation factor combination based on the first value range of each factor value in the calculation factor combination.

[0084] Step 509 may refer to step 205 and will not be described again here.

[0085] Step 511: Check the scene prediction value corresponding to each of the prediction scenes based on each of the second value ranges.

[0086] Step 511 may refer to step 207 and will not be described again here.

[0087] In the disclosed embodiment, the server determines the various calculation factors in the target calculation process based on the logical factors, obtains historical calculation data, and enumerates and removes duplicates from the factor values ​​of each calculation factor to obtain all the factor values ​​of each calculation factor. Next, by permuting and combining the factor values, the calculation factor combinations for different prediction scenarios are determined, and a multidimensional matrix is ​​constructed using the calculation factor combinations. By removing duplicates and enumerating historical calculation data, all the factor values ​​corresponding to each calculation factor can be quickly and comprehensively determined, thereby generating more comprehensive and accurate combinations of calculation factors.

[0088] The foregoing description describes specific embodiments of the present disclosure. Other embodiments are within the scope of the appended claims. In some cases, the actions or steps recited in the claims can be performed in an order different from that described in the embodiments and still achieve the desired results. Furthermore, the processes depicted in the accompanying drawings do not necessarily require the specific order shown or the sequential order to achieve the desired results. In certain embodiments, multitasking and parallel processing are also possible or may be advantageous.

[0089] As another optional embodiment of the present disclosure, please refer to FIG6 , which shows an overall flow chart of another prediction data verification method provided by the embodiment of the present disclosure. The prediction data verification method can be used in a server.

[0090] As shown in FIG6 , the method for verifying the prediction data may include at least the following steps.

[0091] Step 601: Determine the calculation factors corresponding to the target calculation process, and enumerate all factor values ​​corresponding to each calculation factor. Step 601 can refer to step 201 and will not be repeated here.

[0092] Step 603: Based on a combination algorithm, the factor values ​​are arranged and combined to obtain a multi-dimensional matrix. Step 603 can refer to step 203 and will not be repeated here.

[0093] Step 605: For any of the pricing scenarios, calculate the first value range of the factor value on each target date based on the calculation dimension of the factor value.

[0094] The target date includes a standard date and a special date, and the factor value includes a target calculation dimension, which has different values ​​for different target dates.

[0095] A factor value may include at least one calculation dimension, such as the age of a restaurant, the degree of disease variation, the brand of building materials, the degree of damage to building materials, the fuel type of a vehicle, whether the vehicle is a Hong Kong or Macau vehicle, and the availability of additional vehicle maintenance services. Different calculation dimensions can be assigned different calculation values, so the factor value can vary depending on the calculation dimension. The server can calculate the maximum and minimum values ​​of the factor value based on the calculation dimension and set a first value range for the factor value using the maximum and minimum values ​​as upper and lower limits.

[0096] In addition, for some target calculation dimensions (such as house building material brands, automobile fuel models, and additional vehicle maintenance services), the values ​​corresponding to preset special dates (generally promotional dates, such as Double Eleven, Spring Festival, etc.) and ordinary standard dates are different, resulting in different values ​​corresponding to the target calculation dimensions on different dates. If the dates are not distinguished, the first value range corresponding to the factor value finally calculated is the union of the value range of the special date and the value range of the standard date. It is easy to calculate the predicted value that can only be achieved on the special date when the current date is the standard date, but it is not identified as abnormal prediction data because it falls within the range of the union. Therefore, the server will calculate the first value range under different target dates separately to distinguish the first value range of the standard date from the first value range of the special date.

[0097] As an optional embodiment of the present disclosure, the first value range is determined based on an upper limit and a lower limit of a hash value of the factor value, and the hash value is determined based on the numerical product of each of the calculation dimensions.

[0098] The server can calculate the hash value corresponding to the factor value through the hash function. The maximum and minimum values ​​of the hash value can be used as the upper and lower limits of the first value range. The hash value can be determined by multiplying the values ​​corresponding to the various calculation dimensions.

[0099] Step 607: Determine a second value range of the calculation factor combination based on the sum of the first value ranges of the factor values ​​in the calculation factor combination.

[0100] After the server calculates the first value range corresponding to each factor value respectively, it can obtain the second value range corresponding to the calculation factor combination by summing up the first value ranges of each factor value in the calculation factor combination.

[0101] The server may also set different weights for each factor value based on the relevance of the factor values, and finally obtain the second value range by weighted summation of the first value range. In other embodiments, other methods for calculating the second value range may also be used, which are not limited here.

[0102] Step 609: Select each of the second value ranges that matches the current date.

[0103] The server determines the current date when the scenario prediction value is generated, and determines whether the current date corresponds to a standard date or a special date. It then selects a second value range corresponding to the standard date or special date that matches the current date as the judgment range for subsequent verification. This allows the server to more accurately determine whether the scenario prediction value is reasonable for the current date.

[0104] Step 611: Check the scene prediction value corresponding to each of the prediction scenes based on each of the second value ranges.

[0105] Step 611 may refer to step 207 and will not be described again here.

[0106] In this embodiment, the server can calculate the first value range of the factor value using the calculation dimension, and can calculate the second value range corresponding to the standard date and the special date respectively based on the different values ​​of the target calculation dimension on the standard date and the special date. Finally, the server can select the second value range that matches the current date to verify the scenario prediction value, thereby more accurately determining whether the scenario prediction value is reasonable for the current date.

[0107] The foregoing description describes specific embodiments of the present disclosure. Other embodiments are within the scope of the appended claims. In some cases, the actions or steps recited in the claims can be performed in an order different from that described in the embodiments and still achieve the desired results. Furthermore, the processes depicted in the accompanying drawings do not necessarily require the specific order shown or the sequential order to achieve the desired results. In certain embodiments, multitasking and parallel processing are also possible or may be advantageous.

[0108] Next, please refer to FIG. 7 , which shows an overall flow chart of a method for verifying prediction data provided by an embodiment of the present disclosure. The method for verifying prediction data can be used in a server.

[0109] As shown in FIG7 , the method for verifying the prediction data may include at least the following steps.

[0110] Step 701: Determine each pricing factor corresponding to the insurance pricing process, and enumerate all factor values ​​corresponding to each pricing factor.

[0111] Different insurances (such as e-commerce freight insurance, car insurance, card insurance, etc.) can be set up with different insurance pricing processes. Each insurance pricing process can be set up with at least one pricing factor that affects insurance pricing according to the characteristics of its own insurance category, so the server can determine each pricing factor through the insurance pricing process. Each pricing factor can be a key-value pair, containing two elements: factor key and factor value. The factor key can be used to indicate the specific type of factor affecting pricing represented by the pricing factor, and the factor value can be used to indicate the value of the pricing factor under different conditions. Among them, the factor key and the factor value can be a 1:n relationship, that is, each factor key corresponds to one or more factor values. As shown in Figure 3, the factor values ​​corresponding to factor key A can be A1, A2, A3, and the factor values ​​corresponding to factor key B can be B1, B2, B3, and so on.

[0112] In the insurance pricing process, several factors influencing insurance pricing can be manually pre-set based on the specific insurance type. Individual factors derived from these factors can be considered pricing factors determined from the insurance pricing process. In other embodiments, corresponding pricing factors can be manually pre-set for each insurance pricing process. The specific method for determining pricing factors is not limited herein.

[0113] After determining the pricing factors of the insurance pricing process, the server can enumerate each pricing factor in the historical data through offline stored historical data to enumerate all enumeration fields corresponding to each pricing factor, that is, enumerate all factor values ​​corresponding to each pricing factor.

[0114] As an example, taking auto insurance as an example, the pricing factors corresponding to the auto insurance pricing process may include the type of vehicle (for example, motorcycles, private cars, tractors, commercial trucks, commercial buses, non-commercial trucks, non-commercial buses, etc.), the price of the vehicle, the cumulative mileage of the vehicle, the weight range of the vehicle, the nature of the vehicle's use (for example, family use, use by institutions, use by enterprises, rental use, freight use, teaching use, examination use, etc.), the age of the vehicle, the gender of the driver, the driving experience of the driver, etc. For example, for the type of vehicle, motorcycles, private cars, tractors and other specific types can correspond to different factor values. For the price of the vehicle, different price ranges correspond to different factor values. For the age of the vehicle, different years correspond to different factor values, etc.

[0115] Step 703: Arrange and combine the factor values ​​based on a combination algorithm to obtain a multi-dimensional matrix.

[0116] Among them, the multidimensional matrix is ​​used to represent the mapping relationship between pricing scenarios and pricing factor combinations.

[0117] In practice, to improve pricing accuracy, multiple pricing factors are generally considered comprehensively. Different pricing factors are selected for different pricing scenarios and permuted and combined to create a corresponding pricing factor combination for each pricing scenario. Based on the pricing factor combination corresponding to the pricing scenario, the insurance pricing corresponding to that pricing scenario can be quickly determined. As shown in Figure 4, after permuting and combining the factor values ​​of each pricing factor using a combinatorial algorithm to determine the mapping relationship between each pricing scenario and each pricing factor combination, this can be represented as a multidimensional matrix, allowing the server to quickly locate the corresponding pricing factor combination for pricing.

[0118] The combination algorithm can utilize machine learning algorithms, such as cluster analysis, decision trees, and neural networks, to discover hidden patterns and correlations among multiple pricing factors, and then combine these pricing factors into a prediction model. Each prediction model can correspond to a pricing scenario. In other embodiments, the combination algorithm can also utilize other algorithms, such as correlation analysis methods, which analyze the correlation between pricing factors and pricing scenarios and select pricing factors with greater correlation for combination.

[0119] Step 705: For any of the pricing scenarios, determine a fourth value range of the pricing factor combination based on the third value range of each factor value in the pricing factor combination.

[0120] Pricing factors are manually set and determined based on the insurance pricing process to influence pricing for a specific type of insurance. However, there may also be other factors with relatively minor impact that are not considered pricing factors. These factors can serve as derived characteristics of certain factor values. Derived characteristics can be considered different pricing dimensions of factor values, influencing the specific value of the factor values. Therefore, for each pricing factor, the specific price corresponding to at least some factor values ​​is not fixed but rather has a range of values. For example, if the pricing factor is the age of the vehicle, the corresponding factor value could be 1 year, 2 years, and so on. However, even for the same car with one year of use, the degree of depreciation may vary. Therefore, a one-year-old vehicle with high depreciation may be valued at 200,000 yuan, while a one-year-old vehicle with low depreciation may be valued at 210,000 yuan. The degree of depreciation can be a derived characteristic of the factor value. Therefore, the server can calculate the value of the pricing factor combination based on the third value range of each factor value to obtain a fourth value range. This fourth value range represents the range of values ​​for the pricing factor combination under the influence of different derived characteristics.

[0121] Step 707: Verify the actual pricing corresponding to each pricing scenario based on each fourth value range.

[0122] Since the fourth value range is the value range of the pricing factor combination calculated after considering the derivative characteristics of the pricing factors, the maximum and minimum values ​​of the fourth value range can be expressed as the upper and lower limits of the pricing price corresponding to the pricing factor combination. If the server's actual pricing for the pricing factor combination exceeds this range, it means that the current pricing may be unreasonable due to parameter calculation errors. The server can use this to check the actual pricing against the corresponding fourth value range when generating new actual pricing for each pricing scenario to determine whether the current pricing is reasonable.

[0123] As an optional embodiment of the present disclosure, the checking of the actual pricing corresponding to each pricing scenario based on each fourth value range in step 707 includes: for any pricing scenario, obtaining the actual pricing corresponding to the pricing scenario, and comparing the actual pricing with the fourth value range corresponding to the pricing scenario; when the actual pricing is within the fourth value range, generating third verification information, the third verification information is used to indicate that the actual pricing is normal; when the actual pricing is not within the fourth value range, generating fourth verification information, the fourth verification information is used to indicate that the actual pricing is abnormal.

[0124] The server will generate different verification information based on the comparison between the actual price and the fourth value range. If the actual price is within the fourth value range, the server will consider the current pricing to be normal and will generate third verification information. If the actual price is not within the fourth value range, the server will consider the current pricing to be abnormal and will generate fourth verification information. After the verification information is generated, it can be sent to the terminal of the relevant staff so that the staff can promptly identify the abnormal pricing problem. The verification information can be a text message containing a text description of "this pricing is abnormal / normal", or it can be a voice broadcast information, etc.

[0125] In the disclosed embodiment, each pricing factor in the insurance pricing process is determined, and the factor values ​​of each pricing factor are enumerated to obtain the total factor values ​​of each pricing factor. Next, by permuting and combining the factor values, pricing factor combinations for different pricing scenarios are determined, and a fourth value range for the pricing factor combinations is calculated. If the actual price exceeds the fourth value range, it indicates that an error occurred in the calculation process of the actual price, resulting in unreasonable pricing. Therefore, based on the fourth value range, unreasonable pricing can be promptly identified, avoiding losses to users and / or insurance companies.

[0126] The foregoing description describes specific embodiments of the present disclosure. Other embodiments are within the scope of the appended claims. In some cases, the actions or steps recited in the claims can be performed in an order different from that described in the embodiments and still achieve the desired results. Furthermore, the processes depicted in the accompanying drawings do not necessarily require the specific order shown or the sequential order to achieve the desired results. In certain embodiments, multitasking and parallel processing are also possible or may be advantageous.

[0127] As another optional embodiment of the present disclosure, please refer to FIG8 , which shows an overall flow chart of another method for verifying prediction data provided by an embodiment of the present disclosure. The method for verifying prediction data can be used in a server.

[0128] As shown in FIG8 , the method for verifying the prediction data may include at least the following steps.

[0129] Step 801: Obtain logical factors corresponding to the insurance pricing process, and determine various pricing factors based on the logical factors.

[0130] Logical factors can be set in advance for the insurance pricing process. Logical factors are descriptive information about whether a specific event / condition occurs in each process step of the insurance pricing process, such as whether the vehicle is a commercial vehicle, whether the vehicle price is greater than 200,000 yuan, whether the driver's driving experience is greater than 5 years, etc. The server will obtain these logical factors and determine the various pricing factors based on the logical factors. For example, based on whether the vehicle is a commercial vehicle, the pricing factor can be determined as the vehicle usage type, and based on whether the vehicle price is greater than 200,000 yuan, the pricing factor can be determined as the vehicle price, and so on. Among them, the specific way to determine the pricing factor based on the logical factor can be to extract the field keywords, or the staff can set the corresponding pricing factor for each logical factor in advance, and the pricing factor can be directly determined after the logical factor is determined. It can also be other determination methods, which are not limited here.

[0131] Step 803: Obtain historical pricing data corresponding to each pricing factor respectively.

[0132] The server retrieves historical pricing data for each pricing factor from the database based on the specific data category. This allows the server to determine all the factor values ​​corresponding to the pricing factor based on this historical pricing data. For example, for a pricing factor based on vehicle age, the server will retrieve historical pricing data related to vehicle age, meaning that it takes vehicle age into account during the pricing process.

[0133] Step 805: De-duplicate the historical pricing data and enumerate all factor values ​​corresponding to the pricing factors.

[0134] Among them, the pricing factor characteristics of different factor values ​​are different.

[0135] There will be a lot of similar data in the historical pricing data, and the server can deduplicate the historical pricing data based on the pricing factor characteristics. Taking the historical pricing data corresponding to the age of the vehicle as an example, each historical pricing data has a vehicle age feature (for example, 1 year of age, 2 years of age, etc.). The server will traverse the historical pricing data to deduplicate the historical pricing data based on the pricing factor feature (i.e., vehicle age feature), and then deduplicate multiple historical pricing data with the same age of the vehicle, and only retain one historical pricing data with the same age of the vehicle. Through feature deduplication, the server can enumerate all the pricing factor features of the pricing factor in the historical pricing data, and these pricing factor features can be used to represent the factor value corresponding to the pricing factor. For example, if after feature deduplication, there are a total of 20 historical pricing data corresponding to vehicle ages of 1 year to 20 years, then it can be considered that the factor values ​​enumerated for the pricing factor are 20 in total, namely, vehicle ages of 1 year to 20 years.

[0136] Step 807: traverse the pricing factors based on the combination algorithm, arrange and combine the pricing factors into pricing factor combinations, and construct a multidimensional matrix based on the pricing factor combinations.

[0137] Each of the pricing factors has at most one factor value in each of the pricing factor combinations.

[0138] The server traverses and calculates each pricing factor through a combination algorithm, and can determine multiple pricing factors with high correlation. By combining these pricing factors, a pricing factor combination can be obtained. Each pricing factor combination can correspond to a combination under an actual pricing scenario. By summarizing the pricing factor combinations, a multidimensional matrix can be constructed. The multidimensional matrix can represent the mapping relationship between each pricing scenario and each pricing factor combination. During the actual verification process, the server can obtain the pricing scenario contained in the pricing information corresponding to the actual pricing, query it in the multidimensional matrix, and obtain the corresponding pricing factor combination. Then, based on the fourth value range of the pricing factor combination, it can determine whether the actual pricing is abnormal.

[0139] As an example, a pricing factor combination is a combination of pricing factors, where at most one factor value is selected for each pricing factor. For example, the pricing factor combination may be for a one-year-old family vehicle with a driver who has three years of driving experience. In other embodiments, more than one factor value may be selected for at least some of the pricing factors, and this is not limited here.

[0140] Step 809: For any of the pricing scenarios, determine a fourth value range of the pricing factor combination based on the third value range of each factor value in the pricing factor combination.

[0141] Step 809 may refer to step 705 and will not be described again here.

[0142] Step 811: Verify the actual pricing corresponding to each of the pricing scenarios based on each of the second value ranges.

[0143] Step 811 may refer to step 707 and will not be described again here.

[0144] In the disclosed embodiment, the server determines the various pricing factors in the insurance pricing process based on logical factors. It then obtains historical pricing data, enumerates and deduplicates the factor values ​​for each pricing factor, and obtains the complete factor values ​​for each pricing factor. Next, by permuting and combining the factor values, it determines the pricing factor combinations for different pricing scenarios and constructs a multidimensional matrix based on these pricing factor combinations. By deduplicating and enumerating historical pricing data, it is possible to quickly and comprehensively determine the complete factor values ​​for each pricing factor, thereby generating more comprehensive and accurate pricing factor combinations.

[0145] The foregoing description describes specific embodiments of the present disclosure. Other embodiments are within the scope of the appended claims. In some cases, the actions or steps recited in the claims can be performed in an order different from that described in the embodiments and still achieve the desired results. Furthermore, the processes depicted in the accompanying drawings do not necessarily require the specific order shown or the sequential order to achieve the desired results. In certain embodiments, multitasking and parallel processing are also possible or may be advantageous.

[0146] As another optional embodiment of the present disclosure, please refer to FIG9 , which shows an overall flow chart of another method for verifying prediction data provided by an embodiment of the present disclosure. The method for verifying prediction data can be used in a server.

[0147] As shown in FIG9 , the method for verifying the prediction data may include at least the following steps.

[0148] Step 901: Determine each pricing factor corresponding to the insurance pricing process, and enumerate all factor values ​​corresponding to each pricing factor.

[0149] Step 901 may refer to step 701 and will not be described again here.

[0150] Step 903: Arrange and combine the factor values ​​based on a combination algorithm to obtain a multi-dimensional matrix.

[0151] Step 903 may refer to step 703 and will not be described again here.

[0152] Step 905: For any of the pricing scenarios, calculate the third value range of the factor value at each target date based on the pricing dimension of the factor value.

[0153] The target dates include daily dates and promotional dates, and the factor values ​​include a target pricing dimension, which has different values ​​for different target dates.

[0154] The factor value may include at least one pricing dimension, such as the degree of damage, fuel type, whether the vehicle is a Hong Kong or Macau vehicle, and additional maintenance services. Different pricing dimensions can have different pricing values, so the factor value may vary depending on the pricing dimension. The server may calculate the maximum and minimum values ​​of the factor value based on the pricing dimension and set a third value range for the factor value using the maximum and minimum values ​​as upper and lower limits.

[0155] In addition, for some target pricing dimensions (such as fuel models, additional maintenance services), their prices on preset promotional dates (such as Double Eleven, Spring Festival, etc.) and ordinary daily dates are different, resulting in different values ​​corresponding to the target pricing dimensions on different dates. If the dates are not distinguished, the third value range corresponding to the factor value finally calculated is the union of the value range of the promotion date and the value range of the daily date. It is easy to calculate the price that can only be achieved on the promotion date on the daily date, but it is not identified as abnormal pricing because it falls within the range of the union. Therefore, the server will calculate the third value range under different target dates separately to distinguish the third value range of the daily date from the third value range of the promotion date.

[0156] As an optional embodiment of the present disclosure, the third value range is determined based on an upper limit and a lower limit of a hash value of the factor value, and the hash value is determined based on the numerical product of each of the pricing dimensions.

[0157] The server can calculate the hash value corresponding to the factor value through a hash function. The maximum and minimum hash values ​​can serve as the upper and lower limits of the third value range. The hash value can be determined by multiplying the values ​​corresponding to the various pricing dimensions.

[0158] Step 907: Determine a fourth value range of the pricing factor combination based on the sum of the third value ranges of the factor values ​​in the pricing factor combination.

[0159] After the server calculates the third value range corresponding to each factor value respectively, it can obtain the fourth value range corresponding to the pricing factor combination by summing up the third value ranges of each factor value in the pricing factor combination.

[0160] The server may also set different weights for each factor value based on the relevance of the factor values, and finally obtain the fourth value range by weighted summation of the third value range. In other embodiments, other calculation methods for the fourth value range may also be used, which are not limited here.

[0161] Step 909: Select each of the fourth value ranges that matches the current date.

[0162] The server determines the current date when the actual price is generated, and whether the current date corresponds to a regular date or a promotional date. It then selects the fourth value range corresponding to the regular date or promotional date that matches the current date as the subsequent verification range. This allows the server to more accurately determine whether the actual price is reasonable given the current date.

[0163] Step 911: Verify the actual pricing corresponding to each of the pricing scenarios based on each of the fourth value ranges.

[0164] Step 911 may refer to step 707 and will not be described again here.

[0165] In the disclosed embodiment, the server can calculate the third value range of the factor value based on the pricing dimension, and can calculate the fourth value range corresponding to regular dates and promotional dates based on the different values ​​of the target pricing dimension on regular dates and promotional dates, respectively. Ultimately, the server can select the fourth value range that matches the current date to determine the actual price, thereby more accurately determining whether the actual price is reasonable for the current date.

[0166] The foregoing description describes specific embodiments of the present disclosure. Other embodiments are within the scope of the appended claims. In some cases, the actions or steps recited in the claims can be performed in an order different from that described in the embodiments and still achieve the desired results. Furthermore, the processes depicted in the accompanying drawings do not necessarily require the specific order shown or the sequential order to achieve the desired results. In certain embodiments, multitasking and parallel processing are also possible or may be advantageous.

[0167] Next, please refer to Figure 10, which shows a schematic diagram of the structure of a prediction data verification device provided by an embodiment of the present disclosure. It should be noted that the prediction data verification device shown in Figure 10 is used to execute the method of the embodiment shown in Figure 2 of this application. For ease of explanation, only the portion relevant to the embodiment of this application is shown. For specific technical details not disclosed, please refer to the embodiment shown in Figure 2 of this application.

[0168] As shown in Figure 10, the verification device for the predicted data may at least include: a first determination module 1001, used to determine the pricing factors corresponding to the insurance pricing process, and enumerate all factor values ​​corresponding to each of the pricing factors; a first permutation and combination module 1002, used to permutate and combine the factor values ​​based on a combination algorithm to obtain a multidimensional matrix, wherein the multidimensional matrix is ​​used to represent the mapping relationship between the pricing scenario and the pricing factor combination; a second determination module 1003, used to determine the second value range of the pricing factor combination based on the first value range of each factor value in the pricing factor combination for any of the pricing scenarios; a first verification module 1004, used to verify the actual pricing corresponding to each of the pricing scenarios based on the second value ranges.

[0169] As an option of the embodiment of the present disclosure, the first determination module 1001 is specifically configured to: obtain a logical factor corresponding to the insurance pricing process, and determine each pricing factor based on the logical factor.

[0170] As an optional embodiment of the present disclosure, the first determination module is further specifically used to: obtain historical pricing data corresponding to each of the pricing factors respectively; perform feature deduplication on the historical pricing data, and enumerate all factor values ​​corresponding to the pricing factors, wherein the pricing factor features of different factor values ​​are different.

[0171] As an optional embodiment of the present disclosure, the first permutation and combination module 1002 is specifically used to: traverse each of the pricing factors based on a combination algorithm, permutate and combine each of the pricing factors into pricing factor combinations, and construct a multidimensional matrix based on the pricing factor combinations, wherein each of the pricing factors has at most one factor value in each of the pricing factor combinations.

[0172] As an optional embodiment of the present disclosure, the second determination module 1003 is specifically used to: calculate the first value range of the factor value based on the pricing dimension of the factor value, and the factor value includes at least one of the pricing dimensions; determine the second value range of the pricing factor combination based on the sum of the first value ranges of each of the factor values ​​in the pricing factor combination.

[0173] As an optional embodiment of the present disclosure, the first value range is determined based on an upper limit and a lower limit of a hash value of the factor value, and the hash value is determined based on the numerical product of each of the pricing dimensions.

[0174] As an optional embodiment of the present disclosure, the second determination module 1003 is further specifically used to: calculate the first value range of the factor value under each target date based on the pricing dimension of the factor value, wherein the target date includes daily dates and promotion dates, and there is a target pricing dimension in the factor value, and the target pricing dimension has different values ​​on different target dates.

[0175] As an option in the embodiment of the present disclosure, the first checking module 1004 is specifically configured to select each of the second value ranges that matches the current date.

[0176] As an optional embodiment of the present disclosure, the first verification module 1004 is further specifically used to: for any of the pricing scenarios, obtain the actual pricing corresponding to the pricing scenario, and compare the actual pricing with the second value range corresponding to the pricing scenario; when the actual pricing is within the second value range, generate first verification information, and the first verification information is used to indicate that the actual pricing is normal; when the actual pricing is not within the second value range, generate second verification information, and the second verification information is used to indicate that the actual pricing is abnormal.

[0177] Those skilled in the art will clearly understand that the technical solutions of the embodiments of the present application can be implemented with the help of software and / or hardware. "Unit" and "module" in this disclosure refer to software and / or hardware that can independently perform or cooperate with other components to perform specific functions, where the hardware can be, for example, a field programmable gate array (FPGA), an integrated circuit (IC), etc.

[0178] Each processing unit and / or module in the embodiments of the present application may be implemented by an analog circuit that implements the functions described in the embodiments of the present application, or may be implemented by software that executes the functions described in the embodiments of the present application.

[0179] Next, please refer to Figure 11, which shows a schematic diagram of the structure of a prediction data verification device provided by an embodiment of the present disclosure. It should be noted that the prediction data verification device shown in Figure 11 is used to execute the method of the embodiment shown in Figure 7 of this application. For ease of explanation, only the portion relevant to the embodiment of this application is shown. For specific technical details not disclosed, please refer to the embodiment shown in Figure 7 of this application.

[0180] As shown in Figure 11, the verification device for the predicted data may at least include: a third determination module 1101, used to determine the pricing factors corresponding to the insurance pricing process, and enumerate all factor values ​​corresponding to each of the pricing factors; a second permutation and combination module 1102, used to permute and combine the factor values ​​based on a combination algorithm to obtain a multidimensional matrix, wherein the multidimensional matrix is ​​used to represent the mapping relationship between the pricing scenario and the pricing factor combination; a fourth determination module 1103, used to determine the fourth value range of the pricing factor combination for any of the pricing scenarios based on the third value range of each factor value in the pricing factor combination; a second verification module 1104, used to verify the actual pricing corresponding to each of the pricing scenarios based on the fourth value range.

[0181] As an option in the embodiment of the present disclosure, the third determination module 1101 is specifically configured to: obtain a logical factor corresponding to the insurance pricing process, and determine each pricing factor based on the logical factor.

[0182] As an optional embodiment of the present disclosure, the third determination module 1101 is further specifically used to: obtain historical pricing data corresponding to each of the pricing factors respectively; perform feature deduplication on the historical pricing data, and enumerate all factor values ​​corresponding to the pricing factors, wherein the pricing factor features of different factor values ​​are different.

[0183] As an optional embodiment of the present disclosure, the second permutation and combination module 1102 is specifically used to: traverse each of the pricing factors based on a combination algorithm, permutate and combine each of the pricing factors into pricing factor combinations, and construct a multidimensional matrix based on the pricing factor combinations, wherein each of the pricing factors has at most one factor value in each of the pricing factor combinations.

[0184] As an optional embodiment of the present disclosure, the fourth determination module 1103 is specifically used to: calculate the third value range of the factor value based on the pricing dimension of the factor value, and the factor value includes at least one of the pricing dimensions; determine the fourth value range of the pricing factor combination based on the sum of the third value ranges of each factor value in the pricing factor combination.

[0185] As an optional embodiment of the present disclosure, the third value range is determined based on the upper limit and the lower limit of the hash value of the factor value, and the hash value is determined based on the numerical product of each of the pricing dimensions.

[0186] As an optional embodiment of the present disclosure, the fourth determination module 1103 is further specifically used to: calculate the third value range of the factor value under each target date based on the pricing dimension of the factor value, wherein the target date includes daily dates and promotion dates, and there is a target pricing dimension in the factor value, and the target pricing dimension has different values ​​on different target dates.

[0187] As an option in the embodiment of the present disclosure, the second checking module 1104 is specifically configured to select each of the fourth value ranges that matches the current date.

[0188] As an optional embodiment of the present disclosure, the second verification module 1104 is further specifically used to: for any of the pricing scenarios, obtain the actual pricing corresponding to the pricing scenario, and compare the actual pricing with the third value range corresponding to the pricing scenario; when the actual pricing is within the fourth value range, generate third verification information, and the third verification information is used to indicate that the actual pricing is normal; when the actual pricing is not within the fourth value range, generate fourth verification information, and the fourth verification information is used to indicate that the actual pricing is abnormal.

[0189] Those skilled in the art will clearly understand that the technical solutions of the embodiments of the present application can be implemented with the help of software and / or hardware. "Unit" and "module" in this disclosure refer to software and / or hardware that can independently perform or cooperate with other components to perform specific functions, where the hardware can be, for example, a field programmable gate array (FPGA), an integrated circuit (IC), etc.

[0190] Each processing unit and / or module in the embodiments of the present application may be implemented by an analog circuit that implements the functions described in the embodiments of the present application, or may be implemented by software that executes the functions described in the embodiments of the present application.

[0191] Next, please refer to FIG12 , which shows a schematic structural diagram of an electronic device provided by an embodiment of the present disclosure.

[0192] As shown in FIG. 12 , the electronic device 1200 may include: at least one processor 1201 , at least one network interface 1204 , a user interface 1203 , a memory 1205 , and at least one communication bus 1202 .

[0193] The communication bus 1202 can be used to connect and communicate with each of the aforementioned components. The user interface 1203 can include buttons, and optional user interfaces can also include standard wired interfaces and wireless interfaces. The network interface 1204 can include, but is not limited to, a Bluetooth module, an NFC module, a Wi-Fi module, and the like.

[0194] The processor 1201 may include one or more processing cores. The processor 1201 uses various interfaces and lines to connect the various parts of the entire electronic device 1200, and executes various functions of the electronic device 1200 and processes data by running or executing instructions, programs, code sets or instruction sets stored in the memory 1205, and calling data stored in the memory 1205. Optionally, the processor 1201 can be implemented in at least one hardware form of DSP, FPGA, and PLA. The processor 1201 can integrate one or a combination of CPU, GPU, and modem. Among them, the CPU mainly processes the operating system, user interface, and application programs; the GPU is responsible for rendering and drawing the content to be displayed on the display screen; and the modem is used to handle wireless communications. It is understandable that the above-mentioned modem may not be integrated into the processor 1201, but may be implemented separately through a chip.

[0195] The memory 1205 may include RAM or ROM. Optionally, the memory 1205 includes a non-transitory computer-readable medium. The memory 1205 can be used to store instructions, programs, codes, code sets or instruction sets. The memory 1205 may include a program storage area and a data storage area, wherein the program storage area may store instructions for implementing an operating system, instructions for at least one function (such as a touch function, a sound playback function, an image playback function, etc.), instructions for implementing the above-mentioned various method embodiments, etc.; the data storage area may store data involved in the above-mentioned various method embodiments, etc. The memory 1205 may also be at least one storage device located away from the aforementioned processor 1201. As shown in Figure 12, the memory 1205 as a computer storage medium may include an operating system, a network communication module, a user interface module and program instructions.

[0196] Processor 1201 can be used to call a verification application for predictive data stored in memory 1205, and specifically perform the following operations: determine the pricing factors corresponding to the insurance pricing process, and enumerate all factor values ​​corresponding to each of the pricing factors; arrange and combine the factor values ​​based on a combination algorithm to obtain a multidimensional matrix, wherein the multidimensional matrix is ​​used to represent the mapping relationship between pricing scenarios and pricing factor combinations; for any of the pricing scenarios, determine the fourth value range of the pricing factor combination based on the third value range of each factor value in the pricing factor combination; and verify the actual pricing corresponding to each of the pricing scenarios based on the fourth value range.

[0197] As an optional embodiment of the present disclosure, determining the pricing factors corresponding to the insurance pricing process includes: obtaining a logical factor corresponding to the insurance pricing process, and determining the pricing factors based on the logical factor.

[0198] As an optional embodiment of the present disclosure, the enumeration of all factor values ​​corresponding to each of the pricing factors includes: obtaining historical pricing data corresponding to each of the pricing factors; performing feature deduplication on the historical pricing data, and enumerating all factor values ​​corresponding to the pricing factors, wherein pricing factor features of different factor values ​​are different.

[0199] As an optional embodiment of the present disclosure, the method of arranging and combining each of the factor values ​​based on a combination algorithm to obtain a multidimensional matrix includes: traversing each of the pricing factors based on a combination algorithm, arranging and combining each of the pricing factors into pricing factor combinations, and constructing a multidimensional matrix based on the pricing factor combinations, wherein each of the pricing factors has at most one factor value in each of the pricing factor combinations.

[0200] As an optional embodiment of the present disclosure, the determining the fourth value range of the pricing factor combination based on the third value range of each factor value in the pricing factor combination includes: calculating the third value range of the factor value based on the pricing dimension of the factor value, wherein the factor value includes at least one pricing dimension; and determining the fourth value range of the pricing factor combination based on the sum of the third value ranges of each factor value in the pricing factor combination.

[0201] As an optional embodiment of the present disclosure, the third value range is determined based on an upper limit and a lower limit of a hash value of the factor value, and the hash value is determined based on the numerical product of each of the pricing dimensions.

[0202] As an optional embodiment of the present disclosure, the third value range of the factor value is calculated based on the pricing dimension of the factor value, including: calculating the third value range of the factor value under each target date based on the pricing dimension of the factor value, wherein the target date includes daily dates and promotional dates, and there is a target pricing dimension in the factor value, and the target pricing dimension has different values ​​on different target dates.

[0203] As an optional embodiment of the present disclosure, before checking the actual pricing corresponding to each pricing scenario based on each fourth value range, it also includes: selecting each fourth value range that matches the current date.

[0204] As an optional embodiment of the present disclosure, the actual pricing corresponding to each pricing scenario based on each fourth value range includes: for any pricing scenario, obtaining the actual pricing corresponding to the pricing scenario, and comparing the actual pricing with the fourth value range corresponding to the pricing scenario; when the actual pricing is within the fourth value range, generating third verification information, and the third verification information is used to indicate that the actual pricing is normal; when the actual pricing is not within the fourth value range, generating fourth verification information, and the fourth verification information is used to indicate that the actual pricing is abnormal.

[0205] The present disclosure also provides a computer-readable storage medium having a computer program stored thereon, which, when executed by a processor, implements the steps of the above method. The computer-readable storage medium may include, but is not limited to, any type of disk, including a floppy disk, an optical disk, a DVD, a CD-ROM, a microdrive, a magneto-optical disk, a ROM, a RAM, an EPROM, an EEPROM, a DRAM, a VRAM, a flash memory device, a magnetic card or an optical card, a nanosystem (including a molecular memory IC), or any type of medium or device suitable for storing instructions and / or data.

[0206] Next, please refer to FIG13 , which shows a schematic structural diagram of an electronic device provided by an embodiment of the present disclosure.

[0207] As shown in Figure 13, the electronic device 1300 may include: at least one processor 1301, at least one network interface 1304, a user interface 1303, a memory 1305, and at least one communication bus 1302. The communication bus 1302 may be used to implement communication between the various components described above. The user interface 1303 may include buttons, and the optional user interface may also include a standard wired interface or a wireless interface. The network interface 1304 may include, but is not limited to, a Bluetooth module, an NFC module, a Wi-Fi module, etc.

[0208] The processor 1301 may include one or more processing cores. The processor 1301 uses various interfaces and lines to connect the various parts of the entire electronic device 1300, and executes various functions of the electronic device 1300 and processes data by running or executing instructions, programs, code sets or instruction sets stored in the memory 1305, and calling data stored in the memory 1305. Optionally, the processor 1301 can be implemented in at least one hardware form of DSP, FPGA, and PLA. The processor 1301 can integrate one or a combination of CPU, GPU, and modem. Among them, the CPU mainly processes the operating system, user interface, and application programs; the GPU is responsible for rendering and drawing the content to be displayed on the display screen; and the modem is used to handle wireless communications. It is understandable that the above-mentioned modem may not be integrated into the processor 1301, but may be implemented separately through a chip.

[0209] The memory 1305 may include RAM or ROM. Optionally, the memory 1305 includes a non-transitory computer-readable medium. The memory 1305 can be used to store instructions, programs, codes, code sets or instruction sets. The memory 1305 may include a program storage area and a data storage area, wherein the program storage area may store instructions for implementing an operating system, instructions for at least one function (such as a touch function, a sound playback function, an image playback function, etc.), instructions for implementing the above-mentioned various method embodiments, etc.; the data storage area may store data involved in the above-mentioned various method embodiments, etc. The memory 1305 may also be at least one storage device located away from the aforementioned processor 1301. As shown in Figure 13, the memory 1305 as a computer storage medium may include an operating system, a network communication module, a user interface module and program instructions.

[0210] Processor 1301 can be used to call a verification application for predictive data stored in memory 1305, and specifically perform the following operations: determine the pricing factors corresponding to the insurance pricing process, and enumerate all factor values ​​corresponding to each of the pricing factors; arrange and combine the factor values ​​based on a combination algorithm to obtain a multidimensional matrix, wherein the multidimensional matrix is ​​used to represent the mapping relationship between pricing scenarios and pricing factor combinations; for any of the pricing scenarios, determine the second value range of the pricing factor combination based on the first value range of each factor value in the pricing factor combination; and verify the actual pricing corresponding to each of the pricing scenarios based on the second value ranges.

[0211] As an optional embodiment of the present disclosure, determining the pricing factors corresponding to the insurance pricing process includes: obtaining a logical factor corresponding to the insurance pricing process, and determining the pricing factors based on the logical factor.

[0212] As an optional embodiment of the present disclosure, the enumeration of all factor values ​​corresponding to each of the pricing factors includes: obtaining historical pricing data corresponding to each of the pricing factors; performing feature deduplication on the historical pricing data, and enumerating all factor values ​​corresponding to the pricing factors, wherein pricing factor features of different factor values ​​are different.

[0213] As an optional embodiment of the present disclosure, the method of arranging and combining each of the factor values ​​based on a combination algorithm to obtain a multidimensional matrix includes: traversing each of the pricing factors based on a combination algorithm, arranging and combining each of the pricing factors into pricing factor combinations, and constructing a multidimensional matrix based on the pricing factor combinations, wherein each of the pricing factors has at most one factor value in each of the pricing factor combinations.

[0214] As an optional embodiment of the present disclosure, determining the second value range of the pricing factor combination based on the first value range of each factor value in the pricing factor combination includes: calculating the first value range of the factor value based on the pricing dimension of the factor value, the factor value including at least one pricing dimension; and determining the second value range of the pricing factor combination based on the sum of the first value ranges of each factor value in the pricing factor combination.

[0215] As an optional embodiment of the present disclosure, the first value range is determined based on the upper limit and lower limit of the hash value of the factor value, and the hash value is determined based on the numerical product of each of the pricing dimensions.

[0216] As an optional embodiment of the present disclosure, the calculation of the first value range of the factor value based on the pricing dimension of the factor value includes: calculating the first value range of the factor value under each target date based on the pricing dimension of the factor value, wherein the target date includes daily dates and promotional dates, and there is a target pricing dimension in the factor value, and the target pricing dimension has different values ​​on different target dates.

[0217] As an optional embodiment of the present disclosure, before checking the actual pricing corresponding to each pricing scenario based on each second value range, it also includes: selecting each second value range that matches the current date.

[0218] As an optional embodiment of the present disclosure, the actual pricing corresponding to each pricing scenario based on each second value range includes: for any pricing scenario, obtaining the actual pricing corresponding to the pricing scenario, and comparing the actual pricing with the second value range corresponding to the pricing scenario; when the actual pricing is within the second value range, generating first verification information, the first verification information is used to indicate that the actual pricing is normal; when the actual pricing is not within the second value range, generating second verification information, the second verification information is used to indicate that the actual pricing is abnormal.

[0219] The present disclosure also provides a computer-readable storage medium having a computer program stored thereon, which, when executed by a processor, implements the steps of the above method. The computer-readable storage medium may include, but is not limited to, any type of disk, including a floppy disk, an optical disk, a DVD, a CD-ROM, a microdrive, a magneto-optical disk, a ROM, a RAM, an EPROM, an EEPROM, a DRAM, a VRAM, a flash memory device, a magnetic card or an optical card, a nanosystem (including a molecular memory IC), or any type of medium or device suitable for storing instructions and / or data.

[0220] It should be noted that, for the aforementioned method embodiments, for the sake of simplicity of description, they are all expressed as a series of action combinations, but those skilled in the art should be aware that this application is not limited to the order of the actions described, because according to this application, certain steps can be performed in other orders or simultaneously. Secondly, those skilled in the art should also be aware that the embodiments described in the specification are all preferred embodiments, and the actions and modules involved are not necessarily required by this application. In the above embodiments, the description of each embodiment has its own emphasis. For parts that are not described in detail in a certain embodiment, please refer to the relevant description of other embodiments.

[0221] In the several embodiments provided in this application, it should be understood that the disclosed devices can be implemented in other ways. For example, the device embodiments described above are merely schematic. For example, the division of the units is merely a logical function division. In actual implementation, there may be other division methods, such as 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 mutual coupling or direct coupling or communication connection shown or discussed can be through some service interface, and the indirect coupling or communication connection of the device or unit can be electrical or other forms.

[0222] The units described as separate components may or may not be physically separated, and the components displayed as units may or may not be physical units, that is, they may be located in one place, or they may be distributed across multiple network units. Some or all of the units may be selected according to actual needs to achieve the purpose of the scheme of this embodiment. In addition, the functional units in the various embodiments of the present application may be integrated into one processing unit, or each unit may exist physically separately, or two or more units may be integrated into one unit. The above-mentioned integrated units may be implemented in the form of hardware or in the form of software functional units.

[0223] 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 memory. Based on this understanding, the technical solution of the present application is essentially 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, and the computer software product is stored in a memory, including a number of instructions for enabling a computer device (which can be a personal computer, server or network device, etc.) to execute all or part of the steps of the method described in each embodiment of the present application. The aforementioned memory includes: various media that can store program codes, such as a USB flash drive, a read-only memory (ROM), a random access memory (RAM), a mobile hard disk, a magnetic disk or an optical disk.

[0224] Those skilled in the art will appreciate that all or part of the steps in the various methods of the above embodiments can be completed by instructing related hardware through a program, and the program can be stored in a computer-readable memory, which may include: a flash drive, a read-only memory (ROM), a random access memory (RAM), a magnetic disk or an optical disk, etc.

[0225] The foregoing description describes specific embodiments of the present disclosure. Other embodiments are within the scope of the appended claims. In some cases, the actions or steps recited in the claims can be performed in an order different from that described in the embodiments and still achieve the desired results. Furthermore, the processes depicted in the accompanying drawings do not necessarily require the specific order shown or the sequential order to achieve the desired results. In certain embodiments, multitasking and parallel processing are also possible or may be advantageous.

Claims

1. A method for verifying forecast data, comprising: Determine each calculation factor corresponding to the target calculation process, and enumerate all factor values ​​corresponding to each calculation factor; Arrange and combine the factor values ​​based on a combination algorithm to obtain a multidimensional matrix, wherein the multidimensional matrix is ​​used to represent a mapping relationship between the prediction scenario and the calculation factor combination; For any of the prediction scenarios, determining a second value range of the calculation factor combination based on the first value range of each factor value in the calculation factor combination; The scenario prediction value corresponding to each of the prediction scenarios is checked based on each of the second value ranges.

2. The method according to claim 1, wherein determining the calculation factors corresponding to the target calculation process comprises: Obtain a logical factor corresponding to the target calculation process, and determine each calculation factor based on the logical factor.

3. The method according to claim 1, wherein the step of respectively enumerating all factor values ​​corresponding to each of the calculation factors comprises: Respectively obtain historical calculation data corresponding to each of the calculation factors; Deduplication of features is performed on the historical calculation data, and all factor values ​​corresponding to the calculation factors are enumerated to obtain, wherein the calculation factor features of different factor values ​​are different.

4. The method according to claim 1, wherein the permutation and combination of the factor values ​​based on the combination algorithm to obtain a multidimensional matrix comprises: The calculation factors are traversed based on a combination algorithm, the calculation factors are arranged and combined into calculation factor combinations, and a multidimensional matrix is ​​constructed based on the calculation factor combinations, wherein each calculation factor has at most one factor value in each calculation factor combination.

5. The method according to claim 1, wherein determining the second value range of the calculation factor combination based on the first value range of each factor value in the calculation factor combination comprises: Calculating a first value range of the factor value based on a calculation dimension of the factor value, wherein the factor value includes at least one of the calculation dimensions; The second value range of the calculation factor combination is determined based on the sum of the first value ranges of the factor values ​​in the calculation factor combination.

6. The method according to claim 5, wherein the first value range is determined based on an upper limit and a lower limit of a hash value of the factor value, and the hash value is determined based on a product of the numerical values ​​of each of the calculation dimensions.

7. The method according to claim 5, wherein the step of calculating the first value range of the factor value based on the calculation dimension of the factor value comprises: A first value range of the factor value under each target date is calculated based on the calculation dimension of the factor value, wherein the target date includes a standard date and a special date, and the factor value has a target calculation dimension, and the target calculation dimension has different values ​​on different target dates.

8. The method according to claim 7, before checking the scenario prediction value corresponding to each of the prediction scenarios based on each of the second value ranges, further comprising: Select each of the second value ranges that matches the current date.

9. The method according to claim 1, wherein the checking of the scenario prediction value corresponding to each of the prediction scenarios based on each of the second value ranges comprises: For any of the predicted scenarios, obtaining a scenario prediction value corresponding to the predicted scenario, and comparing the scenario prediction value with the second value range corresponding to the predicted scenario; When the scene prediction value is within the second value range, generating first verification information, wherein the first verification information is used to indicate that the scene prediction value is normal; When the scenario prediction value is not within the second value range, second verification information is generated, and the second verification information is used to indicate that the scenario prediction value is abnormal.

10. A method for verifying forecast data, comprising: Determine each pricing factor corresponding to the insurance pricing process, and enumerate all factor values ​​corresponding to each pricing factor; Arrange and combine the factor values ​​based on a combination algorithm to obtain a multidimensional matrix, wherein the multidimensional matrix is ​​used to represent the mapping relationship between the pricing scenario and the pricing factor combination; For any of the pricing scenarios, determining a fourth value range of the pricing factor combination based on the third value range of each factor value in the pricing factor combination; The actual pricing corresponding to each of the pricing scenarios is checked based on each of the fourth value ranges.

11. The method according to claim 10, wherein determining each pricing factor corresponding to the insurance pricing process comprises: Obtain logical factors corresponding to the insurance pricing process, and determine various pricing factors based on the logical factors.

12. The method according to claim 10, wherein the step of respectively enumerating all factor values ​​corresponding to each of the pricing factors comprises: Obtaining historical pricing data corresponding to each of the pricing factors respectively; Deduplication of features is performed on the historical pricing data, and all factor values ​​corresponding to the pricing factors are enumerated, wherein pricing factor features of different factor values ​​are different.

13. The method according to claim 10, wherein the permutation and combination of the factor values ​​based on a combination algorithm to obtain a multidimensional matrix comprises: Based on the combination algorithm, each of the pricing factors is traversed, each of the pricing factors is arranged and combined into pricing factor combinations, and a multidimensional matrix is ​​constructed based on the pricing factor combinations, wherein each of the pricing factors has at most one factor value in each of the pricing factor combinations.

14. The method according to claim 10, wherein determining the fourth value range of the pricing factor combination based on the third value range of each factor value in the pricing factor combination comprises: Calculating a third value range of the factor value based on the pricing dimension of the factor value, wherein the factor value includes at least one of the pricing dimensions; A fourth value range of the pricing factor combination is determined based on the sum of the third value ranges of the factor values ​​in the pricing factor combination.

15. The method according to claim 14, wherein the third value range is determined based on an upper limit and a lower limit of a hash value of the factor value, and the hash value is determined based on a product of numerical values ​​of each of the pricing dimensions.

16. The method according to claim 14, wherein the calculating the third value range of the factor value based on the pricing dimension of the factor value comprises: The third value range of the factor value under each target date is calculated based on the pricing dimension of the factor value, wherein the target date includes daily dates and promotion dates, and the factor value has a target pricing dimension, and the target pricing dimension has different values ​​on different target dates.

17. The method according to claim 16, before verifying the actual price corresponding to each pricing scenario based on each fourth value range, further comprising: Each of the fourth value ranges that matches the current date is selected.

18. The method according to claim 10, wherein the checking of the actual pricing corresponding to each pricing scenario based on each fourth value range comprises: For any of the pricing scenarios, obtaining an actual price corresponding to the pricing scenario, and comparing the actual price with the fourth value range corresponding to the pricing scenario; When the actual price is within the fourth value range, generating third verification information, the third verification information is used to indicate that the actual price is normal; When the actual pricing is not within the fourth value range, fourth verification information is generated, and the fourth verification information is used to indicate that the actual pricing is abnormal.

19. A device for verifying prediction data, comprising: A first determining module is used to determine each calculation factor corresponding to the target calculation process, and enumerate all factor values ​​corresponding to each calculation factor; a first permutation and combination module, configured to permutate and combine the factor values ​​based on a combination algorithm to obtain a multidimensional matrix, wherein the multidimensional matrix is ​​used to represent a mapping relationship between prediction scenarios and calculation factor combinations; A second determining module is configured to determine, for any of the prediction scenarios, a second value range of the calculation factor combination based on the first value range of each factor value in the calculation factor combination; The first checking module is used to check the scene prediction value corresponding to each of the prediction scenes based on each of the second value ranges.

20. The apparatus according to claim 19, wherein the first determining module is specifically configured to: Obtain a logical factor corresponding to the target calculation process, and determine each calculation factor based on the logical factor.

21. The apparatus according to claim 19, wherein the first determining module is further configured to: Respectively obtain historical calculation data corresponding to each of the calculation factors; Deduplication of features is performed on the historical calculation data, and all factor values ​​corresponding to the calculation factors are obtained by enumeration, where: The calculation factor characteristics are different for different factor values.

22. The device according to claim 19, wherein the first permutation and combination module is specifically configured to: The calculation factors are traversed based on a combination algorithm, the calculation factors are arranged and combined into calculation factor combinations, and a multidimensional matrix is ​​constructed based on the calculation factor combinations, wherein: Each of the calculation factors has at most one factor value in each of the calculation factor combinations.

23. The apparatus according to claim 19, wherein the second determining module is specifically configured to: The first value range of the factor value is calculated based on the calculation dimension of the factor value, wherein: The factor value includes at least one of the calculation dimensions; The second value range of the calculation factor combination is determined based on the sum of the first value ranges of the factor values ​​in the calculation factor combination.

24. The apparatus according to claim 23, wherein the first value range is determined based on an upper limit and a lower limit of a hash value of the factor value, and the hash value is determined based on a numerical product of each of the calculation dimensions.

25. The apparatus according to claim 23, wherein the second determining module is further configured to: Based on the calculation dimension of the factor value, the first value range of the factor value under each target date is calculated respectively, wherein: The target date includes a standard date and a special date. The factor value includes a target calculation dimension, and the target calculation dimension has different values ​​for different target dates.

26. The apparatus according to claim 25, wherein the first checking module is specifically configured to: Select each of the second value ranges that matches the current date.

27. The apparatus according to claim 26, wherein the first checking module is further configured to: For any of the predicted scenarios, obtaining a scenario prediction value corresponding to the predicted scenario, and comparing the scenario prediction value with the second value range corresponding to the predicted scenario; When the scene prediction value is within the second value range, generating first verification information, wherein the first verification information is used to indicate that the scene prediction value is normal; When the scenario prediction value is not within the second value range, second verification information is generated, and the second verification information is used to indicate that the scenario prediction value is abnormal.

28. A device for checking prediction data, comprising: The third determining module is used to determine each pricing factor corresponding to the insurance pricing process and enumerate all factor values ​​corresponding to each pricing factor; a second permutation and combination module, configured to permutate and combine the factor values ​​based on a combination algorithm to obtain a multidimensional matrix, wherein the multidimensional matrix is ​​used to represent a mapping relationship between pricing scenarios and pricing factor combinations; a fourth determining module, configured to determine, for any of the pricing scenarios, a fourth value range of the pricing factor combination based on the third value range of each factor value in the pricing factor combination; The second checking module is used to check the actual pricing corresponding to each of the pricing scenarios based on each of the fourth value ranges.

29. The apparatus according to claim 28, wherein the third determining module is specifically configured to: Obtain logical factors corresponding to the insurance pricing process, and determine various pricing factors based on the logical factors.

30. The apparatus according to claim 28, wherein the third determining module is further configured to: Obtaining historical pricing data corresponding to each of the pricing factors respectively; Deduplication of features is performed on the historical pricing data, and all factor values ​​corresponding to the pricing factors are obtained by enumeration, where: The pricing factor characteristics are different for different factor values.

31. The apparatus according to claim 28, wherein the second permutation and combination module is specifically configured to: Based on the combination algorithm, each of the pricing factors is traversed, each of the pricing factors is arranged and combined into each pricing factor combination, and a multidimensional matrix is ​​constructed based on the pricing factor combination, wherein, Each of the pricing factors has at most one factor value in each of the pricing factor combinations.

32. The apparatus according to claim 28, wherein the fourth determining module is specifically configured to: Calculating a third value range of the factor value based on the pricing dimension of the factor value, the factor value including at least one of the pricing dimensions; A fourth value range of the pricing factor combination is determined based on the sum of the third value ranges of the factor values ​​in the pricing factor combination.

33. The apparatus according to claim 32, wherein the third value range is determined based on an upper limit and a lower limit of a hash value of the factor value, and the hash value is determined based on a product of numerical values ​​of each of the pricing dimensions.

34. The apparatus according to claim 32, wherein the fourth determining module is further configured to: Based on the pricing dimension of the factor value, the third value range of the factor value under each target date is calculated respectively, wherein: The target dates include daily dates and promotional dates. The factor values ​​include a target pricing dimension, and the target pricing dimension has different values ​​for different target dates.

35. The apparatus according to claim 34, wherein the second checking module is specifically configured to: Each of the fourth value ranges that matches the current date is selected.

36. The apparatus according to claim 35, wherein the second checking module is further configured to: For any of the pricing scenarios, obtaining an actual price corresponding to the pricing scenario, and comparing the actual price with the fourth value range corresponding to the pricing scenario; When the actual price is within the fourth value range, generating third verification information, the third verification information is used to indicate that the actual price is normal; When the actual pricing is not within the fourth value range, fourth verification information is generated, and the fourth verification information is used to indicate that the actual pricing is abnormal.

37. An electronic device comprising a processor and a memory; The processor is connected to the memory; The memory is used to store executable program code; The processor runs a program corresponding to the executable program code by reading the executable program code stored in the memory, so as to execute the method according to any one of claims 1 to 9.

38. An electronic device comprising a processor and a memory; The processor is connected to the memory; The memory is used to store executable program code; The processor runs a program corresponding to the executable program code by reading the executable program code stored in the memory, so as to execute the method according to any one of claims 10 to 18.

39. A computer-readable storage medium having a computer program stored thereon, wherein the computer-readable storage medium stores instructions, which, when the instructions are executed on a computer or a processor, cause the computer or processor to execute the steps of the method according to any one of claims 1 to 9.

40. A computer-readable storage medium having a computer program stored thereon, wherein the computer-readable storage medium stores instructions, which, when the instructions are executed on a computer or a processor, cause the computer or processor to execute the steps of the method according to any one of claims 10 to 18.

Images