CN111738831A - A business processing method, device and system - Google Patents

Abstract

The embodiment of the specification discloses a service processing method, a device and a system, wherein the method comprises the steps of receiving a current service processing request sent by terminal equipment, wherein the current service processing request comprises object information of a current target object corresponding to service processing; determining an object data set to which the current target object belongs according to the object information; the object data set comprises a data set consisting of calibration values and/or service evaluation values of a plurality of target objects with object attribute characteristics meeting preset requirements in a specified time interval; acquiring a reference value corresponding to the object data set, wherein the reference value is determined according to a calibration value and/or a service evaluation value of at least one target object in the object data set; correcting the reference value by using the correction coefficient of the current target object; and processing the current service processing request by using the service evaluation value of the current target object. With the various embodiments, the processing efficiency of the business processing system may be improved while reducing the financial institution risk of capital loss.

Description

A business processing method, device and system

technical field

This specification relates to the technical field of computer data processing, and in particular, to a business processing method, device, and system.

Background technique

Generally, the processing of risky business such as loans involves the evaluation of resources that can be characterized by collateral, and the efficiency and accuracy of the evaluation of resources that can be characterized by collateral for loan business have a great impact on the processing efficiency of the business processing system. . At present, business processing systems usually evaluate the amount of resources that can be represented by collateral through the historical transaction value of similar collateral. However, for some types of collaterals, due to the low transaction frequency involved, the amount of sample data that can be extracted during the evaluation process is small. Therefore, in order to avoid the high risk of financial institutions losing funds, manual intervention is usually used at present. Carry out the evaluation of the amount of resources that can be represented by the collateral or the adjustment of the evaluation value. With the growth of loan business, the human resource cost required by the above method is relatively high, and the efficiency of data processing of the business system is seriously affected. Therefore, there is an urgent need for a method for further improving the processing efficiency of the business processing system on the basis of reducing the risk of financial institution loss of funds.

SUMMARY OF THE INVENTION

The purpose of the embodiments of the present specification is to provide a business processing method, device and system, which can further improve the processing efficiency of the business processing system, and at the same time reduce the risk of loss of funds of financial institutions.

This specification provides a service processing method, device and system, which are implemented in the following ways:

A service processing method, applied to a server, includes: receiving a current service processing request sent by a terminal device, where the current service processing request includes object information of a current target object corresponding to the service processing. The object data set to which the current target object belongs is determined according to the object information; the object data set includes calibration values and/or business evaluation values of multiple target objects whose attribute characteristics meet preset requirements within a specified time interval. data set. A reference value corresponding to the object data set is acquired, where the reference value is determined according to a calibration value and/or a business evaluation value of at least one target object in the object data set. Correct the reference value by using the correction coefficient of the current target object to obtain the business evaluation value of the current target object; the correction coefficient is based on the difference between the current target object and each target object in the target data set The attribute difference characteristics are determined. The current service processing request is processed by using the service evaluation value of the current target object.

In other embodiments of the method provided in this specification, the reference value corresponding to the object data set is determined in the following manner. Acquiring a first target object including a calibration value within a first specified time interval and a second target object including a business evaluation value within a second specified time interval in the object data set. When it is determined that the number of the first target objects in the object data set is greater than or equal to a first specified threshold, adjust the parameter based on the reference value corresponding to the object data set to the at least one first target object in the object data set. The calibration value is processed to obtain the reference value corresponding to the object data set. When it is determined that the number of the first target objects in the object data set is less than the preset threshold, evaluate the business corresponding to at least one second target object in the object data set based on the reference value adjustment parameter corresponding to the object data set The value is processed to obtain the reference value corresponding to the object data set. Wherein, the reference value adjustment parameter is determined according to the difference characteristic between the calibration value of at least one target object in the object data set or the business evaluation value and the transaction value.

In other embodiments of the method provided in this specification, the method further includes: acquiring the calibration value corresponding to the first target object or the first time corresponding to the generation of the service evaluation value corresponding to the second target object, and the second time corresponding to the current reference value determination. The correction index corresponding to the corresponding first target object or the second target object is determined according to the difference characteristic of the transaction value of the target object in the object data set at the first time and the second time. The calibration value of the corresponding first target object or the business evaluation value of the second target object is corrected by using the correction index to obtain the corrected calibration value of the corresponding first target object or the corrected business evaluation value of the second target object. The reference value corresponding to the object data set is determined by using the revised calibration value of the first target object or the revised service evaluation value of the second target object.

In other embodiments of the method provided in this specification, the method further includes: determining the object according to the transaction volume of the target object in the object data set in a third specified time interval and the difference between the reference value and the transaction value The confidence level of the benchmark value corresponding to the dataset. The data on which the benchmark price calculation of the object data set depends and the benchmark price calculation method are adjusted according to the benchmark value confidence.

In other embodiments of the method provided in this specification, the method further includes: starting from the second time and taking the update cycle as a unit, sequentially iteratively extracting the target object data samples, when the set identifier corresponding to the object data set is Stop data extraction when the number of target object data samples contained in the corresponding extracted target object data samples is greater than the second specified threshold, or stop data extraction when the number of update cycles reaches a third specified threshold. The target object data sample includes data corresponding to a first target object or a second target object. All the extracted target object data samples are stored in the corresponding object data set.

In other embodiments of the method provided in this specification, when the target object data sample includes data corresponding to a first target object, and when the target object data set includes a number of target object data samples greater than the second specified When the threshold value is reached, the method further includes: using the numerical amplitude to adjust the maximum value of the calibration value and the minimum value of the calibration value in all the extracted target object data samples; the value of the numerical amplitude is based on the numerical amplitude of the previous reference value update cycle and The ratio of the maximum value to the minimum value of all the extracted target object data samples is determined. Filter the target object data samples that are not within the value interval formed by the adjusted maximum value of the calibration value and the minimum value of the calibration value. When the number of target object data samples remaining after filtering is greater than the second specified threshold, the remaining target object data samples after filtering are stored in the object data set. Otherwise, continue to iteratively extract the target object data samples, and repeat the above processing steps until the number of remaining target object data samples after filtering is greater than the second specified threshold or the number of update cycles reaches the third specified threshold, stop the iteration, and use the last iteration The remaining target object data samples after filtering are stored in the object dataset.

On the other hand, an embodiment of this specification also provides a service processing apparatus, which is applied to a server, and includes: a request receiving module, configured to receive a current service processing request sent by a terminal device, where the current service processing request includes a request corresponding to the service processing. Object information of the current target object. A data set determination module, configured to determine the object data set to which the current target object belongs according to the object information; the object data set includes calibration values of multiple target objects whose object attribute characteristics meet preset requirements within a specified time interval and/or business evaluations. A data acquisition module, configured to acquire a reference value corresponding to the object data set, where the reference value is determined according to a calibration value and/or a business evaluation value of at least one target object in the object data set. A correction processing module, configured to perform correction processing on the reference value by using the correction coefficient of the current target object to obtain a business evaluation value of the current target object; the correction coefficient is based on the current target object and the object data The attribute difference feature between each target object in the set is determined. A service processing module, configured to process the current service processing request by using the service evaluation value of the current target object.

In other embodiments of the apparatus provided in this specification, the apparatus further includes a reference value determination module, and the reference value determination module includes: a data acquisition unit, configured to acquire the object data concentrated in a first specified time interval The first target object containing the calibration value and the second target object containing the business evaluation value in the second specified time interval. A first reference value determination unit, configured to adjust parameters based on the reference value corresponding to the object data set to the object data set when it is determined that the number of first target objects in the object data set is greater than or equal to a first specified threshold. The calibration value corresponding to at least one first target object is processed to obtain the reference value corresponding to the object data set. A second reference value determination unit, configured to adjust parameters for at least one object in the object dataset based on the reference value corresponding to the object dataset when it is determined that the number of first target objects in the object dataset is less than a preset threshold The business evaluation value corresponding to the second target object is processed to obtain the reference value corresponding to the object data set. Wherein, the reference value adjustment parameter is determined according to the difference characteristic between the calibration value of at least one target object in the object data set or the business evaluation value and the transaction value.

In other embodiments of the apparatus provided in this specification, the apparatus further includes an adjustment module, and the adjustment module includes: a confidence level determination unit, configured to The transaction volume and the difference between the reference value and the transaction value determine the confidence level of the reference value corresponding to the object data set. An adjustment unit, configured to adjust the data and the calculation method on which the benchmark price calculation of the object data set depends, according to the confidence level of the benchmark value.

On the other hand, an embodiment of the present specification further provides a service processing system, the system includes at least one processor and a memory for storing instructions executable by the processor, and when the instructions are executed by the processor, any one of the above-mentioned instructions is implemented or steps of the methods described in the embodiments.

The business processing method, device and system provided by one or more embodiments of this specification use the target object with certain similar characteristics to display the calibration data in a certain historical time interval and/or the business system itself evaluates the business of the target object The data is used as a sample case to evaluate the amount of resources represented by the current target object at the current time, which can further increase the number of sample cases on which the evaluation depends, thereby improving the accuracy of the evaluation. At the same time, it can further improve the processing efficiency of the business processing system and reduce the risk of financial institutions losing funds.

Description of drawings

In order to more clearly illustrate the technical solutions in the embodiments of the present specification or the prior art, the following briefly introduces the accompanying drawings required in the description of the embodiments or the prior art. Obviously, the accompanying drawings in the following description are only These are some embodiments described in this specification. For those of ordinary skill in the art, other drawings can also be obtained according to these drawings without creative labor. In the attached image:

FIG. 1 is a schematic flowchart of an embodiment of a service processing method provided in this specification;

2 is a schematic diagram of a module structure of a service processing device provided in this specification;

FIG. 3 is a schematic structural diagram of a server according to an exemplary embodiment of the present specification.

Detailed ways

In order to make those skilled in the art better understand the technical solutions in this specification, the technical solutions in one or more embodiments of this specification will be clearly and completely described below with reference to the accompanying drawings in one or more embodiments of this specification. It is obvious that the described embodiments are only a part of the embodiments of the specification, but not all of the embodiments. Based on one or more embodiments in the description, all other embodiments obtained by persons of ordinary skill in the art without creative work shall fall within the protection scope of the solutions of the embodiments of this description.

In an example of an application scenario of this specification, a user can initiate a service processing request through a terminal device and send it to the server of the service processing system. After receiving the service processing request, the server of the service processing system may acquire the object information of the target object included in the service processing request. Then, the object dataset corresponding to the target object can be determined according to the object information. The object data set is composed of calibration values or business evaluation values of multiple target objects whose attribute characteristics meet preset requirements. The server may preliminarily determine the reference value corresponding to the object data set according to the calibration value or service evaluation value of the multiple target objects, so as to preliminarily determine the amount of resources that can be represented by the multiple target objects. Then, the reference value is corrected according to the difference characteristics between the multiple target objects, and the resource amount that can be represented by any one of the target objects is obtained, and the evaluation value of the target object is obtained. Then, the server can perform subsequent business processing based on the evaluation value. For example, the server can compare the evaluation value with the amount of resources applied for by the loan business, and if the requirements are met, the loan business can continue to be processed; if the requirements are not met, the corresponding loan business will be rejected or transferred to manual processing. Thereby, the rationality of business processing is improved, the risk of capital loss of financial institutions is reduced, and the user experience is improved.

FIG. 1 is a schematic flowchart of an embodiment of the service processing method provided in this specification. Although the present specification provides method operation steps or device structures as shown in the following embodiments or accompanying drawings, the method or device may include more or less operation steps after partial combination based on routine or without creative work. or modular units. In the steps or structures that logically do not have a necessary causal relationship, the execution sequence of these steps or the module structure of the device are not limited to the execution sequence or module structure shown in the embodiments of the present specification or the accompanying drawings. When the described method or module structure is applied to an actual device, server or terminal product, it can be executed sequentially or in parallel (for example, parallel processor or multi-thread processing) according to the method or module structure shown in the embodiments or the accompanying drawings. environment, and even the implementation environment of distributed processing and server clusters).

A specific embodiment is shown in FIG. 1. In an embodiment of the service processing method provided in this specification, the method can be applied to a server that configures a collection policy, and the method can include the following steps:

S20: Receive a service processing request sent by the terminal device, where the service processing request includes object information of a current target object corresponding to the service processing.

The server can receive the service processing request sent by the terminal device. The service processing request may include object information of the current target object corresponding to the service processing. The target object in the business process can be a direct object of the business process, an indirect object of the business process, or a subordinate object of the direct object, and so on. For example, the business process is a loan application business process, and the target object in the business process can be the collateral set during the loan application, such as a house and a car.

A user can initiate a service processing request through a terminal device. For example, the user can input information in the service processing application program of the terminal device or the service processing application interface displayed by the browser, and initiate a service processing request. The terminal device can collect the information input by the user through the application program or the application interface, and generate a service processing request based on the collected information. Then, the terminal device can send the generated service processing request to the server. When a user initiates a business processing request through a terminal device, he or she can input the object information of the target object at the same time as inputting the application information in the application program or application interface corresponding to the business processing, or upload a certification image containing the object information of the target object, Such as real estate deed images. The terminal device may generate a service processing request after the user completes the input of the application information corresponding to the service processing. Correspondingly, the service processing request includes the object information of the target object. If it is a proof image, the server can obtain the object information of the target object through image recognition.

The object information may include identification information of the target object. The identification information may include, for example, the registration number of the target object, etc. For example, for a house, the object identification may be city number + real estate certificate number. If the target object does not have a registration number, the object identifier may also be a number randomly generated by a terminal device or a server, or a number generated based on user information or business processing information, etc., to identify the target object. For example, the object information may further include other attribute information of the target object, such as corresponding user information, address information, and the like. For example, for a house, the object information may further include homeowner information, information about the administrative area to which it belongs, and information about a residential area.

The current target object is the target object pointed to by the object information contained in the current service processing request received by the server, so as to facilitate distinguishing and expression.

S22: Determine the object data set to which the current target object belongs according to the object information; the object data set includes calibration values and/or service evaluation values corresponding to multiple target objects whose object attribute characteristics meet preset requirements. data set.

The server may determine the object dataset to which the current target object belongs according to the object information. The object data set may include a data set composed of calibration values and/or service evaluation values corresponding to multiple target objects whose attribute characteristics of the objects meet preset requirements. The object attribute feature may be, for example, address information, user information and the like of the object. For a house, the object attribute feature may also include the house orientation, area, number of floors where it is located, and so on. Based on the application scenario, data corresponding to multiple target objects whose attribute characteristics meet certain preset requirements can be stored in a data set to form an object data set. For example, the object data set may be a data set composed of data of multiple target objects with the same regional attribute characteristics. The preset requirements may be set according to actual application scenarios. The object data set may be identified using a set identifier.

The server may determine the object dataset to which the current target object belongs according to the object information. The server may pre-store the association relationship between each object identifier and the set identifier of the corresponding object data set, and the server may determine the object data set corresponding to the target object according to the association relationship.

Alternatively, the server may also extract the object attribute feature on which the object data set division depends from the object information, and then may determine the object data set corresponding to the target object according to the object attribute feature. For example, the object dataset is a collection of data corresponding to each house in the same residential area located in the same administrative area, that is, the object dataset is divided according to the administrative area and residential area information of the target object. The real estate certificate image collected by the terminal device may include the residential area information and the administrative area information to which the house belongs. The server can obtain the residential area information and administrative area information to which the current target object belongs by performing text recognition on the real estate certificate image, such as OCR recognition, and then can determine the object data set to which the current target object belongs according to the residential area information and administrative area information.

The data contained in the object data set may include transaction values, calibration values, and business evaluation values of the target objects in the historical transactions of each target object, and the like during historical business processing. The calibration value is the quantity value of the target object displayed in a certain time point or time interval in a certain way, so that both parties can trade the target object based on the displayed calibration value. The service evaluation value may be the evaluation quantity value of the target object in the historical service processing process. For example, for a house, the calibration value may be the listed price, and the business evaluation value may be the evaluation price of the house in the historical business process.

S24: Acquire a reference value corresponding to the object data set, where the reference value is determined according to a calibration value and/or a service evaluation value of at least one target object in the object data set.

The server can obtain the reference value corresponding to the object data set. The reference value may be determined according to a calibration value and/or a business evaluation value of at least one target object in the object data set. For example, the average value of the calibration values of at least one target object may be used as the preliminary reference value, and then, the preliminary reference value may be corrected by using the business evaluation value of at least one target object to obtain the reference value corresponding to the object data set .

In other embodiments, the reference value corresponding to the object data set may also be determined in the following manner: acquiring the first target object in the object data set that includes the calibration value in the first specified time interval and the first target object in the second specified time interval A second target object containing business evaluation values. When it is determined that the number of the first target objects in the object data set is greater than or equal to a first specified threshold, adjust the parameter based on the reference value corresponding to the object data set to the at least one first target object in the object data set. The calibration value is processed to obtain the reference value corresponding to the object data set. When it is determined that the number of the first target objects in the object data set is less than the preset threshold, evaluate the business corresponding to at least one second target object in the object data set based on the reference value adjustment parameter corresponding to the object data set The value is processed to obtain the reference value corresponding to the object data set.

Wherein, the reference value adjustment parameter is determined according to the difference characteristic between the calibration value of at least one target object in the object data set or the business evaluation value and the transaction value. The lengths of the first specified time interval and the second specified time interval can be set according to the actual application scenario, so as to ensure the timeliness of the calibration value and at the same time increase the number of target objects that contain the calibration value in the object data set as much as possible. . The first specified threshold may be determined according to an actual application scenario.

In practical application scenarios, there are usually few target objects with business evaluation values. When the number of target objects with calibration values meets the preset requirements, the target objects with calibration values can be used as sample cases to determine the benchmark value. In order to ensure the number of sample cases used and improve the accuracy of the determination of the reference value corresponding to the object data set. Usually, the accuracy of the business evaluation value is higher than the calibration value. When the number of target objects with the calibration value is very small, the target object with the business evaluation value can be used as a sample case to determine the benchmark value to further ensure the object data. The accuracy of the determination of the reference value corresponding to the set.

When a sample case for determining the reference value is selected from the aggregated data set according to the above judgment conditions, the calibration value data or business evaluation value data of the sample case can be further adjusted by using the reference value adjustment parameter. The reference value adjustment parameter can be used to characterize the difference between the calibration value of the target object or the business evaluation value and the transaction value. The difference between the evaluation value corresponding to the target object in the current evaluation process and the possible transaction value corresponding to the actual transaction is further improved, so as to ensure the accuracy of the value evaluation of the target object.

In other embodiments, if neither the first target object nor the second target object exists in the object data set, the correction index may be used to correct the historical reference value of the object data set, as the object data set for this evaluation. the corresponding benchmark value. If there is no historical reference value in the object data set, the reference value of the object data set of the current target object can be estimated by using the reference value of the object set with similar characteristics to the object data set of the current target object. Alternatively, an algorithm such as random forest may be further utilized to determine the reference value of the object dataset of the current target object.

In other embodiments, the first time corresponding to the generation of the calibration value corresponding to the first target object or the business evaluation value corresponding to the second target object, and the second time corresponding to the determination of the current reference value may also be obtained. The correction index corresponding to the corresponding first target object or the second target object is determined according to the difference characteristic of the transaction value of the target object in the object data set at the first time and the second time. The calibration value of the corresponding first target object or the business evaluation value of the second target object is corrected by using the correction index to obtain the corrected calibration value of the corresponding first target object or the corrected business evaluation value of the second target object. The reference value corresponding to the object data set is determined by using the revised calibration value of the first target object or the revised service evaluation value of the second target object.

By correcting the calibration value or business evaluation value based on the transaction value difference characteristics of at least one target object in the object dataset over time to correct to the current time point, the historical sample data used can further reduce the impact of time migration on the target object. Quantitative values assess the effect of accuracy.

S26: Correct the reference value by using the correction coefficient of the current target object to obtain a business evaluation value of the current target object; the correction coefficient is based on the current target object and each target object in the target data set The attribute difference between the characteristics is determined.

The server may perform correction processing on the reference value by using the correction coefficient of the current target object to obtain a business evaluation value of the current target object. The correction coefficient may be determined according to the attribute difference feature between the current target object and each target object in the object data set. The attribute difference feature may include a difference feature of the object attribute corresponding to the current target object relative to the object attributes of other target objects in the object data set.

In some embodiments, the correction coefficient may be determined by using a pre-built correction coefficient prediction model, for example. Feature extraction can be performed on attribute data of a large number of target objects to determine the feature type. Then, feature extraction can be performed using the determined feature types to obtain attribute feature vectors of each target object. Then, an unsupervised learning algorithm, such as a cluster analysis algorithm, can be used to cluster attribute feature vectors of multiple target objects to obtain multiple cluster categories with similar attribute features. Then, the cluster categories to which each target object belongs in the object dataset can be analyzed, and the attribute features corresponding to the object dataset can be quantitatively determined based on the number of target objects corresponding to each cluster category and the cluster center value of each cluster category. central value.

Correspondingly, for the current target object, feature extraction can be performed on its attribute data by using a predetermined feature type, and then the extracted feature vector can be used to perform similarity analysis with each clustering category to determine the clustering category to which it belongs. Then, the difference value between the cluster center value of the cluster category to which it belongs and the attribute feature center value of the corresponding object data set can be used as the correction coefficient.

The target objects in the same object dataset only have certain similarity in one or more object data features. By further considering the differences of various attribute characteristics of the target object and modifying the reference value corresponding to the object data set, the evaluation value of the target object can be more in line with the actual attribute characteristics of the current target object, and the accuracy of determining the evaluation value can be improved. .

S28: Process the current service processing request by using the service evaluation value of the current target object.

After obtaining the service evaluation value of the current target object, the server may further process the current service processing request based on the service evaluation value of the current target object. For example, for loan business processing, after determining the business evaluation value of the current target object, the server can use this as a benchmark to determine whether the loan amount applied for by the current loan business process meets the requirements. If the evaluation value is determined, the server can feed back the rejection of the current loan business processing, and feed back the processing result to the terminal device. Or further refer to whether there are other guarantee conditions in the current loan business processing to determine whether the current loan business can continue to be processed. Alternatively, the business evaluation value can also be fed back to the terminal device of the business personnel, and displayed, so that the business personnel can determine whether to continue the current loan business processing.

In practical application scenarios, the transaction frequency of certain types of target objects may be very low or the specific transaction data may be difficult to extract, so that the evaluation of the quantity value of such target objects based on actual transaction data relies on less sample data. Or it is seriously missing, which seriously affects the accuracy of the quantitative value evaluation of the target object. In the embodiment of this specification, the evaluation of the quantity value of the target object at the current time point is carried out by taking the calibration data displayed by the target object to the outside within a certain historical time interval and/or the business evaluation data of the target object by the business system itself as a sample case, It can further increase the number of sample cases that the evaluation relies on, thereby improving the accuracy of the evaluation, and at the same time, it can also reduce the risk of capital loss that may be caused by inaccurate evaluation of the target object in business processing.

In other embodiments, the data in each object data set may be extracted in the following manner: starting from the second time and taking the update cycle as a unit, sequentially extracting the target object data samples iteratively. Stop data extraction when the number of target object data samples extracted corresponding to the set identifier corresponding to the object data set contains target object data samples greater than the second specified threshold, or stop data extraction when the number of update cycles reaches a third specified threshold. The target object data sample includes data corresponding to a first target object or a second target object. All the extracted target object data samples are stored in the corresponding object data set.

Iteratively extracting the target object data through the above method can further improve the extracted target object data to be closer to the current estimated time on the basis of ensuring the number of samples to be extracted, and reduce the difference of the target object data due to time changes. Improve the accuracy of estimates.

In other embodiments, when the target object data sample includes data corresponding to a first target object, and when the number of target object data samples included in the object data set is greater than a second specified threshold, the numerical amplitude may also be used. Adjust the maximum calibration value and the minimum calibration value in all the extracted target object data samples. Filter the target object data samples that are not within the value interval formed by the adjusted maximum value of the calibration value and the minimum value of the calibration value. When the number of target object data samples remaining after filtering is greater than the second specified threshold, the remaining target object data samples after filtering are stored in the object data set. Otherwise, continue to iteratively extract the target object data samples, and repeat the above processing steps until the number of remaining target object data samples after filtering is greater than the second specified threshold or the number of update cycles reaches the third specified threshold, stop the iteration, and use the last iteration The remaining target object data samples after filtering are stored in the object dataset. By further setting the range of the extracted target object data, the influence of abnormal data on the evaluation results can be further cleaned up, and the accuracy of the estimation can be further improved.

In other embodiments, the confidence level of the reference value corresponding to the object data set may also be determined according to the transaction volume of the target object in the object data set and the difference between the reference value and the transaction value in the third specified time interval. The data on which the benchmark price calculation of the object data set depends and the calculation method are adjusted according to the confidence level of the benchmark value. By marking the calculation result of the benchmark value and the data and calculation method it depends on, the accuracy of the calculation result is verified, and the data and calculation method that the calculation process relies on are adjusted according to the accuracy of the calculation result, so as to improve the target object. The accuracy of the quantitative value assessment.

Based on the solutions provided by the foregoing embodiments, in a scenario example provided in this specification, the foregoing solutions can be used to estimate the price of the mortgage-house in the loan business processing.

The user can initiate a loan business processing request through the terminal device, and the loan business processing request includes the related information of the mortgage-house. The terminal device can send the loan business processing request to the server. The server can obtain information such as the property certificate number of the house, the information of the community where the house is located, and the information of the administrative area through the related information of the mortgage-house included in the loan business processing request. Then, the server can determine the corresponding set identifier by using the information of the cell where the house is located and the information of the administrative area. Further, the reference value corresponding to the set identifier can be obtained. The server may also determine the correction coefficient according to the real estate certificate number. The server may perform correction processing on the reference value by using the obtained correction coefficient to obtain the estimated value of the mortgage-house. The server may perform subsequent processing on the loan business processing request based on the estimated value.

The reference value is pre-determined by the server in the following manner. The update period of the reference value may be performed based on the actual application scenario, for example, it may be updated once a week or a month. The benchmark price mentioned in the following description corresponds to the benchmark value, the estimated price corresponds to the estimated value, the listed price corresponds to the calibration value, the listed case refers to the house with the listed price, and the evaluation case refers to the house with the appraisal price.

The main data sources used in the calculation of the benchmark price are as follows: 1. City ID data: This data is relatively stable and can be updated monthly; 2. Community ID data: Community data is added or changed every day, so the data should be updated regularly , to ensure that the latest data is updated before monthly calculation; 3. Community address id data: the data should be updated regularly, and the latest data should be updated before monthly calculation; 4. Listed case data: at least ensure that the listed case data for the past six months, in every Guaranteed to update the latest data before monthly calculation; 5. Evaluation case: Guaranteed to update the latest data before monthly calculation at least for the past three months; 6. Base price calculation model and base price adjustment parameters: update data monthly, Guaranteed to update the latest data before monthly calculation; 7. Aggregate revision index: update data monthly, and ensure that the index data of the current month is available before monthly calculation.

After the data is obtained, data cleaning can be performed first.

Case data cleaning mainly simplifies cases from different sources and converts them into standard data. The cleaning process first performs single-channel deduplication, and then multi-channel aggregation deduplication can be performed to improve cleaning efficiency, reduce case redundancy, and subdivide the reasons for deduplication for subsequent analysis and optimization. The single-channel/multi-channel cleaning process is the same except whether the source channel field is added during the cleaning process, including the following steps:

First synchronize the data to the temporary table in the middle of the database, according to all fields in the temporary table (cell ID, unit price, total price, current floor, area, number of bedrooms, number of living rooms, number of bathrooms, floor type, apartment type, orientation, decoration, total floor) , month, source) to remove duplicates, use MD5 to process data, and compare data with the same time in the database.

The case is linked to the real estate listing area as follows. Attach in the following order until the attachment is successful:

1. According to the name of the main cell in the cell table (same administrative area);

2. Attach according to the community alias in the community alias table (same administrative area);

3. The administrative district + community name in the case is linked according to the manual accumulation table;

4. Attach according to the name of the main cell in the cell table (no administrative area);

5. Attach according to the community alias in the community alias table (no administrative area);

6. Connect according to the community red list provided by the online department;

7. Attach the upper-level name field (same administrative region) according to the upper-level name in the community alias table.

The online red list refers to the comparison table of community relations provided by the online capture department. The manual accumulation rule means that there is no linked data, enter the ledger system, and manually link the community once a month to form the accumulated community data table; the listed data will be automatically executed once a month, the last 3 months The data are linked according to manual accumulation. The transaction data will be automatically executed once a month, and the data of the last 6 months will be linked based on manual accumulation.

Then, the abnormal data can be cleaned based on the following Table 1. Table 1 shows the abnormal data cleaning rule table.

Table 1

Clean the data according to the above table, and supplement the missing values of the fields accordingly, such as:

If the unit price is empty, and the total price and building area are not empty, the total price/building area should be reserved to 2 decimal places;

If the total price is empty, and the unit price and building area are not empty, the unit price * building area, with 2 decimal places;

The unit type is empty, the building area is >150, with more than four bedrooms; the building area >=120, three bedrooms and two halls; the building area >=90, three bedrooms and one hall; the building area>=55, two bedrooms and one hall; , one room and one hall; otherwise, open room;

Secondly, the corresponding field code value conversion can also be performed according to the attribute code value table, and the key fields of the transaction (community ID, total price, current floor, area, total floor, month) can be deduplicated, and the listed key fields (community ID, unit price, total price, area, total floor, month) deduplication, etc. Finally, the data of the successfully connected cases can be entered into the production table of the model in quasi-real time, and the unconnected cases can be processed manually (once a month).

Then, the listing price and the appraisal price can be calculated using the listing case and the appraisal case.

The calculation results of listed cases include the number of listed cases in the community, the minimum value of listed cases in the community, the median of listed cases in the community, the maximum value of listed cases in the community, and the number and minimum value of the low-rise, multi-storey, high-rise, and super-high-rise cases after the community screening. , median, and maximum. Evaluation cases are evaluation case data accumulated based on evaluations.

Listed case statistics are as follows:

First of all, the cases in which the listed price of each district and each month in the calculated city is 1.5 times the standard deviation of the mean value of the cases are excluded.

The data fields read by the case table are shown in Table 2 below. Table 2 shows the comparison table of field names and text descriptions of listed cases.

Table 2

field name Rename definition communityid community_id cell id sqmprice price Evaluate case prices totalfloor total_floor total floor Listingid listing_id case id online_source online_source Case source timeoflisting price_date Case listing month timeoflisting listing_days The number of days from the listing date to 0001-01-01

After screening by 1.5 times the standard deviation, if the number of listed cases in some communities is less than 24, the following iterative supplementary steps can be performed: iteratively increase the number of cases in a monthly unit, and stop when the number of cases reaches more than 24 data in a certain month Iterate, return all cases from this month to the current month. However, if it still fails to meet the minimum quantity requirement in the past 6 months, it will be stopped. Among them, in the case of making up the adjustment, the listed price can be adjusted to the current time point using the correction index. The correction index includes the data calculated based on the data of each house in the community and the data calculated based on the data of each house in the city. The index obtained based on the cell is preferentially used for adjustment. If the index obtained based on the cell does not exist in a certain cell in a certain city, the index obtained based on the city will be used instead for adjustment. Revised listing price = listing price * revised index.

Calculate the minimum value (Min), median (Median), and maximum value (Max) of the listed price of each district, and adjust the listing according to the price amplitude (Ampli (price amplitude) = maximum value of listed price / minimum value of listed price) The maximum and minimum prices. The specific process is as follows:

1. Amplitude determination: If the price amplitude of the previous cycle is less than 1.5, the price amplitude of this cycle will be corrected to 1.5; if the price amplitude of the previous cycle is >2, the price amplitude of this cycle will be corrected to 2; otherwise, the amplitude of this cycle will be equal to the amplitude of the previous cycle.

2. Data cleaning: Adjust the minimum listing price according to the price amplitude: MinAdj=Median-(Median-Min)*Ampli*Min/Max, adjust the maximum listing case: MaxAdj=Median-(Median-Max)*Ampli*Min/ Max. And delete the case corresponding to the listed price outside the interval [MinAdj, MaxAdj].

3. If the number of listed cases in some communities is less than 24, you can perform the above iterative replenishment steps for iterative replenishment. Then, the minimum listed price is adjusted based on the price amplitude, and the data is cleaned.

Since the calculation of the current month's benchmark price will involve the use of the previous month's data, if there is a cycle that has not been evaluated during the retrospective process, that is, the base cycle. The data of the base period can be determined according to the following methods: Calculate the base amplitude by using the maximum and minimum values of the listed prices in the base period. If the base amplitude is less than 1.5, the base price amplitude is corrected to 1.5. If the base price amplitude is greater than 2, then The base price amplitude was revised to 2. Then, use the base amplitude to correct the maximum and minimum base listing prices, and filter the listing cases.

After the processing based on the above steps is completed, the minimum value (Min), the median (Median), the maximum value (Max) and the number of cases (Count) of the listed price of the cell can be further counted. If Count<6, the price of this cycle (Min, Max, Median) = the price of the previous cycle (Min, Max, Median) * the set correction index. After the above processing, high-rise cases (total floors ∈ (7, 33]), multi-storey cases (total floors ∈ (3, 7]), super high-rise cases (total floors ∈ (33, + ∞)), low-rise cases (total floors ∈ (0,3]), and count the corresponding Min, Max, and Median respectively. The statistical results of the following cases are formed as shown in Table 3. Table 3 indicates that the field names of listed cases correspond to the corresponding text descriptions surface.

table 3

field name meaning city_code city code community_id cell id count Number of listed cases screened count_low Number of low-level listing cases screened count_medium Number of Multi-Layer Listing Cases Screened count_high Number of high-level listing cases screened count_sup Number of super high-rise listing cases screened min Listed case minimum min_low Low-level listing case minimum min_medium Minimum value for multi-level listing cases min_high High-rise listing case minimum min_sup Super high-rise listing case minimum median Median listing cases median_low Median low-level listing cases median_medium Median Multi-Layer Listing Cases median_high Median high-rise listing cases median_sup Median Supertall Listing Cases max Listed case maximum max_low Low-level listing case maximum max_medium Maximum Multi-Layer Listing Cases max_high High-rise listing case maximum max_sup Super high-rise listing case maximum mean Average listing case price_date Calculate the month create_date Date created

Evaluate case statistics

Read the data from the assessment case table (assess_case) for the past three months. The read fields are: community_id, sqmprice, contractdate, price_date, city_code, etc. Backtracking the evaluation cases, the specific process is as follows: take the 1st of each month as the division point of different months, if there is no evaluation case in the current month, iteratively increase the case by month until the number of cases > 0, and finally stop after 3 months ; Revised appraisal price = Appraisal price * Revised index. The statistical results of the following cases are formed as shown in Table 4 below. Table 4 shows the correspondence table between the field names of the evaluation cases and the corresponding text descriptions.

Table 4

field name Rename definition community_id community_id cell id sqmprice price Evaluate case prices contractdate contract_date Assessment Case Date contractdate price_date Evaluation Case Month city_code city_code city id

At the same time, the median evaluation price of each evaluation case in the community can also be calculated and stored in the database.

Then, the calculation of the base price can be performed.

The cell id is associated with the cell listing case statistical data set and cell evaluation case statistical data set processed in the above steps, as the object data set.

Then, the number of listed cases in the object set can be determined. If the number meets certain conditions, the benchmark value adjustment parameters can be used to process the listed price data in the object set. For example, the median of the listed price data can be taken as the benchmark value. The product of the adjustment parameter and the median of the listed price data is used as the benchmark price of the community.

If the number of listed cases is small, the benchmark value adjustment parameters can be used to process the appraisal price data in the object set. The product is used as the base price of the community.

If a set of objects has neither a listing case nor an evaluation case, the benchmark price can be determined by using the data of the neighborhoods with similar characteristics. Or the random forest algorithm initially determines the benchmark price.

In some embodiments, the calculation model type configuration may also be performed in advance in combination with the data features of the object set. The field corresponding to the corresponding calculation model type is configured as standard_price_type.

For example, a set of objects whose number of listed cases meets certain conditions can be configured as parameter adjustment. If the standard_price_type field of the object collection is read as parameter adjustment, the benchmark price is calculated in the following way:

When the benchmark price parameter standard_price_parem is less than or equal to 0.5,

price=min+(median-min)*standard_price_parem*2;

When the benchmark price parameter standard_price_parem is greater than 0.5,

price=max+(median-max)*(1-standard_price_parem)*2.

If the number of listed cases is low, it can be further broken down:

If the value of standard_price_type is manual, the current community benchmark price in the previous cycle is corrected by the set correction index as the current community benchmark price.

If the value of standard_price_type is to be determined or empty. If it is judged that there is an evaluation case in the community, the benchmark price will use the median calculated by the evaluation case; otherwise, the benchmark price will be calculated according to the parameter adjustment.

If standard_price_type takes the value of the primary and secondary cells, the benchmark price of similar cells multiplied by the coefficient is used as the benchmark price of the cell.

If the benchmark price cannot be calculated by the above methods, random forest can be used for price prediction. Alternatively, the average price of the administrative region and city where the community is located can be further used to fill the gap, and the average price of the administrative region can be used first. The benchmark price calculated by random forest, administrative area average price, and city average price is corrected by random numbers between -3% and 3% to avoid a lot of repetition of results.

In other embodiments, the calculation result and calculation method can also be marked:

Calculation result marking:

(1) If sale_count (the volume of transactions in 12 months)>=3 and ppe15>=0.8, and in the past three months (only for the month with more than or equal to 3 transactions), the single-month transaction is greater than or equal to 3 and nearly 3 If the PPE15 of the mid-month and single month is not lower than 75%, the confidence level is high; the calculation result of the current benchmark price can be marked as high confidence level.

(2) If sale_count (the volume of transactions in 12 months)>=3 and ppe15>=0.8, and in the past three months (only for the month with more than or equal to 3 transactions), the single-month transaction is greater than or equal to 3 and nearly 3 Mid-month and single-month PPE15 is lower than 75% confidence level is medium-high; the current benchmark price calculation result can be marked as medium-high confidence level.

(3) If sale_coun (12-month transaction volume)<3 and ppe15>=0.8, the confidence level is medium; the calculation result of the current benchmark price can be marked as medium confidence.

(4) If ppe15>=0.6 and ppe15<0.8, the confidence level is medium; the calculation result of the current benchmark price can be marked as medium confidence.

(5) If ppe15<0.6, the confidence is low; the calculation result of the current benchmark price can be marked as low confidence.

(6) If ppe15 is empty, the confidence is medium; the calculation result of the current benchmark price can be marked as medium.

Benchmark price calculation method

The calculation method of the benchmark price is to mark the calculation method of the benchmark price and the corresponding data, so as to find the reason when there is a problem later. The following marking is only for reference. Among them, super high-rise, high-rise, multi-storey, low-rise, flat-storey, and residential means that the listing cases involve all the housing cases under the super-high-rise, high-rise, multi-storey, low-rise, flat-storey and residential.

(1) If standard_price_type=="parameter adjustment" and Count>=6, the calculation method is marked C1A;

(2) If standard_price_type=="parameter adjustment" and Count<6, the calculation method is marked C1B;

(3) If standard_price_type=="high-level" and Count>=6, mark C2A by calculation method;

(4) If standard_price_type=="high-level" and Count<6, the calculation method is used to mark C2B;

(5) If standard_price_type=="multi-layer" and Count>=6, mark C3A by calculation method;

(6) If standard_price_type=="multi-layer" and Count<6, mark C3B by calculation method;

(7) If standard_price_type=="super high-rise" and Count>=6, mark C4A with the calculation method;

(8) If standard_price_type=="super high-rise" and Count<6, the calculation method is used to mark C4B;

(9) If standard_price_type=="low-level" and Count>=6, the calculation method is marked C5A;

(10) If standard_price_type=="low-level" and Count<6, mark C5B with the calculation method;

(11) If standard_price_type=="to be determined" or empty, and the evaluation price is empty, and Count>=6, the calculation method is marked C6A;

(12) If standard_price_type=="to be determined" or empty, and the evaluation price is empty, and Count<6, the calculation method is marked C6B;

(13) If standard_price_type=="to be determined" or empty, and the appraised price is not empty, the calculation method is marked C22;

(14) If standard_price_type=="manual", mark C21 by calculation method;

(15) If standard_price_type=="leveling", the calculation method is marked C20;

(16) If standard_price_type=="residential", the calculation method is marked with C19;

(17) The calculation method of random forest or KNN for price prediction is marked C16;

(18) Use the price calculated by the price of the secondary cell, and mark C18 with the calculation method;

(19) The average price of the administrative region, the calculation method is marked C23;

(20) The average price of the city, the calculation method is marked C24.

After the calculation of the benchmark price in each cycle is completed, the calculation results of the benchmark price can be verified. The verification contents mainly include: 1. The completeness of the benchmark price calculation; 2. The maximum and minimum benchmark prices in each city; Distribution of confidence levels; 4. The distribution of benchmark prices in various calculation methods; 5. Whether the results of benchmark prices in each city are repeated; ) limit, the quantity distribution beyond the up-down parameter, and within the up-down parameter. The field names output by the benchmark price verification mechanism are shown in Table 5 below. Table 5 shows the correspondence table between the field names of the benchmark price verification mechanism and the corresponding text descriptions.

table 5

By marking the confidence of the calculation results, the accuracy of the benchmark price of different calculation methods can be accurately analyzed. For cells with inaccurate calculations, the calculation method and the data used can be obtained to analyze the reasons for the inaccuracy and make timely adjustments, such as adjusting the calculation method to ensure the accuracy of the calculation results.

For the correction factor, it can be determined in the following ways:

In some embodiments, the coefficient correction system can be divided into three levels: city coefficient, building coefficient, and community coefficient.

The city coefficient takes into account the differences in housing market prices caused by factors such as decoration, orientation, area, and floors, and sets a correction coefficient to reflect the impact of these four factors in the valuation.

The building coefficient reflects the influence of different building attributes in the community on the house price. It is a key step in the valuation process and can effectively increase the assessment accuracy.

The community coefficient is a correction coefficient that reflects the pros and cons of different houses in the community, which can effectively eliminate the estimation deviation caused by the different specific conditions of the houses.

Among them, the urban correction coefficient includes four categories: area coefficient, orientation coefficient, decoration coefficient, and floor coefficient. The area coefficient is divided into several segments according to the area, and each segment corresponds to a coefficient. The orientation correction coefficient is divided according to the different orientations of the house, and each orientation corresponds to a coefficient. Decoration is divided into rough, simple, hardcover and high-end, and each decoration grade corresponds to a coefficient. Floor correction coefficient, corresponding to different correction coefficients according to different total floors and current floors.

The building coefficient mainly reflects the value difference brought by the property types, property rights, street conditions, landscape conditions and other building attributes of different buildings. Property types refer to multi-storey, high-rise, super high-rise and villas, and property rights refer to residential, commercial and residential, apartments and other different property rights years. Comparing the pros and cons of buildings is actually comparing the different properties of buildings. When estimating, if it can be matched to a specific building, select the coefficient for calculation; otherwise, take the default value of 1.

The community coefficient includes four categories: property type, property property, plane structure (flat, jump/duplex), and unit structure. Plane structure refers to indoor structures such as flat floor, jump floor, duplex, attic, etc., and unit structure refers to the price difference caused by different unit structures under the same area.

If the residence to be assessed can be attached to a building and the building coefficient is not empty, use this coefficient for calculation, ignoring the property type coefficient and property right property coefficient in the community coefficient.

If the residence to be assessed cannot be attached to the building, judge it as multi-storey, high-rise or super high-rise according to the total floor and then read the corresponding community coefficient (property type); The apartment reads the corresponding community coefficient (property property) for valuation calculation.

Based on the above three levels, the attribute data set corresponding to the house can be constructed, and then feature extraction and cluster analysis can be performed. After that, the determination of the correction coefficient of each target object can be performed based on the clustering result.

Each embodiment in this specification is described in a progressive manner, and the same and similar parts between the various embodiments may be referred to each other, and each embodiment focuses on the differences from other embodiments. For details, reference may be made to the descriptions of the foregoing related processing-related embodiments, which will not be repeated here.

The foregoing describes specific embodiments of the present specification. 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 in the embodiments and still achieve desirable results. Additionally, the processes depicted in the figures do not necessarily require the particular order shown, or sequential order, to achieve desirable results. In some embodiments, multitasking and parallel processing are also possible or may be advantageous.

In the business processing method provided by one or more embodiments of this specification, the calibration data displayed by target objects with certain similar characteristics in a certain historical time interval and/or the business evaluation data of the target object by the business system itself is used as a sample case. , evaluating the amount of resources represented by the current target object at the current time can further increase the number of sample cases on which the evaluation depends, thereby improving the accuracy of the evaluation. At the same time, it can further improve the processing efficiency of the business processing system and reduce the risk of financial institutions losing funds.

Based on the service processing method described above, one or more embodiments of this specification further provide a service processing apparatus. The apparatuses may include systems, software (applications), modules, components, servers, etc. that use the methods described in the embodiments of this specification, in combination with apparatuses that implement necessary hardware. Based on the same innovative idea, the apparatuses in one or more embodiments provided by the embodiments of this specification are described in the following embodiments. Since the implementation solution of the device to solve the problem is similar to the method, the implementation of the specific device in the embodiment of the present specification can refer to the implementation of the foregoing method, and repeated details will not be repeated. As used below, the term "unit" or "module" may be a combination of software and/or hardware that implements a predetermined function. Although the apparatus described in the following embodiments is preferably implemented in software, implementations in hardware, or a combination of software and hardware, are also possible and contemplated. Specifically, FIG. 2 shows a schematic diagram of a module structure of an embodiment of a service processing apparatus provided in the specification. As shown in FIG. 2 , the apparatus may include:

The request receiving module 102 can be configured to receive a current service processing request sent by the terminal device, where the current service processing request includes object information of the current target object corresponding to the service processing;

The data set determination module 104 can be configured to determine the object data set to which the current target object belongs according to the object information; the object data set includes the data of multiple target objects whose object attribute characteristics meet preset requirements within a specified time interval. Data sets consisting of calibration values and/or business evaluation values;

The data acquisition module 106 can be configured to acquire the reference value corresponding to the object data set, the reference value is determined according to the calibration value and/or service evaluation value of at least one target object in the object data set;

The correction processing module 108 can be configured to perform correction processing on the reference value by using the correction coefficient of the current target object to obtain a business evaluation value of the current target object; the correction coefficient is based on the current target object and the Determining the attribute difference characteristics between each target object in the object dataset;

The service processing module 110 may be configured to process the current service processing request by using the service evaluation value of the current target object.

In other embodiments, the apparatus may further include a reference value determination module, and the reference value determination module may include:

a data acquisition unit, which can be used to acquire a first target object that includes a calibration value within a first specified time interval and a second target object that includes a business evaluation value within a second specified time interval in the object data set;

The first reference value determination unit may be configured to adjust parameters based on the reference value corresponding to the object dataset when determining that the number of first target objects in the object dataset is greater than or equal to a first specified threshold. The calibration value corresponding to the at least one first target object is processed to obtain the reference value corresponding to the object data set;

The second reference value determining unit may be configured to, when it is determined that the number of the first target objects in the object data set is less than a preset threshold, adjust parameters for at least one object in the object data set based on the reference value corresponding to the object data set A business evaluation value corresponding to a second target object is processed to obtain a reference value corresponding to the object data set;

Wherein, the reference value adjustment parameter may be determined according to the difference characteristic between the calibration value of at least one target object in the object data set or the business evaluation value and the transaction value.

In other embodiments, the apparatus may further include an adjustment module, and the adjustment module may include:

a confidence level determination unit, which can be configured to determine the confidence level of the reference value corresponding to the object data set according to the transaction volume of the target object in the object data set and the difference between the reference value and the transaction value in the third specified time interval;

The adjustment unit may be configured to adjust the data on which the benchmark price calculation of the object data set depends and the calculation method according to the confidence level of the benchmark value.

It should be noted that the above-mentioned apparatus may also include other implementations according to the description of the method embodiment. For a specific implementation manner, reference may be made to the description of the related method embodiments, which will not be repeated here.

In the service processing apparatus provided by one or more embodiments of this specification, the calibration data displayed by the target objects with certain similar characteristics in a certain historical time interval and/or the service evaluation data of the target objects by the business system itself is used as a sample case. , evaluating the amount of resources represented by the current target object at the current time can further increase the number of sample cases on which the evaluation depends, thereby improving the accuracy of the evaluation. At the same time, it can further improve the processing efficiency of the business processing system and reduce the risk of financial institutions losing funds.

The methods or apparatuses described in the above embodiments provided in this specification can realize business logic through computer programs and record them on a storage medium, and the storage medium can be read and executed by a computer to achieve the effects of the solutions described in the embodiments of this specification. Therefore, this specification also provides a service processing device, including a processor and a memory storing instructions executable by the processor, and when the instructions are executed by the processor, the instructions include the steps of the method described in any one of the foregoing embodiments.

The identity verification device provided by the embodiments of this specification may be a computer terminal, a server, or a similar computing device. Taking the server as an example, FIG. 3 is a block diagram of the hardware structure of the server to which the embodiments of this specification are applied. As shown in FIG. 3 , the server 100 may include one or more (only one is shown in the figure) processor 200 (the processor 200 may include, but is not limited to, a processing device such as a microprocessor MCU or a programmable logic device FPGA), A memory 300 for storing data, and a transmission module 400 for communication functions. Those skilled in the art can understand that the structure shown in FIG. 3 is only a schematic diagram, which does not limit the structure of the above electronic device. For example, the server 100 may also include more or fewer components than those shown in FIG. 3 , for example, may also include other processing hardware, such as databases or multi-level caches, GPUs, or have a different configuration than that shown in FIG. 3 .

The memory 300 may be used to store software programs and modules of application software, such as program instructions/modules corresponding to the search method in the embodiment of the present invention, the processor 200 executes various functions by running the software programs and modules stored in the memory 300 applications and data processing. Memory 300 may include high-speed random access memory, and may also include non-volatile memory, such as one or more magnetic storage devices, flash memory, or other non-volatile solid-state memory. In some instances, the memory 300 may further include memory located remotely from the processor 200, and these remote memories may be connected to the computer terminal through a network. Examples of such networks include, but are not limited to, the Internet, an intranet, a local area network, a mobile communication network, and combinations thereof.

The transmission module 400 is used to receive or transmit data via a network. The specific example of the above-mentioned network may include a wireless network provided by the communication provider of the computer terminal. In one example, the transmission module 400 includes a network adapter (Network Interface Controller, NIC), which can be connected to other network devices through the base station so as to communicate with the Internet. In one example, the transmission module 400 may be a radio frequency (Radio Frequency, RF) module, which is used to communicate with the Internet in a wireless manner.

The storage medium may include a physical device for storing information, and usually the information is digitized and then stored in an electrical, magnetic or optical medium. The storage medium may include: devices that use electrical energy to store information, such as various memories, such as RAM, ROM, etc.; devices that use magnetic energy to store information, such as hard disks, floppy disks, magnetic tapes, magnetic core memories, magnetic bubble memories, etc. USB stick; a device that stores information optically, such as a CD or DVD. Of course, there are other readable storage media, such as quantum memory, graphene memory, and so on.

It should be noted that the above-mentioned device may also include other implementations according to the description of the method embodiment. For a specific implementation manner, reference may be made to the description of the related method embodiments, which will not be repeated here.

The business processing device described in the above embodiment, by taking the calibration data displayed by the target object with certain similar characteristics to the outside in a certain historical time interval and/or the business evaluation data of the target object by the business system itself as a sample case, the current target The evaluation of the resource amount represented by the object at the current time can further increase the number of sample cases on which the evaluation depends, thereby improving the accuracy of the evaluation. At the same time, it can further improve the processing efficiency of the business processing system and reduce the risk of financial institutions losing funds.

The present specification also provides a service processing system, which can be an independent service processing system, and can also be applied to various computer data processing systems. The system described may be a single server, or may include a server cluster, system (including distributed system), software (application), Actual operating devices, logic gate circuit devices, quantum computers, etc., combined with terminal devices that implement necessary hardware. The business processing system may include a resource management system and a server, the resource management system includes one or more business subsystems, wherein the resource management system may be configured to send resource transfer data to the server; the server may It includes at least one processor and a memory that stores computer-executable instructions, and when the processor executes the instructions, the steps of the methods described in one or more of the above embodiments are implemented; one or more service subsystems of the resource management system It can be used to receive identification results and manage and control accounts of the target monitoring account type.

It should be noted that the above-mentioned system may further include other implementation manners according to the description of the method or apparatus embodiment, and the specific implementation manner may refer to the description of the related method embodiment, which will not be repeated here.

The business processing system described in the above-mentioned embodiment uses the calibration data displayed by the target object with certain similar characteristics to the outside within a certain historical time interval and/or the business evaluation data of the target object by the business system itself as a sample case, and the current target The evaluation of the resource amount represented by the object at the current time can further increase the number of sample cases on which the evaluation depends, thereby improving the accuracy of the evaluation. At the same time, it can further improve the processing efficiency of the business processing system and reduce the risk of financial institutions losing funds.

The embodiments of this specification are not limited to those that must conform to standard data models/templates or described in the embodiments of this specification. Some industry standards or implementations described using custom methods or examples with slight modifications can also achieve the same, equivalent or similar, or predictable implementation effects after deformations of the above-mentioned examples. Embodiments obtained by applying these modified or deformed data acquisition, storage, judgment, processing methods, etc., may still fall within the scope of the optional embodiments of this specification.

Each embodiment in this specification is described in a progressive manner, and the same and similar parts between the various embodiments may be referred to each other, and each embodiment focuses on the differences from other embodiments. In particular, as for the system embodiments, since they are basically similar to the method embodiments, the description is relatively simple, and for related parts, please refer to the partial descriptions of the method embodiments. In the description of this specification, description with reference to the terms "one embodiment," "some embodiments," "example," "specific example," or "some examples", etc., mean specific features described in connection with the embodiment or example , structure, material or feature is included in at least one embodiment or example of this specification. In this specification, schematic representations of the above terms are not necessarily directed to the same embodiment or example. Furthermore, the particular features, structures, materials or characteristics described may be combined in any suitable manner in any one or more embodiments or examples. Furthermore, those skilled in the art may combine and combine the different embodiments or examples described in this specification, as well as the features of the different embodiments or examples, without conflicting each other.

The above descriptions are merely examples of the present specification, and are not intended to limit the present specification. Various modifications and variations of this specification are possible for those skilled in the art. Any modification, equivalent replacement, improvement, etc. made within the spirit and principle of this specification shall be included within the scope of the claims of this specification.

Claims (10)

1. A service processing method is applied to a server and comprises the following steps:

receiving a current service processing request sent by terminal equipment, wherein the current service processing request comprises object information of a current target object corresponding to service processing;

determining an object data set to which the current target object belongs according to the object information; the object data set comprises a data set consisting of calibration values and/or service evaluation values of a plurality of target objects with object attribute characteristics meeting preset requirements in a specified time interval;

acquiring a reference value corresponding to the object data set, wherein the reference value is determined according to a calibration value and/or a service evaluation value of at least one target object in the object data set;

correcting the reference value by using the correction coefficient of the current target object to obtain a service evaluation value of the current target object; the correction coefficient is determined according to the attribute difference characteristics between the current target object and each target object in the object data set;

and processing the current service processing request by using the service evaluation value of the current target object.

2. The method of claim 1, wherein the reference value corresponding to the object data set is determined by:

acquiring a first target object containing a calibration value in a first specified time interval and a second target object containing a service evaluation value in a second specified time interval in the object data set;

when the number of the first target objects in the object data set is determined to be larger than or equal to a first designated threshold, processing a calibration value corresponding to at least one first target object in the object data set based on a reference value adjustment parameter corresponding to the object data set to obtain a reference value corresponding to the object data set;

when the number of the first target objects in the object data set is smaller than a preset threshold value, processing a service evaluation value corresponding to at least one second target object in the object data set based on a reference value adjusting parameter corresponding to the object data set to obtain a reference value corresponding to the object data set;

wherein the reference value adjustment parameter is determined according to a difference characteristic between a calibration value or a traffic evaluation value of at least one target object in the object data set and a transaction value.

3. The method of claim 1, further comprising:

acquiring a calibration value corresponding to a first target object or a first time corresponding to generation of a service evaluation value corresponding to a second target object, and determining a second time corresponding to a current reference value;

determining a correction index corresponding to a corresponding first target object or a corresponding second target object according to the trading value difference characteristics of the target objects in the object data set at the first time and the second time;

correcting the calibration value of the corresponding first target object or the service evaluation value of the second target object by using the correction index to obtain a corrected calibration value of the corresponding first target object or a corrected service evaluation value of the second target object;

and determining a reference value corresponding to the object data set by using the corrected calibration value of the first target object or the corrected service evaluation value of the second target object.

4. The method of claim 1, further comprising:

determining a reference value confidence corresponding to the object data set according to the transaction amount of the target object in the object data set in a third designated time interval and the difference value between the reference value and the transaction value;

and adjusting the data and the reference price calculation mode which depend on the reference price calculation of the object data set according to the reference value confidence.

5. The method of claim 3, further comprising:

sequentially and iteratively extracting target object data samples by taking the updating period as a unit from the second time, and stopping data extraction when the number of the target object data samples contained in the target object data samples correspondingly extracted by the set identifier corresponding to the object data set is greater than a second specified threshold value, or stopping data extraction when the number of the updating periods reaches a third specified threshold value; wherein the target object data sample comprises data corresponding to a first target object or a second target object;

and storing all the extracted target object data samples into the corresponding object data sets.

6. The method of claim 5, wherein when the target object data sample comprises data corresponding to a first target object and when the object data set contains a number of target object data samples greater than a second specified threshold, the method further comprises:

adjusting the maximum value and the minimum value of the calibration value in all the extracted target object data samples by using the numerical amplitude; the value of the numerical amplitude is determined according to the numerical amplitude of the last reference value updating period and the ratio of the maximum value to the minimum value of all the extracted target object data samples;

filtering target object data samples which are not in a value range formed by the adjusted calibration value maximum value and the calibration value minimum value;

when the number of the target object data samples left after filtering is larger than a second specified threshold value, storing the target object data samples left after filtering into the object data set;

otherwise, continuing to iteratively extract the target object data samples, repeating the processing steps until the number of the target object data samples remaining after filtering is larger than a second specified threshold or the number of updating cycles reaches a third specified threshold, stopping iteration, and storing the target object data samples remaining after the last iterative filtering into the object data set.

7. A service processing device applied to a server includes:

a request receiving module, configured to receive a current service processing request sent by a terminal device, where the current service processing request includes object information of a current target object corresponding to service processing;

the data set determining module is used for determining an object data set to which the current target object belongs according to the object information; the object data set comprises a data set consisting of calibration values and/or service evaluation values of a plurality of target objects with object attribute characteristics meeting preset requirements in a specified time interval;

the data acquisition module is used for acquiring a reference value corresponding to the object data set, and the reference value is determined according to a calibration value and/or a service evaluation value of at least one target object in the object data set;

the correction processing module is used for correcting the reference value by using the correction coefficient of the current target object to obtain a service evaluation value of the current target object; the correction coefficient is determined according to the attribute difference characteristics between the current target object and each target object in the object data set;

and the service processing module is used for processing the current service processing request by utilizing the service evaluation value of the current target object.

8. The apparatus of claim 7, further comprising a reference value determination module, the reference value determination module comprising:

a data acquisition unit, configured to acquire a first target object containing a calibration value in a first specified time interval and a second target object containing a traffic evaluation value in a second specified time interval in the object data set;

a first reference value determining unit, configured to, when it is determined that the number of the first target objects in the target data set is greater than or equal to a first specified threshold, process a calibration value corresponding to at least one first target object in the target data set based on a reference value adjustment parameter corresponding to the target data set, and obtain a reference value corresponding to the target data set;

a second reference value determining unit, configured to, when it is determined that the number of the first target objects in the object data set is smaller than a preset threshold, process a service evaluation value corresponding to at least one second target object in the object data set based on a reference value adjustment parameter corresponding to the object data set, so as to obtain a reference value corresponding to the object data set;

wherein the reference value adjustment parameter is determined according to a difference characteristic between a calibration value or a traffic evaluation value of at least one target object in the object data set and a transaction value.

9. The apparatus of claim 7, further comprising an adjustment module, the adjustment module comprising:

the confidence degree determining unit is used for determining the confidence degree of the reference value corresponding to the object data set according to the transaction amount of the target object in the object data set in a third designated time interval and the difference value between the reference value and the transaction value;

and the adjusting unit is used for adjusting the data and the calculation mode which are depended on by the reference price calculation of the object data set according to the confidence of the reference value.

10. A business processing system comprising at least one processor and a memory for storing processor-executable instructions which, when executed by the processor, implement steps comprising the method of any one of claims 1 to 6.

Images