WO2019100763A1 - 一种车险定损数据的处理方法、装置和处理设备 - Google Patents

一种车险定损数据的处理方法、装置和处理设备 Download PDF

Info

Publication number
WO2019100763A1
WO2019100763A1 PCT/CN2018/099998 CN2018099998W WO2019100763A1 WO 2019100763 A1 WO2019100763 A1 WO 2019100763A1 CN 2018099998 W CN2018099998 W CN 2018099998W WO 2019100763 A1 WO2019100763 A1 WO 2019100763A1
Authority
WO
WIPO (PCT)
Prior art keywords
damage
component
probability
loss
conclusion
Prior art date
Application number
PCT/CN2018/099998
Other languages
English (en)
French (fr)
Inventor
胡越
郭昕
章海涛
程丹妮
吴博坤
Original Assignee
阿里巴巴集团控股有限公司
胡越
郭昕
章海涛
程丹妮
吴博坤
Priority date (The priority date is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the date listed.)
Filing date
Publication date
Application filed by 阿里巴巴集团控股有限公司, 胡越, 郭昕, 章海涛, 程丹妮, 吴博坤 filed Critical 阿里巴巴集团控股有限公司
Priority to SG11202004728XA priority Critical patent/SG11202004728XA/en
Priority to EP18880390.2A priority patent/EP3716194A4/en
Publication of WO2019100763A1 publication Critical patent/WO2019100763A1/zh
Priority to US16/880,211 priority patent/US20200334638A1/en

Links

Images

Classifications

    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
    • G06Q10/00—Administration; Management
    • G06Q10/10—Office automation; Time management
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
    • G06Q40/00—Finance; Insurance; Tax strategies; Processing of corporate or income taxes
    • G06Q40/08—Insurance
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
    • G06N7/00—Computing arrangements based on specific mathematical models
    • G06N7/01—Probabilistic graphical models, e.g. probabilistic networks

Definitions

  • the embodiment of the present specification belongs to the technical field of computer data processing, and in particular, to a method, a device and a processing device for processing a car insurance fixed loss data.
  • the purpose of the embodiments of the present specification is to provide a method, a device and a processing device for processing a vehicle risk loss data, which can process the damage loss conclusion from the perspective of the combination of the damage components, and can effectively identify the missing damage component in the fixed damage conclusion, and improve the determination. Loss the accuracy of the conclusion and improve the user experience.
  • the method, device and processing device for processing the vehicle risk loss data provided by the embodiments of the present specification are implemented by the following methods:
  • a method for processing car damage loss data comprising:
  • the damaged component combination including at least one damaged component
  • the damage-related component is used as a missing damage component of the damage loss conclusion.
  • a data processing device for displaying interface content comprising:
  • a receiving module configured to receive a fixed loss conclusion of the automobile insurance
  • a probability calculation module configured to calculate a probability of occurrence of a combination of damaged components in the determined loss decision based on historical loss determination data, the damage component combination including at least one damage component;
  • An association component determining module configured to determine whether there is a damage association component that matches the damaged component when the probability is greater than the first threshold
  • the first output module is configured to use the damage-related component as a missing damage component of the fixed loss conclusion when the matched damage-related component is queried.
  • a processing device includes a processor and a memory for storing processor-executable instructions that, when executed by the processor, are implemented:
  • the damaged component combination including at least one damaged component
  • the damage-related component is used as a missing damage component of the damage loss conclusion.
  • the damaged component combination including at least one damaged component
  • the risk prompt message is sent.
  • the method, device and processing device for processing the vehicle risk loss data can calculate the probability of the damage component combination in the occurrence of the loss determination conclusion by combining the case information of the damage component combination in the historical damage loss conclusion data. Probability can represent the reliability of a fixed loss conclusion. If the probability is greater than a certain threshold, it can be said that the damage component combination in the fixed loss conclusion is a common damage combination (also called a high frequency damage combination), which is the probability of the normal component combination.
  • a component if a component is damaged, it can be checked whether the component associated with the component is also damaged, and if so, the recommendation of the missing component can be performed to supplement the damage claim or Correction, solve the problem of outputting unreasonable fixed loss conclusions in some scenarios, effectively improve the accuracy and reliability of the output loss determination conclusion, and improve the user experience.
  • FIG. 2 is a schematic flow chart of another embodiment of the method described in the present specification.
  • FIG. 3 is a block diagram showing the hardware structure of an embodiment of a mobile terminal for processing a car insurance loss data according to the present specification
  • FIG. 4 is a block diagram showing the structure of an embodiment of a vehicle risk determination data processing apparatus provided by the present specification
  • FIG. 5 is a schematic structural diagram of a module of another embodiment of the apparatus provided by the present specification.
  • FIG. 6 is a schematic structural diagram of a module of another embodiment of the apparatus provided by the present specification.
  • FIG. 7 is a schematic diagram of a system framework of a fixed loss decision system constructed using the method described in the present specification.
  • the damaged component may include more than one. If a collision occurs in the right front of the vehicle, the damaged component may usually include multiple components such as a front bumper, a lamp, a tire, a fender, etc., and one or more damaged components damaged in a single car may be called damage.
  • Component combination may include components that are connected at the appearance of the vehicle, or may include components that are externally connected to the interior of the vehicle, or damage of multiple sub-components as a single component.
  • a combination of components such as a combination of damaged components including two damaged components of a rear view mirror frame and a rear view mirror glass.
  • components that are not directly connected in position such as a combination of damaged components of a vehicle taillight and a center console lighting controller, may also be included.
  • the fog lamp is broken, and the high probability fog lamp frame is broken.
  • the right rear fender, the right bottom and the right rear door are common combinations of damage components.
  • the right rear door has a higher probability of simultaneous damage.
  • the probability of occurrence of damage to other components in the occurrence of certain damaged components in the loss-reduction process can be obtained, and the historical damage can be utilized in the processing of the fixed-loss conclusion.
  • the data determines whether the combination of damaged components in the conclusion of the damage is a combination of damage in the normal case (specifically, according to the definition of the scene). If yes, it is possible to further find out whether there is a damage-related component of the damaged component in the determination of the damage, and if so, it can be used as a missing damage component to supplement or correct the damage conclusion, thereby improving the accuracy of the loss determination conclusion.
  • the embodiments provided by this specification may utilize Bayesian inference methods to calculate the probability of occurrence of a combination of damaged components in the determined loss decision.
  • a Bayesian inference engine can be designed, which can obtain historical loss and loss conclusion data from historical damage cases, stored in the database, and based on the prior probability and conditional probability of the damaged components in a large number of historical damage predictions. The reliability of the damage conclusion.
  • the prior probability and the conditional probability the corresponding formula can be as follows:
  • damage combination x) Pro (damage component y, damage combination x) / Pro (damage combination x).
  • a low conditional probability may indicate that the associated damage component combination is less likely to occur in a historical case and can be considered as a suspicious loss decision.
  • the prior probability and conditional probability calculations can be updated by historical loss determination data in the database. Of course, in the case of computer performance, it can also be obtained by real-time flow engine real-time calculation.
  • FIG. 1 is a schematic flowchart diagram of an embodiment of a data processing method for displaying interface content provided by the present specification.
  • the present specification provides method operation steps or device structures as shown in the following embodiments or figures, there may be more or partial merged fewer operational steps in the method or device based on conventional or no inventive labor. Or module unit.
  • the execution order of the steps or the module structure of the device is not limited to the execution order or the module structure shown in the embodiment or the drawings.
  • server or terminal product of the method or module structure When the device, server or terminal product of the method or module structure is applied, it may be executed sequentially or in parallel according to the method or module structure shown in the embodiment or the drawing (for example, parallel processor or multi-thread processing). Environment, even including distributed processing, server cluster implementation environment).
  • the method may include:
  • the determination of the damage may include information of the identified damaged component of the vehicle, such as the name of the component of the damaged component, the degree of damage, the location of the damage, and the like.
  • the user can input the damage loss conclusion into the fixed loss decision system, for example, the fixed loss conclusion obtained by the manual damage.
  • the fixed loss decision may also be transmitted to the fixed loss decision system by other terminal devices.
  • the fixed loss result obtained by the fixed loss server through the fixed loss image recognition process is sent to the fixed loss decision system.
  • the fixed loss decision system can process or persist the post processing.
  • the damaged component combination may be implemented by including one damaged component. If the damage component combination in the damage determination conclusion is that the fog lamp is damaged, then the fog lamp frame with a high probability is also damaged. Therefore, the damage component combination in this embodiment may include one damage component, so that the missing damage component can be obtained based on the subsequent damage of the fog lamp to match the component fog lamp frame.
  • the Bayesian inference method can be used to calculate the probability of the occurrence of the damaged component in the fixed loss conclusion in combination with the historical loss determination data stored in the database.
  • this specification does not exclude other embodiments, other statistical or inductive or predictive algorithms, or custom algorithms or models may be used, and historical loss determination data may be used to obtain the probability of occurrence of a combination of damaged components in the determined loss conclusion.
  • the probability of occurrence of the combination of damaged components is greater than a certain threshold, it can be indicated that the combination of damaged components belongs to a conventional combination of damaged components (or becomes a combination of high frequency damaged components).
  • the analysis result of the historical loss determination data can then be used to query whether there is a damage-related component that matches the damaged component.
  • the historical damage prediction data can be statistically obtained to obtain information about another damaged component when one component is damaged.
  • a training simple learning model can be established, and the historical loss record data can be used as sample data for training learning, such as CNN ((Deep Neural Networks), GBDT (GradientBoosting DecisionTree), iterative decision tree algorithm), SVM (Support) Vector Machine, support vector machine, etc.
  • the present embodiment can further inquire whether there is a damage-related component that matches the damaged component.
  • whether the query has a damage association component that matches the damaged component may include:
  • S40 Query the damage correlation component of the damage component in the historical association rule, and the historical association rule includes determining, according to the historical damage determination data, the second damage component that is damaged when the first damage component is present.
  • Historical association rules may be generated based on information that is damaged by other components (which may be referred to as the second component), depending on the damage to a component (which may be referred to as the first component) in the historical loss determination data. For example, when component A is damaged, in some historical damage conclusions, component B will also be damaged. In other historical damage conclusions, component C will be damaged and B will not be damaged. The situation, of course, can occur when component A is damaged, and parts B and C are also damaged. In this way, the historical association rule can be generated according to the processing of the historical loss determination conclusion data, and the historical association rule can record the information of the damaged second component when the first component is damaged, the second component Can be one or more. If there is a history association rule that "Part A is damaged when component A is damaged", there may be a history association rule of "Part A is damaged when component A is damaged”.
  • a historical association rule of a component may have a corresponding degree of confidence, and the confidence may include when the first damage component is based on historical loss determination data. The probability that the second damaged component is damaged is determined, and the damage-related component with a confidence greater than a threshold may be selected as the matched damage-related component when determining the damaged component.
  • the damage correlation component in the historical association rule is filtered by using the confidence degree, and the high confidence level greater than the threshold is selected as the matched damage correlation component, which can further improve the accuracy of identifying and searching for missing damage components, and improve the damage loss conclusion. Reliability and accuracy.
  • the threshold for selecting the confidence level may be set in combination with the application scenario. For example, the damage correlation component corresponding to the confidence level greater than 90% may be set.
  • the first threshold value may be set to 0.5%
  • the damage component combination includes the damage component A and the damage component B.
  • the probability of occurrence of the combination of the damaged component is calculated by the historical damage loss conclusion data to be 65%, indicating that the combination of the damaged component A and B is a common combination.
  • Querying the history case association rule shows that in 90% of cases, when the damaged parts A and B appear at the same time, the part C is also damaged.
  • the component C may be used as the damage-related component of the damaged component A, or may be the damage-related component of the damaged component B.
  • the damage-related component can be used as a missing damage component appearing in the conclusion of the loss, and the missing damage component can be sent to the designated receiver for manual review by pushing or prompting information.
  • the missing damage component may be re-nucleated as the damage component in the determination of the damage, such that the output damage determination may include missing damage components and the originally included damage components.
  • the problem of the loss determination can be improved and corrected, and the problem of outputting the unreasonable fixed loss conclusion in some scenarios can be solved, and the accuracy and reliability of the output loss determination conclusion can be effectively improved, and the user experience is improved.
  • the actual fixed loss order and a large amount of historical data indicate the right rear fender, the right bottom large side and the right rear door. It is a common combination of damage, and the right rear door damage probability is 90% when the right rear fender and the right bottom are damaged at the same time.
  • another situation that does not conform to a conventional component combination such as human fraud
  • Malicious users can achieve the illegal purpose of car damage determination by forging a loss image, intentionally shooting at an abnormal angle, or even using a fixed-loss image of a different vehicle.
  • the conventional component combination is not met, such as the right front door collision, the right front wheel, and the right front A-pillar are damaged, but the right turn signal is not damaged.
  • it may be car insurance fraud or Collisions that occur at a specific angle.
  • the damage component combination includes at least two damage components, and if a probability of occurrence of the damage component combination is less than a certain threshold, a risk prompt message may be issued, The damage component combination is prompted for manual verification or re-identification processing. Therefore, in another embodiment of the method provided by the present specification, if the damage component combination includes at least two damage components, the method may further include:
  • S8 Send a risk prompt message when the probability is determined to be lower than the second threshold.
  • the damaged component combination may include an implementation scenario of the damaged component. If the probability of occurrence of the combination of damaged components is less than a threshold (which may be referred to herein as a fourth threshold), a risk alert message may also be issued.
  • a threshold which may be referred to herein as a fourth threshold
  • the damaged component is combined into a component inside the vehicle, such as an armrest box. In the case of a large number of vehicles, the possibility of damage to the armrest box is extremely low.
  • a risk alert message may be sent if the damaged component combination includes only one damaged component and the probability of the damaged component combination occurring is less than a fourth threshold.
  • the risk alert message is sent if the damage component combination includes a damage component and the probability is determined to be below a fourth threshold.
  • the calculating the probability of occurrence of the damaged component combination based on historical loss determination data includes:
  • the probability of occurrence of the combination of the damaged component is zero.
  • the third threshold it is determined that the probability of occurrence of the combination of the damaged component A and the damaged component M in the current loss determination conclusion is zero.
  • the specific condition data corresponding to the determination result is also obtained,
  • the specific condition data includes at least one of data of a collision angle, a collision strength, an accident occurrence place, an accident occurrence, and an accident type;
  • the specific condition data corresponding to the determined loss conclusion matches the specific condition data in the historical loss determination conclusion data, determining that the probability of occurrence of the damaged component group in the determined loss conclusion is greater than a first threshold.
  • the specific condition data may include such a scene description: the sunroof is open, the ground is close to the building, and the parabolic object is high. Under such specific conditions, it is possible to send the case that only the dangerous armrest box is damaged. . If such a scene has appeared in history, and the specific condition of the currently processed fixed loss conclusion is also such a scene, the component damage scene environment is the same or similar, then the probability that it can be output is greater than the first threshold, indicating that the currently occurring damage component combination is Such a specific condition meets the normal probability of occurrence.
  • the present embodiment combines the data information under specific conditions to process the loss determination conclusion, which can further improve the reliability of the loss determination conclusion.
  • the data for judging the omission or the risk may include not only historical loss data but also other data such as collision marks.
  • the fixed loss conclusion data after manual review or adding missing damage components can be used as new historical loss determination conclusion data, so that the historical loss determination conclusion data can be improved through continuous data accumulation. Reliable, it also makes the subsequent model loss data processing results more and more accurate and reliable.
  • the method may further include:
  • the first corrected loss determination result obtained by modifying the determined loss conclusion based on the missing damage component
  • the second corrected loss determination result obtained by reviewing the determined loss conclusion based on the risk warning message.
  • the corrected loss determination conclusion may include any one of the first or second corrected loss determination conclusions described above, or both.
  • FIG. 7 is a system framework diagram of a loss determination decision system constructed using the methods described herein, with the dashed line portion indicating portions that may not be included in some embodiments.
  • the embodiment of the present specification provides a set of efficient and accurate processing method for the damage data of the vehicle insurance, which can output a higher precision fixed loss result, and provides a mechanism for automatically diverting the suspicious fixed loss conclusion, and the suspicious item of the fixed loss combination is performed.
  • the risk warning can identify cases with suspected fraud, and at the same time, when the conclusion of the algorithm is unreliable, the human intervention corrects the conclusion of the damage integrity to improve the user experience and reduce the risk of fraud.
  • FIG. 3 is a hardware structural block diagram of a mobile terminal for processing a car insurance loss data according to an embodiment of the present invention.
  • the mobile terminal 10 may include one or more (only one shown) processor 102 (the processor 102 may include, but is not limited to, a processing device such as a microprocessor MCU or a programmable logic device FPGA).
  • the mobile terminal 10 may also include more or fewer components than those shown in FIG. 7, for example, may also include other processing hardware, or have a different configuration than that shown in FIG.
  • the memory 104 can 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, and the processor 102 executes various functions by running software programs and modules stored in the memory 104.
  • Application and data processing that is, the processing method for realizing the above-mentioned auto insurance loss data.
  • Memory 104 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.
  • memory 104 may further include memory remotely located relative to processor 102, which may be coupled to computer terminal 10 via a network. Examples of such networks include, but are not limited to, the Internet, intranets, local area networks, mobile communication networks, and combinations thereof.
  • the transmission module 106 is configured to receive or transmit data via a network.
  • the network specific examples described above may include a wireless network provided by a communication provider of the computer terminal 10.
  • the transport module 106 includes a Network Interface Controller (NIC) that can be connected to other network devices through a base station to communicate with the Internet.
  • the transmission module 106 can be a Radio Frequency (RF) module for communicating with the Internet wirelessly.
  • NIC Network Interface Controller
  • RF Radio Frequency
  • FIG. 4 is a schematic structural diagram of a module of a data processing apparatus for displaying interface content provided by the present disclosure, which may include:
  • the receiving module 101 can be configured to receive a fixed loss conclusion of the automobile insurance
  • the probability calculation module 102 may be configured to calculate a probability of occurrence of the damage component combination in the determination of the damage based on the historical damage determination data, the damage component combination including at least one damage component;
  • the association component determining module 103 may be configured to: when determining that the probability is greater than the first threshold, query whether there is a damage association component that matches the damaged component;
  • the first output module 104 may be configured to use the damage-related component as a missing damage component of the fixed loss conclusion when the matched damage-related component is queried.
  • the apparatus may further include:
  • the second output module 104 may be configured to send the risk prompt message when the probability calculation module 102 obtains that the probability is lower than the second threshold.
  • the probability calculation module 102 may include:
  • the Bayesian inference unit can be used to calculate the probability of the combination of the damaged components based on the prior probability and the conditional probability of the damage component appearing in the historical loss determination data using a Bayesian inference method.
  • the probability calculation module 102 calculates the probability of occurrence of the damaged component combination based on the historical impairment loss data:
  • the probability of occurrence of the combination of the damaged component is zero.
  • the specific condition data corresponding to the fixed loss conclusion is also acquired, the specific condition The data includes at least one of data of a collision angle, a collision strength, an accident occurrence location, an accident occurrence, and an accident type;
  • the specific condition data corresponding to the determined loss conclusion matches the specific condition data in the historical loss determination conclusion data, determining that the probability of occurrence of the damaged component group in the determined loss conclusion is greater than a first threshold.
  • the association component determining module 103 queries whether there is a damage association component that matches the damaged component, including:
  • the damage correlation component of the damage component is queried in a historical association rule, the historical association rule including determining a second damage component that is damaged when the first damage component is in the historical damage loss conclusion data.
  • the associated component determining module 103 may further include:
  • a screening unit configured to select a damage correlation component with a confidence greater than a threshold as the matched damage association component, where the confidence includes the second damage component when the first damage component is included in the historical damage loss conclusion data The probability of damage to the component is determined.
  • FIG. 6 is a block diagram showing another embodiment of the apparatus provided in the present specification. As shown in Figure 6, the apparatus may further include:
  • the historical data update module 106 may obtain the corrected loss determination conclusion, and use the corrected damage determination result as the historical damage loss conclusion data, wherein the corrected damage determination conclusion comprises:
  • the first corrected loss determination result obtained by modifying the determined loss conclusion based on the missing damage component
  • the second corrected loss determination result obtained by reviewing the determined loss conclusion based on the risk warning message.
  • the data processing method for displaying interface content can be implemented by a processor executing a corresponding program instruction in a computer, such as using a C++ language of a Windows operating system on a PC side, or other systems such as Linux, android, and iOS.
  • a processor executing a corresponding program instruction in a computer
  • the processing device may include a processor and a memory for storing processor-executable instructions, when the processor executes the instruction:
  • the damaged component combination including at least one damaged component
  • the damage-related component is used as a missing damage component of the damage loss conclusion.
  • the above instructions may be stored in a variety of computer readable storage media.
  • the computer readable storage medium may include physical means for storing information, which may be digitized and stored in a medium utilizing electrical, magnetic or optical means.
  • the computer readable storage medium of this embodiment may include: means for storing information by means of electrical energy, such as various types of memories, such as RAM, ROM, etc.; means for storing information by magnetic energy means, such as hard disk, floppy disk, magnetic tape, magnetic Core memory, bubble memory, U disk; means for optically storing information such as CD or DVD.
  • electrical energy such as various types of memories, such as RAM, ROM, etc.
  • magnetic energy means such as hard disk, floppy disk, magnetic tape, magnetic Core memory, bubble memory, U disk
  • means for optically storing information such as CD or DVD.
  • quantum memories graphene memories, and the like.
  • the embodiment of the present specification further provides a processing device for car insurance loss data
  • the processing device may include a mobile terminal, a personal applause computer, a smart wear device, a vehicle interaction device, a personal computer, a server, and a server.
  • the processing device can include at least one processor and a memory for storing processor-executable instructions that, when executed by the processor, are implemented:
  • the damaged component combination including at least one damaged component
  • the risk prompt message is sent.
  • processing device and the electronic device described in the foregoing embodiments of the present specification may further include other embodiments according to the description of the related method embodiments, for example.
  • the above-mentioned computer-readable storage medium may further include other embodiments according to the description of the method or the device embodiment.
  • reference may be made to the description of the method embodiments, and details are not described herein.
  • the method, device and processing device for processing the vehicle risk loss data can calculate the probability of the damage component combination in the occurrence of the loss determination conclusion by combining the case information of the damage component combination in the historical damage loss conclusion data. Probability can represent the reliability of a fixed loss conclusion. If the probability is greater than a certain threshold, it can be said that the damage component combination in the fixed loss conclusion is a common damage combination (also referred to as a high frequency damage combination), which is a probability of occurrence of a normal component combination.
  • a component if a component is damaged, it can be checked whether the component associated with the component is also damaged, and if so, the recommendation of the missing component can be performed to supplement the damage claim or Correction, solve the problem of outputting unreasonable fixed loss conclusions in some scenarios, effectively improve the accuracy and reliability of the output loss determination conclusion, and improve the user experience.
  • PLD Programmable Logic Device
  • FPGA Field Programmable Gate Array
  • HDL Hardware Description Language
  • the controller can be implemented in any suitable manner, for example, the controller can take the form of, for example, a microprocessor or processor and a computer readable medium storing computer readable program code (eg, software or firmware) executable by the (micro)processor.
  • computer readable program code eg, software or firmware
  • examples of controllers include, but are not limited to, the following microcontrollers: ARC 625D, Atmel AT91SAM, The Microchip PIC18F26K20 and the Silicone Labs C8051F320, the memory controller can also be implemented as part of the memory's control logic.
  • the controller can be logically programmed by means of logic gates, switches, ASICs, programmable logic controllers, and embedding.
  • Such a controller can therefore be considered a hardware component, and the means for implementing various functions included therein can also be considered as a structure within the hardware component.
  • a device for implementing various functions can be considered as a software module that can be both a method of implementation and a structure within a hardware component.
  • the system, device, module or unit illustrated in the above embodiments may be implemented by a computer chip or an entity, or by a product having a certain function.
  • a typical implementation device is a computer.
  • the computer can be, for example, a personal computer, a laptop computer, a car-mounted human-machine interaction device, a cellular phone, a camera phone, a smart phone, a personal digital assistant, a media player, a navigation device, an email device, a game console, a tablet.
  • the above devices are described as being separately divided into various modules by function.
  • the functions of the modules may be implemented in the same software or software, or the modules that implement the same function may be implemented by multiple sub-modules or a combination of sub-units.
  • the device embodiments described above are merely illustrative.
  • the division of the unit is only a logical function division.
  • there may be another division manner for example, multiple units or components may be combined or integrated. Go to another system, or some features can be ignored or not executed.
  • the mutual coupling or direct coupling or communication connection shown or discussed may be an indirect coupling or communication connection through some interface, device or unit, and may be in an electrical, mechanical or other form.
  • the controller can be logically programmed by means of logic gates, switches, ASICs, programmable logic controllers, and embedding.
  • the computer program instructions can also be stored in a computer readable memory that can direct a computer or other programmable data processing device to operate in a particular manner, such that the instructions stored in the computer readable memory produce an article of manufacture comprising the instruction device.
  • the apparatus implements the functions specified in one or more blocks of a flow or a flow and/or block diagram of the flowchart.
  • These computer program instructions can also be loaded onto a computer or other programmable data processing device such that a series of operational steps are performed on a computer or other programmable device to produce computer-implemented processing for execution on a computer or other programmable device.
  • the instructions provide steps for implementing the functions specified in one or more of the flow or in a block or blocks of a flow diagram.
  • a computing device includes one or more processors (CPUs), input/output interfaces, network interfaces, and memory.
  • processors CPUs
  • input/output interfaces network interfaces
  • memory volatile and non-volatile memory
  • the memory may include non-persistent memory, random access memory (RAM), and/or non-volatile memory in a computer readable medium, such as read only memory (ROM) or flash memory.
  • RAM random access memory
  • ROM read only memory
  • Memory is an example of a computer readable medium.
  • Computer readable media includes both permanent and non-persistent, removable and non-removable media.
  • Information storage can be implemented by any method or technology.
  • the information can be computer readable instructions, data structures, modules of programs, or other data.
  • Examples of computer storage media include, but are not limited to, phase change memory (PRAM), static random access memory (SRAM), dynamic random access memory (DRAM), other types of random access memory (RAM), read only memory. (ROM), electrically erasable programmable read only memory (EEPROM), flash memory or other memory technology, compact disk read only memory (CD-ROM), digital versatile disk (DVD) or other optical storage, Magnetic tape cartridges, magnetic tape storage or other magnetic storage devices or any other non-transportable media can be used to store information that can be accessed by a computing device.
  • computer readable media does not include temporary storage of computer readable media, such as modulated data signals and carrier waves.
  • embodiments of the present specification can be provided as a method, system, or computer program product.
  • embodiments of the present specification can take the form of an entirely hardware embodiment, an entirely software embodiment or a combination of software and hardware.
  • embodiments of the present specification can take the form of a computer program product embodied on one or more computer usable storage media (including but not limited to disk storage, CD-ROM, optical storage, etc.) including computer usable program code.
  • Embodiments of the present description can be described in the general context of computer-executable instructions executed by a computer, such as a program module.
  • program modules include routines, programs, objects, components, data structures, and the like that perform particular tasks or implement particular abstract data types.
  • Embodiments of the present specification can also be practiced in distributed computing environments where tasks are performed by remote processing devices that are connected through a communication network.
  • program modules can be located in both local and remote computer storage media including storage devices.

Landscapes

  • Business, Economics & Management (AREA)
  • Engineering & Computer Science (AREA)
  • Physics & Mathematics (AREA)
  • Theoretical Computer Science (AREA)
  • General Physics & Mathematics (AREA)
  • Finance (AREA)
  • Accounting & Taxation (AREA)
  • Strategic Management (AREA)
  • General Business, Economics & Management (AREA)
  • Marketing (AREA)
  • Economics (AREA)
  • Development Economics (AREA)
  • Technology Law (AREA)
  • Human Resources & Organizations (AREA)
  • Entrepreneurship & Innovation (AREA)
  • Data Mining & Analysis (AREA)
  • Quality & Reliability (AREA)
  • Operations Research (AREA)
  • Tourism & Hospitality (AREA)
  • Pure & Applied Mathematics (AREA)
  • Mathematical Physics (AREA)
  • Evolutionary Computation (AREA)
  • Computational Mathematics (AREA)
  • Algebra (AREA)
  • Software Systems (AREA)
  • General Engineering & Computer Science (AREA)
  • Computing Systems (AREA)
  • Artificial Intelligence (AREA)
  • Mathematical Optimization (AREA)
  • Mathematical Analysis (AREA)
  • Probability & Statistics with Applications (AREA)
  • Financial Or Insurance-Related Operations Such As Payment And Settlement (AREA)

Abstract

一种车险定损数据的处理方法、装置和处理设备，所述方法可以包括：结合历史定损结论数据中损伤部件组合的案件信息来计算得到出现定损结论中损伤部件组合的概率，该概率可以表示定损结论的可靠性。若概率大于一定的阈值，可以表示定损结论中的损伤部件组合是常见的损伤组合，为正常的部件组合出现概率。在上述方法中，如果某个部件受损，则可以查看与该部件相关联的部件是否也出现受损，如果有，则可以进行损伤遗漏部件的推荐，对定损结论进行补充或修正。

Description

一种车险定损数据的处理方法、装置和处理设备 技术领域
本说明书实施例方案属于计算机数据处理的技术领域，尤其涉及一种车险定损数据的处理方法、装置和处理设备。
背景技术
随着机动车辆的普及，车险业务也呈明显增加趋势。当车辆出险时，快速、准确的进行定损可以带来更好的用户体验。
目前业内出现了很多自动定损的方式。该方式中，用户可以对受损车辆进行拍摄，然后通过处理设备识别出受损部件，基于拍摄图像识别出的受损部件得到定损结论。这样的定损结论依赖于机器学习算法，会出现自动定损错误，出现不合常理的定损结果。并且，这样不和常理的定损结果也常常难以区分是由于拍摄角度、现场环境的影响还是人为故意欺诈造成。
因此，业内亟需一种可以进一步评估车险定损结论可靠性的处理方案。
发明内容
本说明书实施例目的在于提供一种车险定损数据的处理方法、装置和处理设备，从损伤部件组合的角度对定损结论进行处理，可以有效识别出定损结论中的遗漏损伤部件，提高定损结论准确性，提高用户体验。
本说明书实施例提供的一种车险定损数据的处理方法、装置和处理设备是包括以下方式实现的：
一种车险定损数据的处理方法，所述方法包括：
接收车险的定损结论；
基于历史定损结论数据计算出现所述定损结论中的损伤部件组合的概率，所述损伤部件组合包括至少一个损伤部件；
确定所述概率大于第一阈值时，查询是否有与所述损伤部件匹配的损伤关联部件；
若有，则将所述损伤关联部件作为所述定损结论的遗漏损伤部件。
一种展示界面内容的数据处理装置，包括：
接收模块，用于接收车险的定损结论；
概率计算模块，用于基于历史定损结论数据计算出现所述定损结论中的损伤部件组合的概率，所述损伤部件组合包括至少一个损伤部件；
关联部件确定模块，用于确定所述概率大于第一阈值时，查询是否有与所述损伤部件匹配的损伤关联部件；
第一输出模块，用于查询到匹配的损伤关联部件时，将所述损伤关联部件作为所述定损结论的遗漏损伤部件。
一种处理设备，包括处理器以及用于存储处理器可执行指令的存储器，所述处理器执行所述指令时实现：
接收车险的定损结论；
基于历史定损结论数据计算出现所述定损结论中的损伤部件组合的概率，所述损伤部件组合包括至少一个损伤部件；
确定所述概率大于第一阈值时，查询是否有与所述损伤部件匹配的损伤关联部件；
若有，则将所述损伤关联部件作为所述定损结论的遗漏损伤部件。
一种电子设备，包括至少一个处理器以及用于存储处理器可执行指令的存储器，所述处理器执行所述指令时实现：
接收车险的定损结论；
基于历史定损结论数据计算出现所述定损结论中的损伤部件组合的概率，所述损伤部件组合包括至少一个损伤部件；
确定所述概率大于第一阈值时，查询是否有与所述损伤部件匹配的损伤关联部件，若有，则将所述损伤关联部件作为所述定损结论的遗漏损伤部件；
确定所述概率低于第二阈值时，发送风险提示消息。
本说明书实施例提供的一种车险定损数据的处理方法、装置和处理设备，可以结合历史定损结论数据中损伤部件组合的案件信息来计算得到出现定损结论中损伤部件组合的概率，该概率可以表示定损结论的可靠性。若概率大于一定的阈值，可以表示定损结论中的损伤部件组合是常见的损伤组合(也可以称为高频损伤组合)，为正常的部件 组合出现概率。在本说明书提供实施方案中，如果某个部件受损，则可以查看与该部件相关联的部件是否也出现受损，如果有，则可以进行损伤遗漏部件的推荐，对定损结论进行补充或修正，解决一些场景下输出不合常理的定损结论的问题，有效提高输出的定损结论的精度和可靠性，提高用户体验。
附图说明
为了更清楚地说明本说明书实施例或现有技术中的技术方案，下面将对实施例或现有技术描述中所需要使用的附图作简单地介绍，显而易见地，下面描述中的附图仅仅是本说明书中记载的一些实施例，对于本领域普通技术人员来讲，在不付出创造性劳动性的前提下，还可以根据这些附图获得其他的附图。
图1是本说明书所述方法实施例的一个处理流程示意图；
图2是本说明书所述方法另一个实施例流程示意图；
图3是本说明书一种车险定损数据的处理方法的移动终端实施例的硬件结构框图；
图4是本说明书提供的一种车险定损数据处理装置实施例的模块结构示意图；
图5是本说明书提供的所述装置另一种实施例的模块结构示意图；
图6是本说明书提供的所述装置另一种实施例的模块结构示意图；
图7是利用本说明书所述方法构建的一种定损决策系统的系统框架示意图。
具体实施方式
为了使本技术领域的人员更好地理解本说明书中的技术方案，下面将结合本说明书实施例中的附图，对本说明书实施例中的技术方案进行清楚、完整地描述，显然，所描述的实施例仅仅是本说明书中的一部分实施例，而不是全部的实施例。基于本说明书中的一个或多个实施例，本领域普通技术人员在没有作出创造性劳动前提下所获得的所有其他实施例，都应当属于本说明书实施例保护的范围。
在车辆受损时，多数情况下是受损的部件可能包括多个。如车辆右前方发生碰撞，受损的部件通常可以包括前保险杠、车灯、车胎、挡泥板等多个部件，这些在单次车险中受损的一个或多个损伤部件可以称为损伤部件组合。在本说明书的一些实施例中，所述的损伤部件组合可以包括在车辆外观位置上相连接的部件，也可以包括外观与车辆内 部相连接的部件，或者作为一个整体部件的多个子部件的损伤部件组合，如包括后视镜框与后视镜玻璃两个损伤部件的损伤部件组合。其他的一些实施例中，也可以包括位置上不直接相连接的部件，如车辆尾灯与中控台的灯光控制器组成的损伤部件组合。
通常情况下，在车辆发生碰撞时，如果一个部件受损，常常会伴随周边位置部件的损坏，雾灯坏了，大概率雾灯框也坏了。例如右后翼子板、右底大边和右后门是常见的损伤部件组合，而且，基于历史定损结论数据可以看出，当右后翼子板和右底大边框同时出现损伤情况下，右后门有较大概率的会同时出现损伤。现有的一些基于拍摄图像识别损伤部件的方法中，由于拍摄角度、识别算法自身逻辑、定损图像质量等，在定损结论中常常会出现遗漏某些损伤部件的情况。本说明书提供的实施方案，至少可以根据历史定损结论数据，得到在定损处理中出现某些损伤部件时伴随着其他部件损伤的概率情况，在进行定损结论处理时，可以利用历史定损结论数据判断定损结论中的损伤部件组合是否为常规情况(具体可以根据场景定义)的损伤组合。如果是，则可以进一步的查找是否有定损结论中损伤部件的损伤关联部件，如果有，则可以将其作为遗漏损伤部件，补充或修正定损结论，进而提高定损结论的准确性。
本说明书提供的实施方案可以利用贝叶斯推理的方法来计算出现所述定损结论中的损伤部件组合的概率。例如可以设计一种贝叶斯推理引擎，可以从历史定损案件中得到历史定损结论数据，存储在数据库中，基于大量历史定损结论中损伤部件出现的先验概率与条件概率，识别定损结论的可靠性。在本说明书一些利用贝叶斯推理实施例中，先验概率和条件概率，对应公式可以如下所示：
先验概率计算：Pro(损伤组合x)＝Num(损伤组合x)/Num(历史定损案件)；
条件概率计算：Pro(损伤部件y|损伤组合x)＝Pro(损伤部件y,损伤组合x)/Pro(损伤组合x)。
条件概率低可以表示相关的损伤部件组合在历史案件中出现可能性低，可以视为可疑定损结论。先验概率以及条件概率计算可以通过数据库中的历史定损结论数据更新，当然，在计算机性能允许的情况下，也可以通过实时流引擎实时计算获得。
下面以一个定损决策系统来处理定损中包括多个损伤部件的受损结论的过程为应用场景对本说明书实施方案进行说明。具体的，图1是本说明书提供的所述一种展示界面内容的数据处理方法实施例的流程示意图。虽然本说明书提供了如下述实施例或附图所示的方法操作步骤或装置结构，但基于常规或者无需创造性的劳动在所述方法或装置中 可以包括更多或者部分合并后更少的操作步骤或模块单元。在逻辑性上不存在必要因果关系的步骤或结构中，这些步骤的执行顺序或装置的模块结构不限于本说明书实施例或附图所示的执行顺序或模块结构。所述的方法或模块结构的在实际中的装置、服务器或终端产品应用时，可以按照实施例或者附图所示的方法或模块结构进行顺序执行或者并行执行(例如并行处理器或者多线程处理的环境、甚至包括分布式处理、服务器集群的实施环境)。
具体的一种实施例如图1所示，本说明书提供的一种页面访问的数据处理方法的一种实施例中，所述方法可以包括：
S0：接收车险的定损结论。
一般的，所述定损结论可以包括识别出的车辆的受损部件的信息，例如定损结论中可以包括受损部件的部件名称、受损程度、受损位置等。用户可以将定损结论输入所述定损决策系统中，例如人工进行定损得到的定损结论。另一些实施场景中，也可以由其他终端设备将定损结论传输给所述定损决策系统，如定损服务器得通过定损图像识别处理得到的定损结论发送给所述定损决策系统，所述定损决策系统可以即时处理或持久化后处理。
S2：基于历史定损结论数据计算出现所述定损结论中的损伤部件组合的概率，所述损伤部件组合包括至少一个损伤部件。
在本实施例损伤部件遗漏的处理场景中，所述的损伤部件组合可以包括一个损伤部件即可实现。如定损结论中的损伤部件组合为雾灯损坏，则通常情况下有较大概率的出现雾灯框也损坏。因此，本实施例中的损伤部件组合可以包括1个损伤部件，这样可以基于后续该损伤部件来雾灯的匹配损伤关联部件雾灯框，得到遗漏的损伤部件。如上述中，可以采用贝叶斯推理方法，结合数据库中存储的历史定损结论数据来计算定损结论中损伤部件组成出现的概率。当然，本说明书不排除其他的实施方式中，可以使用其他统计或归纳或预测算法，或者自定义的算法或模型，并利用历史定损结论数据来得到定损结论中损伤部件组合出现的概率。
S4：确定所述概率大于第一阈值时，查询是否有与所述损伤部件匹配的损伤关联部件。
如果所述损伤部件组合出现的概率大于一定的阈值，则可以表示该损伤部件组合属于常规的损伤部件组合(或者成为高频损伤部件组合)。然后可以结合历史定损结论数 据的分析结果查询是否有与所述损伤部件匹配的损伤关联部件。具体的，可以对历史定损结论数据进行统计得到当某个部件损伤时，同时还会出现另一个损伤部件的信息。或者可以建立训练简单学习模型，以历史定损记录数据作为样本数据进行训练学习，如可以采用CNN((Deep Neural Networks，深度神经网络)、GBDT(GradientBoostingDecisionTree，迭代的决策树算法)、SVM(Support Vector Machine，支持向量机)等。
通常情况下，如果某个部件受损，则与该部件周边的部件也存在受损的可能性。因此，本实施例可以进一步查询是否有与所述受损部件匹配的损伤关联部件。具体的一个实施例中，所述查询是否有与所述损伤部件匹配的损伤关联部件可以包括：
S40：在历史关联规则中查询所述损伤部件的损伤关联部件，所述历史关联规则包括基于历史定损结论数据中当第一损伤部件时出现受损的第二损伤部件确定。
可以根据历史定损结论数据中各当某个部件(可以称为第一部件)受损时，还会伴随着其他部件(可以称为第二部件)受损的信息来生成历史关联规则。例如，当部件A受损时，在一些历史定损结论中，出现还会有部件B也会受损，在另一些历史定损结论中，出现还会有部件C受损而B未受损的情况，当然，可以出现在部件A受损时，部件B和C同样也会受损的历史定损结论。这样，可以根据对历史定损结论数据的处理构建生成历史关联规则，历史关联规则中可以记录有当第一部件受损时，还会发生受损的第二部件的信息，所述第二部件可以为一个或多个。如可以有“部件A受损时，部件B受损”的历史关联规则，也可以有“部件A受损时，部件C受损”的历史关联规则。
在历史定损结论数据中，包含同一个受损部件的不同受损部件组合出现的次数可以不同，则对应着其出现的概率不同。本说明书提供的所述方法的另一个实施例中，某个部件的历史关联规则可以有相应的置信度，所述置信度可以包括基于历史定损结论数据中当所述第一损伤部件时，所述第二损伤部件出现受损的概率确定，在确定损伤部件时可以选取置信度大于阈值的损伤关联部件作为所述匹配的损伤关联部件。
置信度越高，可以表示第一部件损伤时第二部件损伤的概率越高。本实施例中采用置信度对历史关联规则进行中的损伤关联部件进行过滤，选取大于阈值的高置信度作为匹配的损伤关联部件，可以进一步提高识别查找遗漏损伤部件的准确性，提高定损结论的可靠性和准确性。关于选取置信度的阈值，可以结合应用场景进行设置，例如可以设置选择大于90％的置信度对应的损伤关联部件。
具体的一个示例中，例如可以设置第一阈值为0.5％，损伤部件组合中包括损伤部件A和损伤部件B。通过历史定损结论数据计算该损伤部件组合出现的概率为65％，表示损伤部件A与B的组合是常见的组合。查询历史案件关联规则可知，90％情况下，当同时出现损伤部件A和B时，部件C也有损伤。则，此时可以将部件C作为损伤部件A的损伤关联部件，也可以作为损伤部件B的损伤关联部件。
S6：若有，则将所述损伤关联部件作为所述定损结论的遗漏损伤部件。
在一些应用场景中，若查询到损伤关联部件，则可能是用户人为欺诈或者系统自动定损错误。本实施例方案可以将损伤关联部件作为定损结论中出现的遗漏损伤部件，该遗漏损伤部件可以以推送或提示的信息发送给指定接收方进行人工审核。另一些实施例中，可以将改遗漏损伤部件作为定损结论中的损伤部件再次核损，这样，输出的定损结论中可以包括遗漏损伤部件和原始包括的损伤部件。
利用上述实施例提供的实施方案，可以对定损结论进行完善和修正，解决一些场景下输出不合常理的定损结论的问题，有效提高输出的定损结论的精度和可靠性，提高用户体验。例如某个车险定损案件中，只输出了右后翼子板和右底大边损伤，而实际的定损单和大量历史数据中都表明右后翼子板、右底大边和右后门是常见的损伤组合，且当右后翼子板和右底大边同时出现损伤情况下，右后门损伤概率为90％。通过本说明书提供的实施例，可以有效解决这类“不合常理输出”的问题，可以有效提升车险定损输出结论的精度，带来更好的用户体验。
本申请提供的所述方法的另一个应用场景中，可能会出现另一种不符合常规部件组合的情况，如人为欺诈。恶意用户可以通过伪造定损图像、故意非正常角度拍摄，甚至使用不同车辆的定损图像来实现车险定损的非法目的。这种情况下可能会出现一些不符合常规部件组合的情况，如右前门碰撞、右前轮、右前A柱发生损坏，但右侧转向灯确没有损坏，这样的情况下有可能是车险欺诈或者特定角度发生的碰撞。基于此，本说明书提供的所述方法的另一个实施例中，所述损伤部件组合包括至少两个损伤部件，若出现损伤部件组合的概率小于一定的阈值，则可以发出风险提示消息，对该损伤部件组合进行提示，以便进行人工核实或再次识别处理等。因此，本说明书提供的所述方法的另一个实施例中，若所述损伤部件组合包括至少两个损伤部件，则所述方法还可以包括：
S8：确定所述概率低于第二阈值时，发送风险提示消息。
图2是本说明书所述方法另一个实施例流程示意图。通常来说，不符合常规部件组 合不完全排除不可能出现的情况，但整体而言在历史定损结论数据中出现的次数少。
当然，需要说明书的，本说明书的其他实施例中，所述损伤部件组合可以包括一个损伤部件的实施方场景。若该损伤部件组合出现的概率小于一个阈值(在此可以称为第四阈值)，则同样可以发出风险提示消息。如，损伤部件组合为一个车辆内部的部件，如扶手箱。而在绝大数车辆出险情况下，仅仅出现扶手箱的损坏的可能性极低。因此，在一些实施场景中，如果损伤部件组合仅包括一个损伤部件并且该损伤部件组合出现的概率小于第四阈值，则可以发送风险提示消息。因此，所述方法的另一个实施例中，若所述损伤部件组合包括一个损伤部件，且确定所述概率低于第四阈值时，则发送风险提示消息。
本说明书提供的所述方法的一个实施例中，所述基于历史定损结论数据计算出现所述损伤部件组合的概率包括：
若所述定损结论中包括的损伤部件在历史定损结论数据中出现的次数低于第三阈值，则判定所述损伤部件组合出现的概率为0。
例如，损伤部件A和损伤部件M的组合，如果在10000次历史定损结论中进行出现过1次，低于设定的阈值(为了区别不同阈值，这里可以称为第三阈值)，则可以判定当前定损结论中的损伤部件A和损伤部件M的组合出现的概率为0。
如前所述，有些不符合常规部件组合的情况还是有可能出现的，例如在某些特定碰撞角度、或撞击位置、或季节等可能发生极少出现的损伤部件组合情况。因此，本说明书的另一个实施方案中，还可以结合历史定损结论数据在特定条件下的数据信息，如指碰撞角度、碰撞强度、地域，车型，时间(季节)、天气、事故类型等特征，来与当前定损结论的特定条件进行匹配。如果当前定损结论对应的特定条件与历史定损结论数据的特定条件相匹配，则可以表示事故发生的环境(特定条件)相同或类型，则是有较大可能出现不符合常规部件组合的情况。因此，本说明书提供的另一个实施例中，在基于历史定损结论数据计算出现所述定损结论中的损伤部件组合的概率时，还获取所述定损结论对应的特定条件数据，所述特定条件数据至少包括碰撞角度、碰撞强度、事故发生地、事故发生、事故类型中的至少一种数据信息；
相应的，若所述定损结论对应的特定条件数据与所述历史定损结论数据中的特定条件数据相匹配，则确定出现所述定损结论中的损伤部件组的概率大于第一阈值。
如上述扶手箱损坏的实施场景中，特定条件数据可以包括这样的场景描述：天窗打 开、停在地面靠近楼房、高空抛物，在这样的特定条件下是有可能发送仅仅出险扶手箱损坏的情况的。如果历史上出现过这样的场景，并且当前处理的定损结论的特定条件也是这样的场景，部件损坏现场环境相同或类似，则可以输出其概率大于第一阈值，表示当前出现的损伤部件组合在这样特定条件下符合正常的发生概率的。这样，本实施例结合特定条件下的数据信息来对定损结论进行处理，可以进一步提高定损结论的可靠性。本说明书的实施例中，用于判断遗漏或者风险的数据不仅仅可以历史定损单数据，还包括碰撞痕迹等其他数据。
所述方法的另一个实施例中，经过人工审核或者添加遗漏损伤部件后的定损结论数据可以作为新的历史定损结论数据，这样通过不断的数据积累，可以使得历史定损结论数据更加完善可靠，也使得后续的车型定损数据处理结果越来越准确、可靠。具体的，所述方法的另一个实施例中，还可以包括：
S10：获取修正后定损结论，将所述修正后定损结论作为历史定损结论数据，其中，所述修正后定损结论包括：
在所述概率大于第一阈值时，基于所遗漏损伤部件对所述定损结论进行修改得到的第一修正后定损结论；
或者，
在所述概率低于第二阈值时，基于所述风险提示消息对所述定损结论进行审核确认得到的第二修正后定损结论。
修正后定损结论可以包括上述第一或第二修正后定损结论中的任意一个，或者包括两者。
图7是利用本说明书所述方法构建的一种定损决策系统的系统框架示意图，其中虚线部分表示在一些实施例中可以不必包括的部分。本说明书实施例提供一套高效准确的车险定损数据的处理方法，可以输出精度更高的定损结果，并且提供了一套自动化分流可疑定损结论的机制，对定损组合的可疑项目进行风险提示，可以识别存在欺诈嫌疑的案件，同时在算法结论不可靠的情况，人工介入修正结论定损完整性做补充，提升用户体验，降低欺诈风险。
本申请实施例所提供的方法实施例可以在移动终端、计算机终端、服务器或者类似的运算装置中执行。以运行在移动终端上为例，图3是本发明实施例的一种车险定损数据的处理方法的移动终端的硬件结构框图。如图3所示，移动终端10可以包括一个或 多个(图中仅示出一个)处理器102(处理器102可以包括但不限于微处理器MCU或可编程逻辑器件FPGA等的处理装置)、用于存储数据的存储器104、以及用于通信功能的传输模块106。本领域普通技术人员可以理解，图3所示的结构仅为示意，其并不对上述电子装置的结构造成限定。例如，移动终端10还可包括比图7中所示更多或者更少的组件，例如还可以包括其他的处理硬件，或者具有与图3所示不同的配置。
存储器104可用于存储应用软件的软件程序以及模块，如本发明实施例中的搜索方法对应的程序指令/模块，处理器102通过运行存储在存储器104内的软件程序以及模块，从而执行各种功能应用以及数据处理，即实现上述车险定损数据的处理方法。存储器104可包括高速随机存储器，还可包括非易失性存储器，如一个或者多个磁性存储装置、闪存、或者其他非易失性固态存储器。在一些实例中，存储器104可进一步包括相对于处理器102远程设置的存储器，这些远程存储器可以通过网络连接至计算机终端10。上述网络的实例包括但不限于互联网、企业内部网、局域网、移动通信网及其组合。
传输模块106用于经由一个网络接收或者发送数据。上述的网络具体实例可包括计算机终端10的通信供应商提供的无线网络。在一个实例中，传输模块106包括一个网络适配器(Network Interface Controller，NIC)，其可通过基站与其他网络设备相连从而可与互联网进行通讯。在一个实例中，传输模块106可以为射频(Radio Frequency，RF)模块，其用于通过无线方式与互联网进行通讯。
基于上述所述的图像物体定位的方法，本说明书还提供一种展示界面内容的数据处理装置。所述的装置可以包括使用了本说明书实施例所述方法的系统(包括分布式系统)、软件(应用)、模块、组件、服务器、客户端等并结合必要的实施硬件的设备装置。基于同一创新构思，本说明书提供的一种实施例中的处理装置如下面的实施例所述。由于装置解决问题的实现方案与方法相似，因此本说明书实施例具体的处理装置的实施可以参见前述方法的实施，重复之处不再赘述。尽管以下实施例所描述的装置较佳地以软件来实现，但是硬件，或者软件和硬件的组合的实现也是可能并被构想的。具体的，如图4所示，图4是本说明书提供的一种展示界面内容的数据处理装置实施例的模块结构示意图，可以包括：
接收模块101，可以用于接收车险的定损结论；
概率计算模块102，可以用于基于历史定损结论数据计算出现所述定损结论中的损伤部件组合的概率，所述损伤部件组合包括至少一个损伤部件；
关联部件确定模块103，可以用于确定所述概率大于第一阈值时，查询是否有与所述损伤部件匹配的损伤关联部件；
第一输出模块104，可以用于查询到匹配的损伤关联部件时，将所述损伤关联部件作为所述定损结论的遗漏损伤部件。
图5是本说明书提供的所述装置另一种实施例的模块结构示意图，如图5所示，所述装置还可以包括：
第二输出模块104，可以用于概率计算模块102得到所述概率低于第二阈值时，发送风险提示消息。
所述装置另一种实施例中，所述概率计算模块102可以包括：
贝叶斯推理单元，可以用于采用贝叶斯推理方法，基于历史定损结论数据中损伤部件出现的先验概率与条件概率，计算所述损伤部件组合的概率。
所述装置另一种实施例中，概率计算模块102基于历史定损结论数据计算出现所述损伤部件组合的概率包括：
若所述定损结论中包括的损伤部件在历史定损结论数据中出现的次数低于第三阈值，则判定所述损伤部件组合出现的概率为0。
所述装置另一种实施例中，在基于历史定损结论数据计算出现所述定损结论中的损伤部件组合的概率时，还获取所述定损结论对应的特定条件数据，所述特定条件数据至少包括碰撞角度、碰撞强度、事故发生地、事故发生、事故类型中的至少一种数据信息；
相应的，若所述定损结论对应的特定条件数据与所述历史定损结论数据中的特定条件数据相匹配，则确定出现所述定损结论中的损伤部件组的概率大于第一阈值。
所述装置另一种实施例中，关联部件确定模块103查询是否有与所述损伤部件匹配的损伤关联部件包括：
在历史关联规则中查询所述损伤部件的损伤关联部件，所述历史关联规则包括基于历史定损结论数据中当第一损伤部件时出现受损的第二损伤部件确定。
所述装置另一种实施例中，关联部件确定模块103还可以包括：
筛选单元，可以用于选取置信度大于阈值的损伤关联部件作为所述匹配的损伤关联部件，所述置信度包括基于历史定损结论数据中当所述第一损伤部件时，所述第二损伤部件出现受损的概率确定。
图6是本说明书提供的所述装置另一种实施例的模块结构示意图，如图6所示，所述装置还可以包括：
历史数据更新模块106，可以获取修正后定损结论，将所述修正后定损结论作为历史定损结论数据，其中，所述修正后定损结论包括：
在所述概率大于第一阈值时，基于所遗漏损伤部件对所述定损结论进行修改得到的第一修正后定损结论；
或者，
在所述概率低于第二阈值时，基于所述风险提示消息对所述定损结论进行审核确认得到的第二修正后定损结论。
需要说明的是，本说明书实施例上述所述的处理装置，具体的实现方式可以参照方法实施例的描述，在此不作一一赘述。
本说明书实施例提供的展示界面内容的数据处理方法可以在计算机中由处理器执行相应的程序指令来实现，如使用windows操作系统的c++语言在PC端实现，或其他例如Linux、android、iOS系统相对应的应用设计语言集合必要的硬件实现，或者基于量子计算机的处理逻辑实现等。具体的，本说明书提供的一种处理设备的一种实施例中，所述处理设备可以包括处理器以及用于存储处理器可执行指令的存储器，所述处理器执行所述指令时实现：
接收车险的定损结论；
基于历史定损结论数据计算出现所述定损结论中的损伤部件组合的概率，所述损伤部件组合包括至少一个损伤部件；
确定所述概率大于第一阈值时，查询是否有与所述损伤部件匹配的损伤关联部件；
若有，则将所述损伤关联部件作为所述定损结论的遗漏损伤部件。
上述的指令可以存储在多种计算机可读存储介质中。所述计算机可读存储介质可以包括用于存储信息的物理装置，可以将信息数字化后再以利用电、磁或者光学等方式的媒体加以存储。本实施例所述的计算机可读存储介质有可以包括：利用电能方式存储信息的装置如，各式存储器，如RAM、ROM等；利用磁能方式存储信息的装置如，硬盘、软盘、磁带、磁芯存储器、磁泡存储器、U盘；利用光学方式存储信息的装置如， CD或DVD。当然，还有其他方式的可读存储介质，例如量子存储器、石墨烯存储器等等。
基于前述所述，本说明书实施例还提供一种车险定损数据的处理设备，所述的处理设备可以包括移动终端、个人掌声电脑、智能穿戴设备、车机交互设备、个人电脑、服务器、服务器集群等。所述处理设备可以包括至少一个处理器以及用于存储处理器可执行指令的存储器，所述处理器执行所述指令时实现：
接收车险的定损结论；
基于历史定损结论数据计算出现所述定损结论中的损伤部件组合的概率，所述损伤部件组合包括至少一个损伤部件；
确定所述概率大于第一阈值时，查询是否有与所述损伤部件匹配的损伤关联部件，若有，则将所述损伤关联部件作为所述定损结论的遗漏损伤部件；
确定所述概率低于第二阈值时，发送风险提示消息。
需要说明的是，本说明书实施例上述所述的处理装置、电子设备，根据相关方法实施例的描述还可以包括其他的实施方式，例如。具体的实现方式可以参照方法实施例的描述，在此不作一一赘述。
本说明书中的各个实施例均采用递进的方式描述，各个实施例之间相同相似的部分互相参见即可，每个实施例重点说明的都是与其他实施例的不同之处。尤其，对于硬件+程序类实施例而言，由于其基本相似于方法实施例，所以描述的比较简单，相关之处参见方法实施例的部分说明即可。
上述对本说明书特定实施例进行了描述。其它实施例在所附权利要求书的范围内。在一些情况下，在权利要求书中记载的动作或步骤可以按照不同于实施例中的顺序来执行并且仍然可以实现期望的结果。另外，在附图中描绘的过程不一定要求示出的特定顺序或者连续顺序才能实现期望的结果。在某些实施方式中，多任务处理和并行处理也是可以的或者可能是有利的。
需要说明的，上述所述的计算机可读存储介质根据方法或装置实施例的描述还可以包括其他的实施方式，具体的实现方式可以参照方法实施例的描述，在此不作一一赘述。
本说明书实施例提供的一种车险定损数据的处理方法、装置和处理设备，可以 结合历史定损结论数据中损伤部件组合的案件信息来计算得到出现定损结论中损伤部件组合的概率，该概率可以表示定损结论的可靠性。若概率大于一定的阈值，可以表示定损结论中的损伤部件组合是常见的损伤组合(也可以称为高频损伤组合)，为正常的部件组合出现概率。在本说明书提供实施方案中，如果某个部件受损，则可以查看与该部件相关联的部件是否也出现受损，如果有，则可以进行损伤遗漏部件的推荐，对定损结论进行补充或修正，解决一些场景下输出不合常理的定损结论的问题，有效提高输出的定损结论的精度和可靠性，提高用户体验。
虽然本申请提供了如实施例或流程图所述的方法操作步骤，但基于常规或者无创造性的劳动可以包括更多或者更少的操作步骤。实施例中列举的步骤顺序仅仅为众多步骤执行顺序中的一种方式，不代表唯一的执行顺序。在实际中的装置或客户端产品执行时，可以按照实施例或者附图所示的方法顺序执行或者并行执行(例如并行处理器或者多线程处理的环境)。
尽管本说明书实施例内容中提到采用贝叶斯推理计算概率、DNN作为学习模型、多个阈值的设置等之类的数据获取、定义、交互、计算、判断等操作和数据描述，但是，本说明书实施例并不局限于必须是符合行业通信标准、标准计算机数据处理协议、通信协议和标准数据模型/模板或本说明书实施例所描述的情况。某些行业标准或者使用自定义方式或实施例描述的实施基础上略加修改后的实施方案也可以实现上述实施例相同、等同或相近、或变形后可预料的实施效果。应用这些修改或变形后的数据获取、存储、判断、处理方式等获取的实施例，仍然可以属于本说明书的可选实施方案范围之内。
在20世纪90年代，对于一个技术的改进可以很明显地区分是硬件上的改进(例如，对二极管、晶体管、开关等电路结构的改进)还是软件上的改进(对于方法流程的改进)。然而，随着技术的发展，当今的很多方法流程的改进已经可以视为硬件电路结构的直接改进。设计人员几乎都通过将改进的方法流程编程到硬件电路中来得到相应的硬件电路结构。因此，不能说一个方法流程的改进就不能用硬件实体模块来实现。例如，可编程逻辑器件(Programmable Logic Device,PLD)(例如现场可编程门阵列(Field Programmable Gate Array，FPGA))就是这样一种集成电路，其逻辑功能由用户对器件编程来确定。由设计人员自行编程来把一个数字系统“集成”在一片PLD上，而不需要请芯片制造厂商来设计和制作专用的集成电路芯片。而且，如今，取代手工地制作集成电路芯片，这种编程也多半改用“逻辑编译器(logic compiler)”软件来实现，它与程序开发撰写时所用的软件编译器相类似，而要编译之前的原始代码也得用特定的编程 语言来撰写，此称之为硬件描述语言(Hardware Description Language，HDL)，而HDL也并非仅有一种，而是有许多种，如ABEL(Advanced Boolean Expression Language)、AHDL(Altera Hardware Description Language)、Confluence、CUPL(Cornell University Programming Language)、HDCal、JHDL(Java Hardware Description Language)、Lava、Lola、MyHDL、PALASM、RHDL(Ruby Hardware Description Language)等，目前最普遍使用的是VHDL(Very-High-Speed Integrated Circuit Hardware Description Language)与Verilog。本领域技术人员也应该清楚，只需要将方法流程用上述几种硬件描述语言稍作逻辑编程并编程到集成电路中，就可以很容易得到实现该逻辑方法流程的硬件电路。
控制器可以按任何适当的方式实现，例如，控制器可以采取例如微处理器或处理器以及存储可由该(微)处理器执行的计算机可读程序代码(例如软件或固件)的计算机可读介质、逻辑门、开关、专用集成电路(Application Specific Integrated Circuit，ASIC)、可编程逻辑控制器和嵌入微控制器的形式，控制器的例子包括但不限于以下微控制器：ARC 625D、Atmel AT91SAM、Microchip PIC18F26K20以及Silicone Labs C8051F320，存储器控制器还可以被实现为存储器的控制逻辑的一部分。本领域技术人员也知道，除了以纯计算机可读程序代码方式实现控制器以外，完全可以通过将方法步骤进行逻辑编程来使得控制器以逻辑门、开关、专用集成电路、可编程逻辑控制器和嵌入微控制器等的形式来实现相同功能。因此这种控制器可以被认为是一种硬件部件，而对其内包括的用于实现各种功能的装置也可以视为硬件部件内的结构。或者甚至，可以将用于实现各种功能的装置视为既可以是实现方法的软件模块又可以是硬件部件内的结构。
上述实施例阐明的系统、装置、模块或单元，具体可以由计算机芯片或实体实现，或者由具有某种功能的产品来实现。一种典型的实现设备为计算机。具体的，计算机例如可以为个人计算机、膝上型计算机、车载人机交互设备、蜂窝电话、相机电话、智能电话、个人数字助理、媒体播放器、导航设备、电子邮件设备、游戏控制台、平板计算机、可穿戴设备或者这些设备中的任何设备的组合。
虽然本说明书实施例提供了如实施例或流程图所述的方法操作步骤，但基于常规或者无创造性的手段可以包括更多或者更少的操作步骤。实施例中列举的步骤顺序仅仅为众多步骤执行顺序中的一种方式，不代表唯一的执行顺序。在实际中的装置或终端产品执行时，可以按照实施例或者附图所示的方法顺序执行或者并行执行(例如并行处 理器或者多线程处理的环境，甚至为分布式数据处理环境)。术语“包括”、“包含”或者其任何其他变体意在涵盖非排他性的包含，从而使得包括一系列要素的过程、方法、产品或者设备不仅包括那些要素，而且还包括没有明确列出的其他要素，或者是还包括为这种过程、方法、产品或者设备所固有的要素。在没有更多限制的情况下，并不排除在包括所述要素的过程、方法、产品或者设备中还存在另外的相同或等同要素。
为了描述的方便，描述以上装置时以功能分为各种模块分别描述。当然，在实施本说明书实施例时可以把各模块的功能在同一个或多个软件和/或硬件中实现，也可以将实现同一功能的模块由多个子模块或子单元的组合实现等。以上所描述的装置实施例仅仅是示意性的，例如，所述单元的划分，仅仅为一种逻辑功能划分，实际实现时可以有另外的划分方式，例如多个单元或组件可以结合或者可以集成到另一个系统，或一些特征可以忽略，或不执行。另一点，所显示或讨论的相互之间的耦合或直接耦合或通信连接可以是通过一些接口，装置或单元的间接耦合或通信连接，可以是电性，机械或其它的形式。
本领域技术人员也知道，除了以纯计算机可读程序代码方式实现控制器以外，完全可以通过将方法步骤进行逻辑编程来使得控制器以逻辑门、开关、专用集成电路、可编程逻辑控制器和嵌入微控制器等的形式来实现相同功能。因此这种控制器可以被认为是一种硬件部件，而对其内部包括的用于实现各种功能的装置也可以视为硬件部件内的结构。或者甚至，可以将用于实现各种功能的装置视为既可以是实现方法的软件模块又可以是硬件部件内的结构。
本发明是参照根据本发明实施例的方法、设备(系统)、和计算机程序产品的流程图和/或方框图来描述的。应理解可由计算机程序指令实现流程图和/或方框图中的每一流程和/或方框、以及流程图和/或方框图中的流程和/或方框的结合。可提供这些计算机程序指令到通用计算机、专用计算机、嵌入式处理机或其他可编程数据处理设备的处理器以产生一个机器，使得通过计算机或其他可编程数据处理设备的处理器执行的指令产生用于实现在流程图一个流程或多个流程和/或方框图一个方框或多个方框中指定的功能的装置。
这些计算机程序指令也可存储在能引导计算机或其他可编程数据处理设备以特定方式工作的计算机可读存储器中，使得存储在该计算机可读存储器中的指令产生包括指令装置的制造品，该指令装置实现在流程图一个流程或多个流程和/或方框图一个方框或多个方框中指定的功能。
这些计算机程序指令也可装载到计算机或其他可编程数据处理设备上，使得在计算机或其他可编程设备上执行一系列操作步骤以产生计算机实现的处理，从而在计算机或其他可编程设备上执行的指令提供用于实现在流程图一个流程或多个流程和/或方框图一个方框或多个方框中指定的功能的步骤。
在一个典型的配置中，计算设备包括一个或多个处理器(CPU)、输入/输出接口、网络接口和内存。
内存可能包括计算机可读介质中的非永久性存储器，随机存取存储器(RAM)和/或非易失性内存等形式，如只读存储器(ROM)或闪存(flash RAM)。内存是计算机可读介质的示例。
计算机可读介质包括永久性和非永久性、可移动和非可移动媒体可以由任何方法或技术来实现信息存储。信息可以是计算机可读指令、数据结构、程序的模块或其他数据。计算机的存储介质的例子包括，但不限于相变内存(PRAM)、静态随机存取存储器(SRAM)、动态随机存取存储器(DRAM)、其他类型的随机存取存储器(RAM)、只读存储器(ROM)、电可擦除可编程只读存储器(EEPROM)、快闪记忆体或其他内存技术、只读光盘只读存储器(CD-ROM)、数字多功能光盘(DVD)或其他光学存储、磁盒式磁带，磁带磁磁盘存储或其他磁性存储设备或任何其他非传输介质，可用于存储可以被计算设备访问的信息。按照本文中的界定，计算机可读介质不包括暂存电脑可读媒体(transitory media)，如调制的数据信号和载波。
本领域技术人员应明白，本说明书的实施例可提供为方法、系统或计算机程序产品。因此，本说明书实施例可采用完全硬件实施例、完全软件实施例或结合软件和硬件方面的实施例的形式。而且，本说明书实施例可采用在一个或多个其中包含有计算机可用程序代码的计算机可用存储介质(包括但不限于磁盘存储器、CD-ROM、光学存储器等)上实施的计算机程序产品的形式。
本说明书实施例可以在由计算机执行的计算机可执行指令的一般上下文中描述，例如程序模块。一般地，程序模块包括执行特定任务或实现特定抽象数据类型的例程、程序、对象、组件、数据结构等等。也可以在分布式计算环境中实践本说明书实施例，在这些分布式计算环境中，由通过通信网络而被连接的远程处理设备来执行任务。在分布式计算环境中，程序模块可以位于包括存储设备在内的本地和远程计算机存储介质中。
本说明书中的各个实施例均采用递进的方式描述，各个实施例之间相同相似的部分互相参见即可，每个实施例重点说明的都是与其他实施例的不同之处。尤其，对于系统实施例而言，由于其基本相似于方法实施例，所以描述的比较简单，相关之处参见方法实施例的部分说明即可。在本说明书的描述中，参考术语“一个实施例”、“一些实施例”、“示例”、“具体示例”、或“一些示例”等的描述意指结合该实施例或示例描述的具体特征、结构、材料或者特点包含于本说明书实施例的至少一个实施例或示例中。在本说明书中，对上述术语的示意性表述不必须针对的是相同的实施例或示例。而且，描述的具体特征、结构、材料或者特点可以在任一个或多个实施例或示例中以合适的方式结合。此外，在不相互矛盾的情况下，本领域的技术人员可以将本说明书中描述的不同实施例或示例以及不同实施例或示例的特征进行结合和组合。
以上所述仅为本说明书实施例的实施例而已，并不用于限制本说明书实施例。对于本领域技术人员来说，本说明书实施例可以有各种更改和变化。凡在本说明书实施例的精神和原理之内所作的任何修改、等同替换、改进等，均应包含在本说明书实施例的权利要求范围之内。

Claims (19)

  1. 一种车险定损数据的处理方法，所述方法包括：
    接收车险的定损结论；
    基于历史定损结论数据计算出现所述定损结论中的损伤部件组合的概率，所述损伤部件组合包括至少一个损伤部件；
    确定所述概率大于第一阈值时，查询是否有与所述损伤部件匹配的损伤关联部件；
    若有，则将所述损伤关联部件作为所述定损结论的遗漏损伤部件。
  2. 如权利要求1所述的方法，若所述损伤部件组合包括至少两个损伤部件，则所述方法还包括：
    确定所述概率低于第二阈值时，发送风险提示消息。
  3. 如权利要求1或2所述的方法，所述基于历史定损结论数据计算出现所述损伤部件组合的概率包括：
    采用贝叶斯推理方法，基于历史定损结论数据中损伤部件出现的先验概率与条件概率，计算所述损伤部件组合的概率。
  4. 如权利要求1或2所述的方法，所述基于历史定损结论数据计算出现所述损伤部件组合的概率包括：
    若所述定损结论中包括的损伤部件在历史定损结论数据中出现的次数低于第三阈值，则判定所述损伤部件组合出现的概率为0。
  5. 如权利要求4所述的方法，在基于历史定损结论数据计算出现所述定损结论中的损伤部件组合的概率时，还获取所述定损结论对应的特定条件数据，所述特定条件数据至少包括碰撞角度、碰撞强度、事故发生地、事故发生、事故类型中的至少一种数据信息；
    相应的，若所述定损结论对应的特定条件数据与所述历史定损结论数据中的特定条件数据相匹配，则确定出现所述定损结论中的损伤部件组的概率大于第一阈值。
  6. 如权利要求1所述的方法，所述查询是否有与所述损伤部件匹配的损伤关联部件包括：
    在历史关联规则中查询所述损伤部件的损伤关联部件，所述历史关联规则包括基于历史定损结论数据中当第一损伤部件时出现受损的第二损伤部件确定。
  7. 如权利要求6所述的方法，所述方法还包括：
    选取置信度大于阈值的损伤关联部件作为所述匹配的损伤关联部件，所述置信度包括基于历史定损结论数据中当所述第一损伤部件时，所述第二损伤部件出现受损的概率 确定。
  8. 如权利要求1或2所述的方法，所述方法还包括：
    获取修正后定损结论，将所述修正后定损结论作为历史定损结论数据，其中，所述修正后定损结论包括：
    在所述概率大于第一阈值时，基于所遗漏损伤部件对所述定损结论进行修改得到的第一修正后定损结论；
    或者，
    在所述概率低于第二阈值时，基于风险提示消息对所述定损结论进行审核确认得到的第二修正后定损结论。
  9. 一种展示界面内容的数据处理装置，包括：
    接收模块，用于接收车险的定损结论；
    概率计算模块，用于基于历史定损结论数据计算出现所述定损结论中的损伤部件组合的概率，所述损伤部件组合包括至少一个损伤部件；
    关联部件确定模块，用于确定所述概率大于第一阈值时，查询是否有与所述损伤部件匹配的损伤关联部件；
    第一输出模块，用于查询到匹配的损伤关联部件时，将所述损伤关联部件作为所述定损结论的遗漏损伤部件。
  10. 如权利要求9所述的装置，所述装置还包括：
    第二输出模块，用于概率计算模块得到所述概率低于第二阈值时，发送风险提示消息。
  11. 如权利要求9或10所述的装置，所述概率计算模块包括：
    贝叶斯推理单元，用于采用贝叶斯推理方法，基于历史定损结论数据中损伤部件出现的先验概率与条件概率，计算所述损伤部件组合的概率。
  12. 如权利要求9或10所述的装置，所述概率计算模块基于历史定损结论数据计算出现所述损伤部件组合的概率包括：
    若所述定损结论中包括的损伤部件在历史定损结论数据中出现的次数低于第三阈值，则判定所述损伤部件组合出现的概率为0。
  13. 如权利要求12所述的装置，概率计算模块在基于历史定损结论数据计算出现所述定损结论中的损伤部件组合的概率时，还获取所述定损结论对应的特定条件数据，所述特定条件数据至少包括碰撞角度、碰撞强度、事故发生地、事故发生、事故类型中的至少一种数据信息；以及，
    相应的，若所述定损结论对应的特定条件数据与所述历史定损结论数据中的特定条件数据相匹配，则确定出现所述定损结论中的损伤部件组的概率大于第一阈值。
  14. 如权利要求9所述的装置，所述关联部件确定模块查询是否有与所述损伤部件匹配的损伤关联部件包括：
    在历史关联规则中查询所述损伤部件的损伤关联部件，所述历史关联规则包括基于历史定损结论数据中当第一损伤部件时出现受损的第二损伤部件确定。
  15. 如权利要求14所述的装置，所述关联部件确定模块还包括：
    筛选单元，用于选取置信度大于阈值的损伤关联部件作为所述匹配的损伤关联部件，所述置信度包括基于历史定损结论数据中当所述第一损伤部件时，所述第二损伤部件出现受损的概率确定。
  16. 如权利要求9或10所述的装置，所述装置还包括：
    历史数据更新模块，获取修正后定损结论，将所述修正后定损结论作为历史定损结论数据，其中，所述修正后定损结论包括：
    在所述概率大于第一阈值时，基于所遗漏损伤部件对所述定损结论进行修改得到的第一修正后定损结论；
    或者，
    在所述概率低于第二阈值时，基于风险提示消息对所述定损结论进行审核确认得到的第二修正后定损结论。
  17. 一种处理设备，包括处理器以及用于存储处理器可执行指令的存储器，所述处理器执行所述指令时实现：
    接收车险的定损结论；
    基于历史定损结论数据计算出现所述定损结论中的损伤部件组合的概率，所述损伤部件组合包括至少一个损伤部件；
    确定所述概率大于第一阈值时，查询是否有与所述损伤部件匹配的损伤关联部件；
    若有，则将所述损伤关联部件作为所述定损结论的遗漏损伤部件。
  18. 一种电子设备，包括至少一个处理器以及用于存储处理器可执行指令的存储器，所述处理器执行所述指令时实现：
    接收车险的定损结论；
    基于历史定损结论数据计算出现所述定损结论中的损伤部件组合的概率，所述损伤部件组合包括至少一个损伤部件；
    确定所述概率大于第一阈值时，查询是否有与所述损伤部件匹配的损伤关联部件， 若有，则将所述损伤关联部件作为所述定损结论的遗漏损伤部件；
    确定所述概率低于第二阈值时，发送风险提示消息。
  19. 如权利要求1所述的方法，其中，若所述损伤部件组合包括一个损伤部件，且确定所述概率低于第四阈值时，发送风险提示消息。
PCT/CN2018/099998 2017-11-21 2018-08-10 一种车险定损数据的处理方法、装置和处理设备 WO2019100763A1 (zh)

Priority Applications (3)

Application Number Priority Date Filing Date Title
SG11202004728XA SG11202004728XA (en) 2017-11-21 2018-08-10 Method and apparatus for processing loss assessment data for car insurance and processing device
EP18880390.2A EP3716194A4 (en) 2017-11-21 2018-08-10 METHOD AND DEVICE FOR PROCESSING VEHICLE LOSS EVALUATION DATA AND PROCESSING DEVICE
US16/880,211 US20200334638A1 (en) 2017-11-21 2020-05-21 Method and apparatus for processing loss assessment data for car insurance and processing device

Applications Claiming Priority (2)

Application Number Priority Date Filing Date Title
CN201711166508.9A CN108062712B (zh) 2017-11-21 2017-11-21 一种车险定损数据的处理方法、装置和处理设备
CN201711166508.9 2017-11-21

Related Child Applications (1)

Application Number Title Priority Date Filing Date
US16/880,211 Continuation US20200334638A1 (en) 2017-11-21 2020-05-21 Method and apparatus for processing loss assessment data for car insurance and processing device

Publications (1)

Publication Number Publication Date
WO2019100763A1 true WO2019100763A1 (zh) 2019-05-31

Family

ID=62134906

Family Applications (1)

Application Number Title Priority Date Filing Date
PCT/CN2018/099998 WO2019100763A1 (zh) 2017-11-21 2018-08-10 一种车险定损数据的处理方法、装置和处理设备

Country Status (6)

Country Link
US (1) US20200334638A1 (zh)
EP (1) EP3716194A4 (zh)
CN (1) CN108062712B (zh)
SG (1) SG11202004728XA (zh)
TW (1) TWI686759B (zh)
WO (1) WO2019100763A1 (zh)

Families Citing this family (11)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN108062712B (zh) * 2017-11-21 2020-11-06 创新先进技术有限公司 一种车险定损数据的处理方法、装置和处理设备
CN110569701B (zh) * 2018-12-29 2020-08-07 阿里巴巴集团控股有限公司 计算机执行的车辆定损方法及装置
CN109920158A (zh) * 2019-04-23 2019-06-21 许灵辉 租车管理系统及方法
CN110443718B (zh) * 2019-07-17 2022-11-04 卓尔智联(武汉)研究院有限公司 车险理赔方法、计算机装置及可读存储介质
US11164260B1 (en) 2019-09-10 2021-11-02 MagMutual Intermediate Holding Company Systems and methods for simulating and visualizing loss data
US11435884B1 (en) * 2019-09-10 2022-09-06 MagMutual Intermediate Holding Company Impactor, impactor mitigator, and entity structure graphical object visualization system and corresponding methods
CN111311175A (zh) * 2020-01-20 2020-06-19 深圳壹账通智能科技有限公司 评残条款智能预测方法、装置、计算机设备及存储介质
CN111461266B (zh) * 2020-06-18 2020-12-01 爱保科技有限公司 一种车辆定损异常识别方法、装置、服务器以及存储介质
CN112070250B (zh) * 2020-11-13 2021-05-04 深圳壹账通智能科技有限公司 车辆定损方法、装置、终端设备及存储介质
CN112464164A (zh) * 2020-11-25 2021-03-09 北京航空航天大学 一种人因可靠性评估的方法、装置和信息处理设备
CN114842205B (zh) * 2022-05-30 2024-05-07 平安科技（深圳）有限公司 车损检测方法、装置、设备及存储介质

Citations (5)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN102180170A (zh) * 2011-04-22 2011-09-14 林力 车辆数据采集存储分析显示装置
CN103163877A (zh) * 2011-12-19 2013-06-19 通用汽车环球科技运作有限责任公司 用于系统级故障的根本原因分析和质量监控的方法和系统
CN103493051A (zh) * 2011-02-25 2014-01-01 奥达特克斯有限责任公司 用于估计对汽车的碰撞损伤的系统和方法
CN106776681A (zh) * 2016-11-04 2017-05-31 中国平安财产保险股份有限公司 一种车险配件数据库中配件数据的维护方法和系统
CN108062712A (zh) * 2017-11-21 2018-05-22 阿里巴巴集团控股有限公司 一种车险定损数据的处理方法、装置和处理设备

Family Cites Families (5)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
JP4640475B2 (ja) * 2008-09-11 2011-03-02 トヨタ自動車株式会社 車両の修理交換情報管理システム、車両の修理交換情報管理装置、車両の異常原因情報管理システム、車両の異常原因情報管理装置、複数組の教師データの処理方法
CN104123312B (zh) * 2013-04-28 2018-02-16 国际商业机器公司 一种数据挖掘方法和装置
CN105719188B (zh) * 2016-01-22 2017-12-26 平安科技（深圳）有限公司 基于多张图片一致性实现保险理赔反欺诈的方法及服务器
US10692050B2 (en) * 2016-04-06 2020-06-23 American International Group, Inc. Automatic assessment of damage and repair costs in vehicles
CN106600422A (zh) * 2016-11-24 2017-04-26 中国平安财产保险股份有限公司 一种车险智能定损方法和系统

Patent Citations (5)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN103493051A (zh) * 2011-02-25 2014-01-01 奥达特克斯有限责任公司 用于估计对汽车的碰撞损伤的系统和方法
CN102180170A (zh) * 2011-04-22 2011-09-14 林力 车辆数据采集存储分析显示装置
CN103163877A (zh) * 2011-12-19 2013-06-19 通用汽车环球科技运作有限责任公司 用于系统级故障的根本原因分析和质量监控的方法和系统
CN106776681A (zh) * 2016-11-04 2017-05-31 中国平安财产保险股份有限公司 一种车险配件数据库中配件数据的维护方法和系统
CN108062712A (zh) * 2017-11-21 2018-05-22 阿里巴巴集团控股有限公司 一种车险定损数据的处理方法、装置和处理设备

Non-Patent Citations (1)

* Cited by examiner, † Cited by third party
Title
See also references of EP3716194A4 *

Also Published As

Publication number Publication date
CN108062712B (zh) 2020-11-06
EP3716194A4 (en) 2021-12-01
SG11202004728XA (en) 2020-06-29
CN108062712A (zh) 2018-05-22
EP3716194A1 (en) 2020-09-30
US20200334638A1 (en) 2020-10-22
TW201928846A (zh) 2019-07-16
TWI686759B (zh) 2020-03-01

Similar Documents

Publication Publication Date Title
WO2019100763A1 (zh) 一种车险定损数据的处理方法、装置和处理设备
US10817732B2 (en) Automated assessment of collision risk based on computer vision
CN111133447B (zh) 适于自主驾驶的对象检测和检测置信度的方法和系统
US11475770B2 (en) Electronic device, warning message providing method therefor, and non-transitory computer-readable recording medium
US10395377B2 (en) Systems and methods for non-obstacle area detection
JP6394735B2 (ja) 階層型コンテキストアウェアによる手足の検出
US10373024B2 (en) Image processing device, object detection device, image processing method
US10311704B1 (en) Passenger-related item loss mitigation
CN106952303B (zh) 车距检测方法、装置和系统
WO2020042984A1 (zh) 一种车辆行为检测方法及装置
WO2019214320A1 (zh) 车辆损伤识别的处理方法、处理设备、客户端及服务器
US20210018590A1 (en) Perception system error detection and re-verification
US11904853B2 (en) Apparatus for preventing vehicle collision and method thereof
US11250279B2 (en) Generative adversarial network models for small roadway object detection
US20180362037A1 (en) Accident probability calculator, accident probability calculation method, and accident probability calculation program
KR20170106823A (ko) 부분적인 깊이 맵에 기초하여 관심 객체를 식별하는 영상 처리 장치
CN110852261B (zh) 目标检测方法、装置、电子设备和可读存储介质
Velez et al. Embedded platforms for computer vision-based advanced driver assistance systems: a survey
US20220340170A1 (en) Driving risk identification method, storage medium and electronic device
EP3989031A1 (en) Systems and methods for fusing road friction data to enhance vehicle maneuvering
US11328601B1 (en) Prevention of low-speed sideswipe collisions with non-moving objects
CN114708498A (zh) 图像处理方法、装置、电子设备以及存储介质
CN113822146A (zh) 目标检测方法、终端设备及计算机存储介质
US20230386222A1 (en) Method for detecting three-dimensional objects in roadway and electronic device
CN116872840A (zh) 车辆防撞预警方法、装置、车辆和存储介质

Legal Events

Date Code Title Description
121 Ep: the epo has been informed by wipo that ep was designated in this application

Ref document number: 18880390

Country of ref document: EP

Kind code of ref document: A1

NENP Non-entry into the national phase

Ref country code: DE

ENP Entry into the national phase

Ref document number: 2018880390

Country of ref document: EP

Effective date: 20200622