CN111886619B - Vehicle collision damage assessment method and system based on historical cases - Google Patents
Vehicle collision damage assessment method and system based on historical cases Download PDFInfo
- Publication number
- CN111886619B CN111886619B CN201980021736.8A CN201980021736A CN111886619B CN 111886619 B CN111886619 B CN 111886619B CN 201980021736 A CN201980021736 A CN 201980021736A CN 111886619 B CN111886619 B CN 111886619B
- Authority
- CN
- China
- Prior art keywords
- vehicle
- damage
- impairment
- assessment
- historical
- Prior art date
- Legal status (The legal status 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 status listed.)
- Active
Links
- 238000000034 method Methods 0.000 title claims abstract description 80
- 238000010801 machine learning Methods 0.000 claims abstract description 14
- 230000006735 deficit Effects 0.000 claims description 133
- 238000012423 maintenance Methods 0.000 claims description 23
- 238000004364 calculation method Methods 0.000 claims description 19
- 230000008569 process Effects 0.000 claims description 14
- 230000008859 change Effects 0.000 claims description 12
- 239000003973 paint Substances 0.000 claims description 8
- 230000002776 aggregation Effects 0.000 claims description 4
- 238000004220 aggregation Methods 0.000 claims description 4
- 238000009826 distribution Methods 0.000 claims description 4
- 238000005507 spraying Methods 0.000 claims description 3
- 238000004458 analytical method Methods 0.000 claims description 2
- 238000005516 engineering process Methods 0.000 abstract description 5
- 230000008439 repair process Effects 0.000 description 43
- 230000001771 impaired effect Effects 0.000 description 13
- 238000010586 diagram Methods 0.000 description 7
- 238000007781 pre-processing Methods 0.000 description 7
- 238000013528 artificial neural network Methods 0.000 description 5
- 238000004519 manufacturing process Methods 0.000 description 5
- 230000006870 function Effects 0.000 description 4
- 230000007246 mechanism Effects 0.000 description 4
- 231100000809 Damage Assessment Model Toxicity 0.000 description 3
- 230000005540 biological transmission Effects 0.000 description 3
- 238000013136 deep learning model Methods 0.000 description 3
- 238000012545 processing Methods 0.000 description 3
- 238000003860 storage Methods 0.000 description 3
- 238000013473 artificial intelligence Methods 0.000 description 2
- 238000006073 displacement reaction Methods 0.000 description 2
- 239000000463 material Substances 0.000 description 2
- 238000012360 testing method Methods 0.000 description 2
- 238000012549 training Methods 0.000 description 2
- 230000006978 adaptation Effects 0.000 description 1
- 230000004931 aggregating effect Effects 0.000 description 1
- 230000008485 antagonism Effects 0.000 description 1
- 238000004140 cleaning Methods 0.000 description 1
- 238000004891 communication Methods 0.000 description 1
- 230000001010 compromised effect Effects 0.000 description 1
- 238000010276 construction Methods 0.000 description 1
- 238000007405 data analysis Methods 0.000 description 1
- 238000013135 deep learning Methods 0.000 description 1
- 238000011156 evaluation Methods 0.000 description 1
- 230000007774 longterm Effects 0.000 description 1
- 239000002184 metal Substances 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 230000003287 optical effect Effects 0.000 description 1
- 239000003208 petroleum Substances 0.000 description 1
- 230000002787 reinforcement Effects 0.000 description 1
- 239000004065 semiconductor Substances 0.000 description 1
- 238000000638 solvent extraction Methods 0.000 description 1
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
- G06Q40/00—Finance; Insurance; Tax strategies; Processing of corporate or income taxes
- G06Q40/08—Insurance
Landscapes
- Business, Economics & Management (AREA)
- Accounting & Taxation (AREA)
- Finance (AREA)
- Engineering & Computer Science (AREA)
- Development Economics (AREA)
- Economics (AREA)
- Marketing (AREA)
- Strategic Management (AREA)
- Technology Law (AREA)
- Physics & Mathematics (AREA)
- General Business, Economics & Management (AREA)
- General Physics & Mathematics (AREA)
- Theoretical Computer Science (AREA)
- Financial Or Insurance-Related Operations Such As Payment And Settlement (AREA)
Abstract
The invention provides a method for assessing damage after a vehicle collision, which comprises the following steps: (a) Acquiring a vehicle loss assessment historical case, and establishing a vehicle loss assessment historical case database; (b) receiving vehicle information to be assessed; (c) And calculating and determining the damage rating of the vehicle to be damaged according to the vehicle information to be damaged based on the damage history case data. The damage assessment method can rapidly, accurately and individually provide damage assessment by utilizing big data and machine learning technology, and is not only suitable for independent use of each insurance company, but also suitable for a third-party damage assessment platform.
Description
Technical Field
The invention relates to the field of vehicle insurance, in particular to a method and a system for damage assessment after vehicle collision.
Background
The loss estimation (loss estimation) after a vehicle collision usually considers a plurality of technical factors including part price, repair man-hour, local manual rate and the like.
The impairment method relies on a combination of impairment tools based on technical factors and subsequent macroscopic manual adjustments. Thus, in many cases, existing impairment methods and systems are used as references only, are not adaptable to dynamic changes in the market, and may result in price negotiations for multiple rounds. For example, it sometimes takes days, even weeks, to agree on using existing impairment systems, thus presenting problems of inefficiency, high labor costs, and high failure rates, reducing customer satisfaction.
Accordingly, there is a need for an improved impairment method and system that overcomes one or more of the problems of the prior art as discussed above.
Disclosure of Invention
One aspect of the present invention provides a vehicle impairment determination method adapted to be executed on a computer, the method comprising: (a) Acquiring a vehicle loss assessment historical case, and establishing a vehicle loss assessment historical case database; (b) receiving vehicle information to be assessed; and (c) calculating an impairment amount of the impairment-to-be-impaired vehicle based on the impairment history case data according to the impairment-to-be-impaired vehicle information.
In some embodiments, the method further comprises: and establishing an assessment model according to the vehicle assessment historical case. In the case of building the impairment model, the step (c) specifically includes: and calculating the damage amount of the vehicle to be damaged according to the information of the vehicle to be damaged and the damage model.
In some embodiments, the method further comprises: in the process of calculating the damage amount of the vehicle to be damaged, calculating a self-evaluation confidence index for indicating the accuracy of the damage amount.
In some embodiments, the method further comprises, after step (c): and adding the vehicle information to be damaged and the damage amount thereof into the vehicle damage assessment historical case database.
In some embodiments, the method may further comprise: and after the vehicle information to be damaged and the damage amount thereof are added into the vehicle damage history case database, updating the damage model according to the vehicle information to be damaged and the damage amount thereof.
In some embodiments, the hierarchy of the vehicle impairment history case database established in step (a) comprises a case level, an impairment element level, and an impairment factor level.
In some embodiments, the vehicle impairment history case database includes, but is not limited to, data associated with impairment factors and case-level data such as vehicle brand, train, model, year of vehicle production, vehicle impaired photograph, impaired part name and number, impaired part picture, number of impaired parts, impaired degree of impaired parts, replacement part price, repair man-hours, local labor rates and impairment rates. In some embodiments, the vehicle impairment history case database further comprises data associated with impairment element layers such as time of collision, location of collision, repair manufacturer name, location of collision, and additional cost.
In some embodiments, the method further comprises: and preprocessing the vehicle damage assessment historical case data in the vehicle damage assessment historical case database before establishing the damage assessment model according to the vehicle damage assessment historical case. The preprocessing may be performed off-line. The preprocessing modes include, but are not limited to, classifying, clustering, aggregating, sorting, summarizing, counting, cleaning, etc. the data, for example, classifying the historical cases according to collision location.
In some embodiments, the establishing the damage assessment model according to the vehicle damage assessment historical case specifically includes: the impairment elements are associated with impairment factors by feature analysis and classification/clustering algorithms and/or machine learning to build an impairment model.
In some embodiments, the vehicle information to be damaged includes, but is not limited to, damaged vehicle information, damaged part information, including damage level, case attribution information, repair information, and the like.
In some embodiments, the calculating the damage rating of the vehicle to be damaged according to the information of the vehicle to be damaged and the damage model includes: the calculation is performed at a case level to obtain at least one impairment history case not lower than a predetermined similarity. In this case, the calculation results in at least one complete impairment history case comprising a reference payout amount for a plurality of impairment factors.
Alternatively, in some embodiments, the calculating the damage rating of the vehicle to be damaged according to the information of the vehicle to be damaged and the damage model includes: the calculation is performed at the level of the impairment factor to obtain different historical data of one impairment factor of the vehicle to be impaired. In this case, the calculation is based on the historical payouts in the impairment factors, resulting in reference payouts for the respective impairment factors for the vehicle to be impaired.
Alternatively, in some embodiments, the calculating the damage amount of the vehicle to be damaged according to the information of the vehicle to be damaged and the damage model may include: the impairment history cases are time-ordered and the impairment scores of the most recent one or more impairment history cases and/or factors are averaged when more than one impairment history cases and/or impairment factors are identical or not below a predetermined similarity. In other embodiments, the impairment is determined by other algorithms when there is more than one impairment history case and/or factor that is the same or not less than a predetermined similarity.
Alternatively, in other embodiments, calculating the damage rating of the vehicle to be damaged according to the information of the vehicle to be damaged and the damage model may include: the determination of the impairment amount may involve the use of relevant data in other historical cases. For example, if historical pay data for one or more damaged parts corresponding to a vehicle to be damaged is missing in one or more highly similar historical cases, historical pay data for one or more identical parts may be found from other historical cases (e.g., historical cases having a similarity below a predetermined similarity) for the damage of the part, and the damage rating of the vehicle to be damaged may be determined.
Alternatively, in some embodiments, in step (c), the computation is performed at both the case level and the impairment factor level, i.e., the computation does not involve the use of a single complete historical case data. For example, the calculation determines the impairment amount based on impairment factor data from different categories/attributes of different historical cases.
In some embodiments, the calculating a confidence index for indicating the accuracy of the current impairment comprises: and searching for factors such as the number and frequency of historical occurrences of the damage assessment factors and the damage assessment factors matching the damage assessment factors of the vehicle to be assessed and the distribution and quality of the historical pay amounts in the damage assessment factor aggregation set according to the information (which can comprise one or more damage assessment factors and damage assessment factors) of the vehicle to be assessed, and calculating a confidence index.
In some embodiments, the method further comprises: determining whether the confidence index is below a predetermined threshold, and generating prompt data to trigger a manual process flow when the confidence index is below the threshold.
In some embodiments, the method further comprises: one or more adjustments are applied based on the data of the at least one impairment history case to determine an impairment amount of the vehicle to be impaired. In some embodiments, the one or more adjustments are applied based on one or more factors of part price change rate, manual rate change rate, KPI (Key Performance Indicator, key performance indicators) achievement rate, insurance business contribution rate, and the like. In some embodiments, the one or more adjustments comprise manual adjustments.
In some embodiments, the resulting impairment of the conventional impairment method may be used as a test point for the resulting impairment to verify the accuracy of the method of the invention.
In some embodiments, the present invention provides a vehicle impairment system comprising a processor; and a memory for storing vehicle impairment instructions adapted to be loaded by the processor to perform any of the vehicle impairment methods described above.
In some embodiments, the present invention provides a computer readable non-volatile medium storing computer readable instructions adapted to be loaded by a processor to perform any of the vehicle impairment methods described above.
According to the method provided by the invention, the damage rating of the vehicle to be damaged is calculated and determined based on the historical case data and/or factor data according to the information of the vehicle to be damaged and records in the historical case database. The damage assessment method can rapidly, accurately and individually provide damage assessment by utilizing big data and machine learning technology, and is not only suitable for independent use of each insurance company, but also suitable for a third-party damage assessment platform.
Drawings
FIG. 1 is a flow chart of an exemplary method for impairment according to one embodiment.
Fig. 2 is a flowchart of an exemplary method of creating a vehicle impairment history case database according to one embodiment.
Fig. 3 is a flow chart of an exemplary method for impairment according to another embodiment.
Fig. 4 is a table showing an exemplary relationship of an impairment element to an impairment factor.
Fig. 5 is a block diagram of an exemplary impairment system according to one embodiment.
Fig. 6 is a schematic diagram of a data structure of a vehicle impairment history case according to one embodiment.
FIG. 7 is a schematic diagram of a maintenance project table according to one embodiment.
FIG. 8 is a schematic diagram of an impairment model according to one embodiment.
FIG. 9 is a schematic diagram of a machine learning algorithm according to one embodiment.
Fig. 10 is a schematic diagram of a loss assessment flow according to one embodiment.
Detailed Description
The invention will now be described in detail with reference to exemplary embodiments, some examples of which are illustrated in the accompanying drawings. The following description refers to the accompanying drawings in which the same numbers in different drawings represent the same or similar elements, unless otherwise indicated. The schemes described in the following exemplary embodiments do not represent all schemes of the present invention. Rather, these are merely examples of systems and methods of various aspects of the present invention that are set forth in the following claims.
The loss based on technical factors requires that after determining damaged parts and the degree of damage, it is determined from a variety of technical factors, which requires that insurance companies or third parties continuously maintain databases of various vehicle brands, years, models, part types, part prices, maintenance man-hours, manual rates, corresponding maintenance logics, etc. There are challenges to achieving acceptable rates for both insurance companies and repair parties. For example, it is often difficult for insurance companies and loss tool providers to obtain accurate and high coverage databases as the basis for loss, data sources for third parties are often not approved, and so on. These data volumes are large (e.g., various vehicle brands, years, models, part types, part prices, repair man-hours, labor rates, etc.), wide coverage, fast updating, and difficult to agree (e.g., on repair man-hours) are challenges for accurate loss assessment. Furthermore, in china, for example, insurance companies actually still adopt a macroscopic claim management mode, rather than a case-by-case microscopic accurate loss assessment mode, the former taking into account many non-technical factors. For example, if a dealer (e.g., a 4S shop) servicing a vehicle to be assessed could bring a lot of vehicle insurance traffic, the insurer may alter parameters or manually adjust the assessment results accordingly. As another example, if the annual KPI (key performance indicators) of an insurance company has reached the standard, the impairment may relax to increase. This macro management mode of insurance companies may lead to neglecting fine management at the individual case level.
FIG. 1 illustrates an exemplary method 100 for vehicle impairment according to one embodiment of the invention. The method 100 begins at step 110 with building a vehicle impairment history case database. The exemplary method 100 is implemented based on the creation of a database of vehicle impairment history cases. Various insurance companies form a large amount of historical case data in long-term vehicle damage claim business, and the damage amounts of the historical cases are obtained through calculation of conventional damage assessment tools based on technical factors and combination with manual adjustment based on non-technical factors according to a conventional damage assessment mode. Therefore, the damage rates of the historical cases are the most realistic reflection of market conditions, experience and knowledge of damage fighters and nuclear damage fighters are precipitated, the damage rates of the vehicles to be damaged are determined based on the damage rates of the historical cases, and the accuracy of the damage rates can be ensured.
The method of creating the historical case database may be implemented using database creation techniques that are common in the art, for example, a relational database may be used. Method 200 illustrates an exemplary method of establishing a historical case database. The method 200 begins with an impairment factor 210 defining an impairment history case for a vehicle. The damage-assessment factors may include one or more of a vehicle brand, a train, a model, a collision location, a part name, a part number, a part price, a repair man-hour, a local labor rate, a collision occurrence time, a collision occurrence place, a repair manufacturer name, an additional cost expenditure, a year of vehicle production, a version of a model, a body color, a degree of damage to a part, and a body trim. The impairment factor may be represented using a data structure used by a relational database. In some particular embodiments, the impairment factor may be represented in the form of one or more database tables. In step 220, impairment elements of the vehicle impairment history case are defined. In some embodiments, the impairment element may include a collision type, including, but not limited to, a front right case, a rear right case, a front left case, a front right case, a left side case, a right side case, a rear left case, a rear right case, a mild case, a severe case, or various combinations thereof. In other embodiments, the damage-assessment element may include, but is not limited to, damage-assessment vehicle information (brand, model, year, configuration, etc.), damaged accessory information (accessory, damage level, etc.), affiliated insurance agency information (area, discount rate, etc.), repair information (repair shop level, accessory source, discount rate, etc.), and the like. Steps 210 and 220 may also be performed in reverse order. Finally, in step 230, one or more impairment elements are respectively associated with one or more impairment factors. For example, one impairment factor is associated with multiple impairment elements. As another example, one or more identical impairment factors are shared between each impairment element. The relationship between the impairment factor and the impairment element is illustratively shown in table 400 shown in fig. 4.
These impairment factors and impairment elements may be updated, e.g., added, deleted, or modified, so that the database can be maintained from more, updated dimensions. When the method of the present invention is implemented in a third party loss assessment platform, historical cases from multiple insurance companies may be aggregated and a historical case database may be built in accordance with the exemplary method described above.
In a specific embodiment, as shown in fig. 6, the impairment element may include: vehicle model/configuration, damaged parts, damaged extent, accessory source, accessory discount rate, man-hour discount rate, and the like. The impairment factor may include: loss items, maintenance items, disassembly items, and paint items, wherein loss items may include loss item names (e.g., loss item 1), accessory numbers, unit prices, usage, and accessory fees. The associated attribute information included in the loss item is stored in association in the relational database, and the attribute information may correspond to an attribute value, for example, the configuration number may be an identification number of a combination of letters and numbers, and the unit price, the usage amount, and the accessory fee may be arabic numerals. The maintenance items may include a maintenance item name (e.g., maintenance item 1), a job type, a maintenance level, a maintenance man-hour, and a maintenance fee. The maintenance item-related attribute information is stored in association with a database, and for example, the names of these attributes can be used as keys, and the values corresponding to the attributes can be used as values.
In some embodiments, the vehicle impairment history case database may include a claim case information table, a branch office table, a repair shop information table, a damaged vehicle information table, and a repair project table. For example, the claim information table may include information shown in the reduced table 1 in the database, including: loss assessment list number, time of risk, place of risk, institution, repair shop, final case setting time, VIN code, vehicle type and reason of risk.
TABLE 1
Loss assessment list number | Time of risk | Dangerous place | Mechanism | Repair shop | Final time of case setting | VIN code | Vehicle model | Reasons for danger |
xxxx08022016010750-0202-3-1 | 2018/12/5 20:20 | Chongqing | Chongqing division | XX0351001596 | 2018/12/27 15:47:09 | WP0AB29879U780xxx | 2006-Cayman S 3.4L | Crash of collision |
The branch office table may include information in the database as shown in simplified table 2, including: institutions, 4S store discount rates, market price discount rates, and applicable price discount rates.
TABLE 2
Mechanism | 4S store discount rate | Market price discount rate | Applicable price discount rate |
Suzhou division Co Ltd | 85.00 | 65.00 | 80.00 |
The repair shop information table may include information in the database as shown in the simplified table 3, including: a repair shop code, a repair shop name, an association, a repair shop type, an accessory channel, an accessory discount rate, a man-hour unit price type, and the like.
TABLE 3 Table 3
Repair shop code | Repair shop name | Association mechanism | Repair shop type | Fitting channel | Fitting discount rate | Time discount rate | Type of man-hour unit price |
XX0351001596 | Chongqing XX automobile maintenance service Co.Ltd | Chongqing division | Three types of repair shops | Market price | 90% | 100% | Three types of repair shops |
XX0361002265 | Xuzhou XX automobile trade Co.Ltd | Jiangsu division Co Ltd | 4S shop | 4S store price | 100.00 | Repair shop | |
XX0367000606 | Zhangqiu XX Daqi repair works | Shandong division Co Ltd | Class II repair shop | Market price | 100.00 | Class II repair shop | |
XX0367001468 | Solar XX automobile sales service Co.Ltd | Shandong division Co Ltd | 4S shop | 4S store price | 100.00 | 4S shop | |
XX0383002331 | XX Petroleum equipment technologies Co.Ltd | Middle energizer support of the middle energizer | Repair shop | Market price | 100.00 | Repair shop | |
XX0393001122 | Daqing city salon district XX automobile repair shop | Heilongjiang division Co Ltd | Repair shop | Market price | 100.00 | Repair shop |
The damaged vehicle information table may include in the database the information shown in the simplified table 4, including: number plate type, engine number, new vehicle purchase price, date of first date, actual value, vehicle use property, country, long plate, train, vehicle model, power source, engine model, transmission type, displacement, overturning, whether an airbag can run, loss degree, main collision point, secondary collision point and the like.
TABLE 4 Table 4
Loss assessment list number | Bulletin number | Number plate variety | Engine number | New vehicle acquisition price | Date of initial sign-up | Actual value | Vehicle use Properties | Country of China |
xxxx0308022016010750-0202-3-1 | Small-sized automobile | MA121C900839 | 1200000 | 2009/5/12 | 300000 | Self-use | Germany |
Table 4 (subsequent)
Plant brand | Vehicle system | Vehicle model | Style type | Power source |
Porsche | Kaimen [ Cayman ] (2006-2008) | 2006- Cayman S 3.4L | 2006- Cayman S 3.4L M97.21 A87.21 | Gasoline |
Table 4 (subsequent)
Engine type | Transmission type | Displacement volume | Capsizing | Whether or not the air bag is opened | Whether or not to run | Degree of loss | Main collision Point | Secondary collision point |
M97.21 | 5-Gear automatic transmission | 3.40 | Whether or not | Is that | Cannot run | Heavy weight | Left front part |
The repair items table may include information in the database as shown in fig. 7, including: repair shop, damage item name, accessory number, operation type, reference unit price, usage, post-folding accessory, work type, maintenance level, post-folding maintenance, reference disassembly and assembly, paint type, reference paint, auxiliary material cost, accessory external repair, accessory residual value, accessory depreciation, management rate, disassembly and assembly discount rate, paint superposition discount rate and the like.
The historical cases in the fixed vehicle damage assessment historical database provided by the embodiment of the application can be divided from different layers, for example, the fixed vehicle damage assessment historical database can be claim case information, damage assessment factors and damage assessment factors. For example, the above-described partitioning of historical cases may be accomplished using machine learning. For example, as shown in fig. 6, from the claim case itself, the historical case data may include accident vehicle base information, repair shop information, and insurance branch information, where the accident vehicle base information may include: VIN, engine number, vehicle model, new purchase price, license initial registration date, vehicle use property, occurrence cause, accident handling mode, accident type, collision level, loss photo, etc., the insurance branch may set different accessories and man-hour discounts according to the location, and the repair shop information may include repair shop code, repair shop name, repair shop type, association mechanism and cooperation type, etc. From the impairment element, the historical case data may include: vehicle/configuration, damaged parts, damaged extent, accessory source, accessory discount rate, formula discount rate, and the like. From the loss factor, the historical case data may include: the replacement item, auxiliary material item, sheet metal item, disassembly and assembly item, paint spraying item and the like of each claim, and the pay amount of each item can be associated with the damage assessment element classification. The loss assessment element is an assessment element that determines and affects the amount of payable of the loss assessment factor. And (3) associating the damage assessment factors in a large number of historical claim cases with the damage assessment factors, and calculating to obtain the damage assessment factor pay pricing reference value under the condition of the same/similar damage assessment factors, thereby obtaining the damage assessment model. And calculating the damage amount of the vehicle to be damaged according to the information of the vehicle to be damaged and the damage model.
In some embodiments, the damage factors in the historical cases are classified according to the damage factors, a set of damage factors for the same damage factor can be calculated, and one or more historical payouts in the set reflect historical payouts for the same damage factor. The reference payoff amount of the loss factor in the loss element set can be obtained by using rules and algorithm to calculate and process the payoff amount set. Reference payoff calculation rules and algorithms include, but are not limited to, median; taking an average value; taking the number of identical payouts in the collection when the number exceeds a predetermined proportion (e.g., 50%); removing singular values, such as the first 10% and the last 10% of the payoff amount of an impairment factor according to the arrangement from large to small; etc. Based on the tabular impairment model, an impairment amount of the impairment-to-be-impaired vehicle may be calculated. Fig. 8 is a simplified model of the above-mentioned damage, the left columns 2-4 (damage item name, accessory number, and operation type) are historic case damage factors, the columns 5-8 (branch, repair shop type, accessory channel, man-hour unit price type) are damage factors, the columns 9-11 (accessory fee, dismounting fee, and paint fee) are pay amounts of historic damage factors, and the columns 12-14 are calculated damage factor reference pay amounts.
In some embodiments, the impairment model may be built using machine learning, using algorithms including, but not limited to, neural networks, reinforcement learning neural networks, generation networks in antagonism neural networks, and the like. The first stage is a model generation stage, and the specific structure of the neural network can be set by those skilled in the art according to the actual requirements of the damage assessment element, such as parts, maintenance, paint spraying, disassembly and assembly, part context, maintenance unit context, accessory channels, regions, branches, etc., which are not limited in this embodiment of the present invention. In the embodiment of the invention, the loss assessment factors and the historical pay amounts of the corresponding loss assessment factors are input into the machine learning algorithm for learning training, and are output as the reference price (including parts and working hours) of each item of content in the corresponding loss assessment claim case so as to establish the loss assessment model. The second stage is a discrimination model stage in which the rationality of the price predicted by the first stage is judged based on the deep learning model, and the output of the step is 0 and 1, i.e., whether the price is reasonable. In the training process, the object of generating the model is to generate the cheating judgment model of each price in the loss-assessment claim case as much as possible. The object of the discrimination model is to separate the price obtained by generating the model and the real price as much as possible. Thus, the generated model and the discrimination model form a dynamic game process. And finally, generating a price which is enough to be real.
In a specific embodiment, the machine learning algorithm model described above may be as shown in fig. 9. The machine learning algorithm incorporates generating a multi-layer deep learning model of the countermeasure network and the neural network to make a claim price prediction. As shown in fig. 9, the left side is a generating model in which a part set and a maintenance item set of a claim list are generated, and all parts and maintenance items appearing in the claim list are constructed into an N-dimensional input vector containing N parts and maintenance items, wherein N is an integer greater than 1; and carrying out deep learning through Encoder to finally generate a predicted price corresponding to the N-dimensional input vector, simultaneously mixing real claim list (GI) data, sending the data to a right-side countermeasure network model, and judging the generated predicted price through a deep learning model of a Decoder part, thereby judging the rationality of the generated price and obtaining a confidence index.
Due to the deviation in the integrity of the historical case and the actual impairment, the present invention calculates a confidence index for indicating the accuracy of the current impairment, comprising: and searching the historical occurrence times and frequencies of the damage elements and the damage factors of the vehicle to be damaged, the distribution and quality of the historical pay amount in the damage factor aggregation set and other factors according to the information (which can comprise one or more damage elements and damage factors) of the vehicle to be damaged, and calculating the confidence index. Taking fig. 7 as an example, when a certain damage factor in the damage claim case does not appear in the damage model, the confidence index of the damage factor is 0; the historical payoff amount variance of the factor is inversely related to the confidence index, and so on. In the historical case, when the payoff amount corresponding to a certain loss factor is in a relatively fixed value interval, the distribution of the loss factor is considered to be more stable, and accordingly, the confidence index is improved; in the historical case, where a loss factor occurs only twice and the payouts of the two times differ significantly, the confidence index in this case is also low. And combining the confidence indexes of each loss assessment factor in the loss assessment claim according to the proportion of the amount of the payable to obtain the loss assessment confidence index of the loss assessment claim.
In some embodiments, to solve the problem that the damage factor in the damage-assessment claim case cannot be matched with the proper damage reference value in the calculation process when the number of the historical damage cases is insufficient, a clustering algorithm is adopted to cluster similar and similar variables of certain damage-assessment factors, for example, branches or regions with the same or similar man-hour standards are clustered in the calculation process, so that the success rate and the accuracy of damage assessment of the damage-assessment factor are improved.
In some embodiments, to solve the problem that the damage factor in the damage-assessment case does not match the proper damage reference value in the calculation process when the number of the historical damage cases is insufficient, a conditional search method is adopted, such as searching at first the set of damage factors of the most ideal damage-assessment factors, such as searching failure, searching at the suboptimal set obtained by calculation, such as re-failure, searching at the re-set until success or complete failure is achieved, so as to improve the success rate and accuracy of damage assessment of the damage factor. For each stage of calculation success, a corresponding confidence index is assigned, for example, one step success is 1, and the total failure is 0. The method can improve the success rate and accuracy of damage assessment.
In some embodiments, it is determined whether the confidence index is below a predetermined threshold, and when the confidence index is below a threshold, prompt data is generated to trigger a manual process flow.
In some embodiments, the result of the artificial damage is added to the vehicle damage history case database, and the damage model is updated according to the vehicle information to be damaged and the damage amount thereof. For example, the artifacts would be added to the impairment model and give a higher computational weight to adjust the reference payout of the impairment factor.
In step 120, the method receives information about a vehicle to be damaged. Such relevant information may include one or more of a damage factor and damage factor, such as license plate number, vehicle brand, train, vehicle type, impact location, damaged part name and number, vehicle year of production, version of vehicle type, body color, degree of damage to parts, and body trim. As can be expected by those skilled in the art, the more adequate the pending damage vehicle information is provided, the easier it is to match records in the historical case database, thereby retrieving more matching claims/damage records, providing a basis for outputting cases with high similarity. Of course, on the other hand, matching of cases also depends on the breadth and integrity of records in the database. In actual operation, the input of relevant information of the vehicle to be damaged may not depend on the piece-by-piece input of data. For example, in some cases, when entering a license plate number of a vehicle to be compromised, the method 100 may automatically retrieve information from an official platform or other platform for the corresponding vehicle brand, train, model, year of vehicle production, version of model, body color, body trim, etc.
In step 130, the vehicle to be damaged is calculated from the historical case database for similar historical cases. The calculation comprises traversing a vehicle damage assessment historical case database, and calculating the similarity between the vehicle to be damaged and each case or/and factor in the vehicle damage assessment historical case database according to a preset calculation mode and based on the received vehicle information to be damaged. The predetermined calculation mode may be a machine learning algorithm or may be determined by one skilled in the art as desired. An exemplary calculation mode is to calculate a weighted sum of the similarity of the values of the impairment factors. Another exemplary calculation mode is to calculate a weighted average of the similarity of the values of the impairment factors. In the exemplary calculation mode, the person skilled in the art needs to determine the weight of each impairment factor, for example the weight of the vehicle brand, train, model may be higher than the weight of the vehicle production year, version of the model, body color, or the weight of the impact location, the degree of damage to the part may be higher than the weight of the impact location. The determination of the weights may be determined by a person skilled in the art with a limited number of tests.
In some cases, the resulting historical cases are ranked in a high-to-low order of similarity. Or taking the preset similarity as a threshold value, and only outputting the historical cases above the threshold value. When the loss history cases are not lower than the predetermined similarity by more than one, the loss history cases are sorted in time. The predetermined similarity may be set and adjusted by those skilled in the art according to the actual circumstances. In some cases, the similarity can be adjusted in real time, and the output similar cases correspondingly increase or decrease in real time, so that the user can determine the number of similar historical cases according to the actual case.
In step 140, the impairment is calculated based on the obtained data of the historical cases. An exemplary algorithm determines the impairment of the vehicle to be impaired based on the impairment of the resulting one or more impairment history cases or an average thereof. When there are multiple similar impairment history cases above the similarity threshold, the nearest case or cases may be selected to determine the impairment of the vehicle to be impaired (e.g., averaged). Alternatively, when there are multiple similar impairment history cases above the threshold, the impairment of one of the history cases may be subject to. For example, if a certain historical case matches exactly the information of the vehicle to be damaged, it may not be necessary to average as described above.
Another exemplary method of determining the balance of a vehicle to be evaluated involves the use of relevant data in other historical cases. For example, if one or more historical odds of a damaged part corresponding to a vehicle to be damaged is missing in one or more historical cases with high similarity, one or more historical odds of the same part may be found from other historical cases for use in the damage assessment of the part, thereby determining the damage assessment of the vehicle to be damaged.
Another exemplary method of determining the balance of a vehicle to be evaluated does not involve the use of a single complete historical case data. In this case, the method 100 determines the impairment amount based on data from different categories/attributes for different historical cases.
In step 140, one or more adjustments may be applied over the determined impairment to ultimately determine the impairment for the vehicle to be impaired. These adjustments are applied, for example, based on one or more factors such as part price change rate, manual rate change rate, KPI compliance rate, insurance service contribution rate, etc. These adjustments may be computer-implemented. For example, the adjustment is made according to the part price change rate, for example, according to the following formula: some accessory key full price = the accessory reference price the rate of change of the accessory price for the average lead time period of the historical claim.
The adjustment is performed according to the rate of change of the manual rate, and can be according to the following formula: a certain man-hour fee key full price =the man-hour fee reference price is the man-hour fee change rate of the average lead time period of the historical claim case.
Because the historical damage assessment case is a case before a certain time, has certain hysteresis in time, and the damage assessment made by referring to the historical case has certain error, the accuracy and the real-time performance of the damage assessment determination can be improved by adjusting the current factors such as the part price change rate, the manual rate change rate, the KPI standard reaching rate, the insurance service contribution rate and the like on the damage assessment made by referring to the historical case.
Method 300 illustrates another exemplary impairment method. The method 300 begins by building a vehicle impairment history case database (step 310) that is substantially identical to step 110, but during the building process, the method 300 further performs data preprocessing on the database. The pretreatment may typically be performed off-line/off-line. The data preprocessing can be based on big data analysis aggregation algorithm, and can also be performed according to different purposes. For example, historical cases may be categorized based on collision type, so that matching records may be more quickly and accurately screened according to collision type when matching the historical case database, without traversing all data records. Steps 320 through 340 are substantially identical to steps 120 through 140 of method 100. In step 350, the method 300 adds the impairment vehicle information and its impairment amount to the historical case database, thereby enriching and updating the records of the database, providing more accurate and timely matching records for the impairment of new cases. When the historical case database is updated (step 352), the data preprocessing step therein needs to be performed again. Thus, the data preprocessing is dynamic in nature. When the present invention is implemented in artificial intelligence and machine learning, the addition of new data records may allow the machine to dynamically adjust its algorithm, thereby continually improving accuracy.
A specific embodiment of the present disclosure is described below in conjunction with fig. 10. Rectangular box 1000 identifies the architecture of the impairment system provided by the present disclosure. Vehicle impairment history cases are constructed according to the massive number of impairment sheets already completed (step 1001), and then big data machine learning impairment models are performed according to the impairment history cases (step 1002). After the impairment model learning, maintenance inventory data for the vehicle to be impaired is received, which may include one or more maintenance items, for example. These repair checklist data are obtained by image recognition of images of the human survey and/or the collision site. And (3) performing artificial intelligence algorithm damage assessment according to the damage assessment module obtained by machine learning and the maintenance list data of the damaged vehicle (step 1003) to obtain a separated damage assessment list and an intelligent confidence index (step 1004). If the confidence index is greater than a predetermined threshold, then automatic impairment may be performed (step 1005), otherwise, impairment is manually interposed (step 1006). The results of the manual impairment may be put into impairment model learning, updating the impairment model (step 1007). After automatic loss assessment, the loss assessment results are put into historical case learning.
Fig. 5 shows a block diagram of an impairment system 500 according to some disclosed embodiments. The system may include a processor 521, an input/output (I/O) device 522, a memory 523, a storage 526, a database 527, and a display 528.
The processor 521 may be one or more known processing devices, such as a Pentium ™ family of microprocessors manufactured by Intel ™, or a Turion ™ family of microprocessors manufactured by AMD ™. Processor 521 may comprise a single-core processor system or a multi-core processor system capable of parallel processing. For example, processor 521 may be a single-core processor with virtual processing techniques. In some implementations, the processor 521 may utilize a logical processor to execute and control multiple processes simultaneously. The processor 521 may execute virtual machine technology, or other similar known technology, to enable execution, control, admission, manipulation, storage of a plurality of software processes, applications, programs, etc. In another embodiment, processor 521 includes a multi-core processor configuration (e.g., dual-core or quad-core) configured to provide parallel processing functionality, allowing lossy system 500 to execute multiple processes simultaneously. Those skilled in the art will appreciate that other types of processor configurations may also be implemented to provide the functionality described herein.
Memory 523 may include one or more storage devices configured to store instructions for use by processor 521 to perform the functions of the disclosed embodiments. For example, memory 523 may be configured with one or more software instructions, such as instructions 524, that when executed by processor 521 may perform one or more operations. The disclosed embodiments are not limited to a single program or computer configured to perform specialized tasks. For example, the memory 523 may include a single instruction 524 that performs the functions of the impairment system 500, or the instruction 524 may include multiple instructions.
Memory 523 may also store data 525, which data 525 may reflect any type of information in any form that performs the functions in the disclosed embodiments. For example, the data 525 may include metadata of a historical case database related to impairment computations, as well as other data enabling the processor 521 to perform functions in the disclosed embodiments.
The I/O device 522 may be configured to allow data to be received and/or transmitted. I/O devices 522 may include one or more digital and/or analog communication devices that allow damage-assessment system 500 to communicate with other machines and devices. The impairment system 500 may also include one or more databases 527, or be communicatively coupled to one or more databases 527 via a network. For example, the database 527 may include an Oracle ™ database, a Sybase ™ database, or other relational or non-relational database, such as Hadoop sequence files, HBase, or Cassandra. In an exemplary embodiment, database 527 may store historical case data for impairment. For example, the metadata may be created by a user and stored in database 527.
The present invention also provides a computer readable medium storing computer readable instructions adapted to be loaded by a processor to perform any of the vehicle impairment determination methods of the present invention. The computer readable medium may include a removable medium as a package medium including a magnetic disk (including a floppy disk), an optical disk (including a CD-ROM (compact disc read only memory) and a DVD (digital versatile disc)), a magneto-optical disk (including an MD (mini disc)), or a semiconductor memory. In some embodiments, the computer readable medium resides, for example, in an application store to provide an application program, such as a mobile terminal application, encoding any of the vehicle impairment methods illustrated by the present invention.
Other embodiments of the invention will be apparent to those skilled in the art from consideration of the specification and practice of the invention disclosed herein. The scope of the invention is intended to cover any variations, uses, or adaptations of the invention following its general principles and including such departures from the present disclosure as come within known or customary practice within the art to which the invention pertains. It is intended that the specification and examples be considered as exemplary only, with a true scope and spirit of the invention being indicated by the following claims.
It will be appreciated that the invention is not limited to the precise construction described above and illustrated in the accompanying drawings, and that various modifications and variations may be made without departing from the scope of the invention. The scope of the invention is limited only by the claims.
Claims (9)
1. A method of vehicle impairment, adapted to be executed on a computer, the method comprising:
(a) Acquiring a vehicle damage assessment historical case, and establishing a vehicle damage assessment historical case database, wherein the database comprises damage assessment factors and damage assessment elements, and the damage assessment elements comprise vehicle types/configurations, damaged parts, damage degree, accessory sources, accessory discount rates and man-hour discount rates; the damage assessment factors comprise a loss project, a maintenance project, a disassembly project and a paint spraying project; the loss assessment historical case is a previous case and is constructed according to the completed loss assessment list;
(b) Establishing an assessment model according to the vehicle assessment historical case;
(c) Receiving vehicle information to be assessed, wherein the vehicle information comprises one or more assessment factors and assessment factors; and
(D) Calculating the damage amount of the vehicle to be damaged according to the information of the vehicle to be damaged and the damage model;
(e) In the process of calculating the damage amount of the vehicle to be damaged, calculating a confidence index for indicating the accuracy of the damage amount, comprising: searching the number and frequency of historical occurrences of the damage assessment factors and the damage assessment factors of the to-be-assessed vehicle and the distribution and quality of the historical pay amount in the damage assessment factor aggregation set according to the to-be-assessed vehicle information, and calculating a confidence index; wherein:
(i) When the loss factor in the loss history case does not appear in the loss model, the confidence index of the loss factor is 0;
(ii) The variance of the historical payoff amount of the loss assessment factor is inversely related to the confidence index;
(iii) When the historical payoff amount corresponding to the loss-determining factor is in a relatively fixed value interval, the confidence index of the loss-determining factor is improved;
(iv) When the loss factor appears only twice and the payouts of the two times differ greatly, the confidence index of the loss factor is reduced;
(v) When the number of the historical cases is insufficient, a condition searching method is adopted, and corresponding confidence indexes are assigned to each stage of calculation success.
2. The method of claim 1, wherein the method further comprises, after step (d): (f) And adding the vehicle information to be damaged and the damage amount thereof into the vehicle damage assessment historical case database.
3. The method of claim 1, wherein the data in the vehicle impairment history case database is preprocessed in step (a).
4. The method of claim 1, involving applying one or more adjustments in step (d) to determine a final balance of the vehicle to be damaged.
5. The method of claim 4, wherein the one or more adjustments are applied based on one or more factors of part price change rate, manual rate change rate, KPI achievement rate, insurance business contribution rate, or the one or more adjustments include manual adjustments.
6. The method of claim 2, the method further comprising: and after the vehicle information to be damaged and the damage amount thereof are added into the vehicle damage history case database, updating the damage model according to the vehicle information to be damaged and the damage amount thereof.
7. The method of claim 1, the method further comprising: the impairment elements are associated with impairment factors by feature analysis and classification/clustering algorithms and/or machine learning to build an impairment model.
8. A vehicle impairment system, comprising:
A processor; and
A memory for storing vehicle impairment instructions adapted to be loaded by a processor to perform the method of any one of claims 1 to 7.
9. A computer readable non-transitory medium storing computer readable instructions adapted to be loaded by a processor to perform the method of any one of claims 1 to 7.
Applications Claiming Priority (5)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201810250608.8A CN110363669A (en) | 2018-03-26 | 2018-03-26 | The automatic damage identification method of vehicle collision and system based on non-technique factors |
CN201810250609.2A CN110363670A (en) | 2018-03-26 | 2018-03-26 | Vehicle collision damage identification method and system based on history case |
CN2018102506092 | 2018-03-26 | ||
CN2018102506088 | 2018-03-26 | ||
PCT/CN2019/079629 WO2019184899A1 (en) | 2018-03-26 | 2019-03-26 | Vehicle collision damage assessment method and system based on historical cases |
Publications (2)
Publication Number | Publication Date |
---|---|
CN111886619A CN111886619A (en) | 2020-11-03 |
CN111886619B true CN111886619B (en) | 2024-07-02 |
Family
ID=68062521
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN201980021736.8A Active CN111886619B (en) | 2018-03-26 | 2019-03-26 | Vehicle collision damage assessment method and system based on historical cases |
Country Status (2)
Country | Link |
---|---|
CN (1) | CN111886619B (en) |
WO (1) | WO2019184899A1 (en) |
Families Citing this family (5)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN113362189A (en) * | 2021-06-11 | 2021-09-07 | 爱保科技有限公司 | Vehicle claim processing method and device, medium and electronic equipment |
CN113361424A (en) * | 2021-06-11 | 2021-09-07 | 爱保科技有限公司 | Intelligent loss assessment image acquisition method, device, medium and electronic equipment for vehicle |
CN114241398A (en) * | 2022-02-23 | 2022-03-25 | 深圳壹账通科技服务有限公司 | Vehicle damage assessment method, device, equipment and storage medium based on artificial intelligence |
CN115018658A (en) * | 2022-08-09 | 2022-09-06 | 太平金融科技服务(上海)有限公司深圳分公司 | Loss price early warning method, loss price early warning device, loss price early warning equipment, storage medium and program product |
CN116935293B (en) * | 2023-09-15 | 2024-01-02 | 国任财产保险股份有限公司 | Automatic processing method and system for remote video exploration and damage assessment |
Citations (3)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN106570759A (en) * | 2016-11-09 | 2017-04-19 | 中国平安财产保险股份有限公司 | Intelligent risk scoring method and system for loss assessment of auto insurance case |
CN106600422A (en) * | 2016-11-24 | 2017-04-26 | 中国平安财产保险股份有限公司 | Car insurance intelligent loss assessment method and system |
CN106779799A (en) * | 2016-11-17 | 2017-05-31 | 中国平安财产保险股份有限公司 | Local setting loss price processing method and system based on vehicle insurance intelligence setting loss platform |
Family Cites Families (10)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
US9875414B2 (en) * | 2014-04-15 | 2018-01-23 | General Electric Company | Route damage prediction system and method |
US8630878B1 (en) * | 2011-06-17 | 2014-01-14 | Zenith Insurance Company | Determining likely outcomes of active insurance claims by calculating and examining aggregated outcomes of matching historic claims |
CN105488258B (en) * | 2015-11-24 | 2020-03-27 | 大连楼兰科技股份有限公司 | CAE technology-based automatic damage assessment method for vehicle collision |
CN105550756B (en) * | 2015-12-08 | 2017-06-16 | 优易商业管理成都有限公司 | A kind of quick damage identification method of automobile being damaged based on simulating vehicle |
CN105956667B (en) * | 2016-04-14 | 2018-09-25 | 平安科技(深圳)有限公司 | Vehicle insurance setting loss Claims Review method and system |
CN106021548A (en) * | 2016-05-27 | 2016-10-12 | 大连楼兰科技股份有限公司 | Remote damage assessment method and system based on distributed artificial intelligent image recognition |
CN106056452A (en) * | 2016-05-27 | 2016-10-26 | 大连楼兰科技股份有限公司 | Building method of remote unmanned damage assessment system and remote unmanned damage assessment method |
US10902525B2 (en) * | 2016-09-21 | 2021-01-26 | Allstate Insurance Company | Enhanced image capture and analysis of damaged tangible objects |
CN107180389B (en) * | 2017-05-10 | 2018-06-15 | 平安科技(深圳)有限公司 | People hinders Claims Resolution setting loss fee calculating method, device, server and medium |
CN108961084A (en) * | 2018-07-07 | 2018-12-07 | 北京精友世纪软件技术有限公司 | A kind of AI car damage identification Claims Resolution system and method for supporting cell phone client |
-
2019
- 2019-03-26 WO PCT/CN2019/079629 patent/WO2019184899A1/en active Application Filing
- 2019-03-26 CN CN201980021736.8A patent/CN111886619B/en active Active
Patent Citations (3)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN106570759A (en) * | 2016-11-09 | 2017-04-19 | 中国平安财产保险股份有限公司 | Intelligent risk scoring method and system for loss assessment of auto insurance case |
CN106779799A (en) * | 2016-11-17 | 2017-05-31 | 中国平安财产保险股份有限公司 | Local setting loss price processing method and system based on vehicle insurance intelligence setting loss platform |
CN106600422A (en) * | 2016-11-24 | 2017-04-26 | 中国平安财产保险股份有限公司 | Car insurance intelligent loss assessment method and system |
Also Published As
Publication number | Publication date |
---|---|
WO2019184899A1 (en) | 2019-10-03 |
CN111886619A (en) | 2020-11-03 |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
CN111886619B (en) | Vehicle collision damage assessment method and system based on historical cases | |
EP3043447B1 (en) | Distribution transformer heavy loading and overloading mid-term and short-term pre-warning analytics model | |
CN106557882A (en) | Power consumer screening technique and system based on various dimensions Risk Evaluation Factors | |
JP2008537266A (en) | Adaptive data cleaning | |
WO2010051433A1 (en) | Automated specification, estimation, discovery of causal drivers and market response elasticities or lift factors | |
US11367142B1 (en) | Systems and methods for clustering data to forecast risk and other metrics | |
US11367141B1 (en) | Systems and methods for forecasting loss metrics | |
CN112860769B (en) | Energy planning data management system | |
CN113837886B (en) | Knowledge-graph-based vehicle insurance claim fraud risk identification method and system | |
CN112579900B (en) | Second-hand vehicle replacement information recommendation method, system and equipment | |
KR102402782B1 (en) | Price rule integrated comparison engince | |
CN108416619B (en) | Consumption interval time prediction method and device and readable storage medium | |
CN111680382A (en) | Grade prediction model training method, grade prediction device and electronic equipment | |
KR20230122525A (en) | Method for credit evaluation based on external data and apparatus for performing the method | |
CN114187125A (en) | Claims case distribution method, device, equipment and storage medium | |
KR20220092306A (en) | Method for providing subscription economy based rental price comparison service | |
Laurentia et al. | YBM University tourism building location selection with a combination of cut off point and AHP Topsis method | |
US11822562B2 (en) | Unstructured text processing for geographical location determination system | |
KR102464993B1 (en) | Method for credit evaluation based on order data generated between online seller and customer on OMS(order management system) | |
KR102433934B1 (en) | Method for determination of credit evaluation score card based on e-commerce data and apparatus for performing the method | |
CN115907719A (en) | Hierarchical operation and maintenance management method and device for charging station | |
US20170186095A1 (en) | Centralized GAAP approach for multidimensional accounting to reduce data volume and data reconciliation processing costs | |
CN113362191A (en) | Equipment application data processing method and device, computer equipment and storage medium | |
CN116308158B (en) | National asset supervision and management system | |
CN114331595B (en) | Product recommendation method, device, equipment and storage medium based on artificial intelligence |
Legal Events
Date | Code | Title | Description |
---|---|---|---|
PB01 | Publication | ||
PB01 | Publication | ||
SE01 | Entry into force of request for substantive examination | ||
SE01 | Entry into force of request for substantive examination | ||
GR01 | Patent grant |