CN114091903A - Training method and device of loss assessment model, and loss assessment method and device - Google Patents
Training method and device of loss assessment model, and loss assessment method and device Download PDFInfo
- Publication number
- CN114091903A CN114091903A CN202111386073.5A CN202111386073A CN114091903A CN 114091903 A CN114091903 A CN 114091903A CN 202111386073 A CN202111386073 A CN 202111386073A CN 114091903 A CN114091903 A CN 114091903A
- Authority
- CN
- China
- Prior art keywords
- evaluated
- information
- loss
- loss value
- model
- 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.)
- Pending
Links
- 238000012549 training Methods 0.000 title claims abstract description 107
- 238000000034 method Methods 0.000 title claims abstract description 78
- 238000013210 evaluation model Methods 0.000 claims abstract description 103
- 231100000809 Damage Assessment Model Toxicity 0.000 claims description 29
- 238000011156 evaluation Methods 0.000 claims description 19
- 238000012545 processing Methods 0.000 claims description 19
- 238000004364 calculation method Methods 0.000 claims description 7
- 238000013139 quantization Methods 0.000 claims description 5
- 230000006735 deficit Effects 0.000 claims 3
- 238000004590 computer program Methods 0.000 description 14
- 230000006870 function Effects 0.000 description 10
- 238000010586 diagram Methods 0.000 description 8
- 230000006399 behavior Effects 0.000 description 7
- 230000008520 organization Effects 0.000 description 3
- 230000009471 action Effects 0.000 description 2
- 238000006243 chemical reaction Methods 0.000 description 2
- 238000004891 communication Methods 0.000 description 2
- 230000008569 process Effects 0.000 description 2
- 238000013473 artificial intelligence Methods 0.000 description 1
- 230000001413 cellular effect Effects 0.000 description 1
- 230000007812 deficiency Effects 0.000 description 1
- 239000000284 extract Substances 0.000 description 1
- 238000010801 machine learning Methods 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
- 238000005457 optimization Methods 0.000 description 1
- 230000001105 regulatory effect Effects 0.000 description 1
- 230000004044 response Effects 0.000 description 1
- 239000004984 smart glass Substances 0.000 description 1
- 238000012360 testing method Methods 0.000 description 1
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
- G06Q10/06—Resources, workflows, human or project management; Enterprise or organisation planning; Enterprise or organisation modelling
- G06Q10/063—Operations research, analysis or management
- G06Q10/0639—Performance analysis of employees; Performance analysis of enterprise or organisation operations
-
- 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
- G06Q10/04—Forecasting or optimisation specially adapted for administrative or management purposes, e.g. linear programming or "cutting stock problem"
-
- 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
- G06Q40/12—Accounting
Landscapes
- Business, Economics & Management (AREA)
- Engineering & Computer Science (AREA)
- Human Resources & Organizations (AREA)
- Strategic Management (AREA)
- Economics (AREA)
- Development Economics (AREA)
- Theoretical Computer Science (AREA)
- Physics & Mathematics (AREA)
- Marketing (AREA)
- General Physics & Mathematics (AREA)
- Entrepreneurship & Innovation (AREA)
- General Business, Economics & Management (AREA)
- Quality & Reliability (AREA)
- Tourism & Hospitality (AREA)
- Operations Research (AREA)
- Game Theory and Decision Science (AREA)
- Educational Administration (AREA)
- Accounting & Taxation (AREA)
- Finance (AREA)
- Technology Law (AREA)
- Management, Administration, Business Operations System, And Electronic Commerce (AREA)
Abstract
The embodiment of the specification provides a method and a device for training a resource loss assessment model, and a method and a device for assessing resource loss, wherein the method for training the resource loss assessment model comprises the following steps: obtaining multi-dimensional information to be evaluated of a party to be evaluated and a sample asset loss value of the party to be evaluated; inputting the multi-dimensional information to be evaluated into a resource loss evaluation model, and obtaining a predicted asset loss value output by the resource loss evaluation model; calculating a model loss value from the sample asset loss value and the predicted asset loss value; and adjusting model parameters of the resource loss evaluation model based on the model loss value, and continuing to train the resource loss evaluation model until a training stop condition is reached. By the method, an accurate and efficient resource loss evaluation model can be trained.
Description
Technical Field
The embodiment of the specification relates to the technical field of wind control, in particular to a training method of a resource loss assessment model. One or more embodiments of the present specification also relate to a training apparatus for a damage assessment model, a damage assessment method and apparatus, a computing device, a computer-readable storage medium, and a computer program.
Background
The asset loss is asset loss, asset loss assessment and asset loss assessment, the asset loss is usually due to asset loss caused by some behaviors or information of a party to be assessed, the asset loss caused by some behaviors or information of the party to be assessed is unpredictable at present, or a loss estimated according to human experience is usually inaccurate, and the asset loss possibly caused by behaviors, words, information and the like of the party to be assessed cannot be effectively assessed.
Disclosure of Invention
In view of this, the embodiments of the present disclosure provide a method for training a damage assessment model. One or more embodiments of the present disclosure also relate to a training apparatus for a damage assessment model, a damage assessment method and apparatus, a computing device, a computer-readable storage medium, and a computer program, so as to solve technical deficiencies of the prior art.
According to a first aspect of embodiments herein, there is provided a method for training a damage assessment model, including:
obtaining multi-dimensional information to be evaluated of a party to be evaluated and a sample asset loss value of the party to be evaluated;
inputting the multi-dimensional information to be evaluated into a resource loss evaluation model, and obtaining a predicted asset loss value output by the resource loss evaluation model;
calculating a model loss value from the sample asset loss value and the predicted asset loss value;
and adjusting model parameters of the resource loss evaluation model based on the model loss value, and continuing to train the resource loss evaluation model until a training stop condition is reached.
Optionally, the obtaining of the multidimensional information to be evaluated of the party to be evaluated includes:
and obtaining the information of the size to be evaluated and the violation information to be evaluated of the party to be evaluated.
Optionally, the obtaining the information of the amount to be evaluated and the information of the violation to be evaluated of the party to be evaluated includes:
acquiring information of a party to be evaluated, customer information and fund information of the party to be evaluated, and forming the information of the party to be evaluated, the customer information and the fund information into information of the amount of the party to be evaluated;
and acquiring violation information and law and regulation information of the party to be evaluated, and combining the violation information and the law and regulation information into the violation information to be evaluated of the party to be evaluated.
Optionally, obtaining a sample asset loss value of the party to be evaluated includes:
and collecting the capital loss value, the market value loss value and the reputation loss value of the party to be evaluated, and converting the capital loss value, the market value loss value and the reputation loss value into a sample asset loss value according to a preset quantization rule.
Optionally, the multidimensional information to be evaluated includes at least one piece of information to be evaluated;
inputting the multi-dimensional information to be evaluated into a resource loss evaluation model, and acquiring a predicted asset loss value output by the resource loss evaluation model, wherein the method comprises the following steps:
inputting the at least one piece of information to be evaluated into the loss evaluation model;
the resource loss evaluation model carries out embedding processing on the at least one piece of information to be evaluated to obtain at least one piece of vector to be evaluated;
obtaining a predicted capital loss value, a predicted market value loss value and a predicted reputation loss value of the party to be evaluated based on the at least one vector to be evaluated;
and forming a predicted asset loss value output by the asset loss evaluation model according to the predicted capital loss value, the predicted market value loss value and the predicted reputation loss value.
Optionally, the multidimensional information to be evaluated includes at least one piece of information to be evaluated, and each piece of information to be evaluated corresponds to an evaluation weight;
inputting the multi-dimensional information to be evaluated into a resource loss evaluation model, wherein the method comprises the following steps:
and inputting each piece of information to be evaluated and the corresponding evaluation weight into the asset loss evaluation model.
Optionally, the training stop condition includes:
the model loss value is smaller than a preset threshold value; or
The training round reaches the preset training round.
According to a second aspect of embodiments herein, there is provided a method for asset damage assessment, comprising:
determining a party to be evaluated;
acquiring multi-dimensional information to be evaluated of the party to be evaluated;
and inputting the multi-dimensional information to be evaluated into a resource loss evaluation model for processing, and obtaining an asset loss value output by the resource loss evaluation model, wherein the resource loss evaluation model is obtained by training through the resource loss evaluation model training method.
According to a third aspect of embodiments herein, there is provided a training apparatus for a damage assessment model, including:
the system comprises an acquisition module, a storage module and a processing module, wherein the acquisition module is configured to acquire multi-dimensional information to be evaluated of a party to be evaluated and a sample asset loss value of the party to be evaluated;
the prediction module is configured to input the multi-dimensional information to be evaluated into a resource loss evaluation model and obtain a predicted asset loss value output by the resource loss evaluation model;
a calculation module configured to calculate a model loss value from the sample asset loss value and the predicted asset loss value;
a training module configured to adjust model parameters of the resource damage assessment model based on the model loss value, and continue training the resource damage assessment model until a training stop condition is reached.
Optionally, the obtaining module is further configured to:
and obtaining the information of the to-be-evaluated body quantity and the to-be-evaluated violation information of the to-be-evaluated party.
Optionally, the obtaining module is further configured to:
acquiring information of a party to be evaluated, customer information and fund information of the party to be evaluated, and forming the information of the party to be evaluated, the customer information and the fund information into information of the amount of the party to be evaluated;
and acquiring violation information and law and regulation information of the party to be evaluated, and combining the violation information and the law and regulation information into the violation information to be evaluated of the party to be evaluated.
Optionally, the obtaining module is further configured to:
and collecting the capital loss value, the market value loss value and the reputation loss value of the party to be evaluated, and converting the capital loss value, the market value loss value and the reputation loss value into a sample asset loss value according to a preset quantization rule.
Optionally, the multidimensional information to be evaluated includes at least one piece of information to be evaluated;
the prediction module further configured to:
inputting the at least one piece of information to be evaluated into the loss evaluation model;
the resource loss evaluation model carries out embedding processing on the at least one piece of information to be evaluated to obtain at least one piece of vector to be evaluated;
obtaining a predicted capital loss value, a predicted market value loss value and a predicted reputation loss value of the party to be evaluated based on the at least one vector to be evaluated;
and forming a predicted asset loss value output by the asset loss evaluation model according to the predicted capital loss value, the predicted market value loss value and the predicted reputation loss value.
Optionally, the multidimensional information to be evaluated includes at least one piece of information to be evaluated, and each piece of information to be evaluated corresponds to an evaluation weight;
the prediction module further configured to:
and inputting each information to be evaluated and the corresponding evaluation weight into the resource loss evaluation model.
Optionally, the training stop condition includes:
the model loss value is smaller than a preset threshold value; or
The training round reaches the preset training round.
According to a fourth aspect of embodiments herein, there is provided a damage assessment apparatus including:
a determination module configured to determine a party to be evaluated;
the acquisition module is configured to acquire multi-dimensional information to be evaluated of the party to be evaluated;
and the prediction module is configured to input the multidimensional information to be evaluated into a resource loss evaluation model for processing, and obtain an asset loss value output by the resource loss evaluation model, wherein the resource loss evaluation model is obtained by training through a training method of the resource loss evaluation model.
According to a fifth aspect of embodiments herein, there is provided a computing device comprising a memory, a processor and computer instructions stored on the memory and executable on the processor, the processor implementing the method for training the damage assessment model or the steps of the method for assessing damage when executing the computer instructions.
According to a sixth aspect of embodiments herein, there is provided a computer-readable storage medium storing computer instructions which, when executed by a processor, implement the steps of the method for training the damage assessment model or the method for assessing damage.
According to a seventh aspect of embodiments herein, there is provided a computer program, wherein when the computer program is executed in a computer, the computer program causes the computer to execute the steps of the aforementioned method for training the damage assessment model or the damage assessment method.
The method for training the asset loss assessment model, provided by the specification, acquires multi-dimensional information to be assessed of a party to be assessed and a sample asset loss value of the party to be assessed; inputting the multi-dimensional information to be evaluated into a resource loss evaluation model, and obtaining a predicted asset loss value output by the resource loss evaluation model; calculating a model loss value from the sample asset loss value and the predicted asset loss value; and adjusting model parameters of the resource loss evaluation model based on the model loss value, and continuing to train the resource loss evaluation model until a training stop condition is reached.
The training method for the resource loss assessment model provided in an embodiment of the specification realizes acquisition of multidimensional assessment information of a party to be assessed, inputs the multidimensional assessment information of the party to be assessed into the resource loss assessment model to be trained for processing, performs comprehensive calculation in the resource loss assessment model according to a characteristic value of assessment information of each dimension, outputs a predicted asset loss value, compares the predicted asset loss value with a sample asset loss value, and further continues training the model until a training stop condition is reached, so that an accurate and efficient resource loss assessment model can be trained.
Drawings
FIG. 1 is a flow chart of a method for training a damage-assessment model according to an embodiment of the present disclosure;
FIG. 2 is a flow chart of a method for asset loss assessment according to an embodiment of the present disclosure;
FIG. 3 is a schematic structural diagram of a training apparatus for a damage assessment model according to an embodiment of the present disclosure;
fig. 4 is a schematic structural diagram of a damage assessment apparatus according to an embodiment of the present disclosure;
fig. 5 is a block diagram of a computing device according to an embodiment of the present disclosure.
Detailed Description
In the following description, numerous specific details are set forth in order to provide a thorough understanding of the present specification. This description may be embodied in many different forms and should not be construed as limited to the embodiments set forth herein, as those skilled in the art will be able to make and use the present disclosure without departing from the spirit and scope of the present disclosure.
The terminology used in the description of the one or more embodiments is for the purpose of describing the particular embodiments only and is not intended to be limiting of the description of the one or more embodiments. As used in one or more embodiments of the present specification and the appended claims, the singular forms "a," "an," and "the" are intended to include the plural forms as well, unless the context clearly indicates otherwise. It should also be understood that the term "and/or" as used in one or more embodiments of the present specification refers to and encompasses any and all possible combinations of one or more of the associated listed items.
It will be understood that, although the terms first, second, etc. may be used herein in one or more embodiments to describe various information, these information should not be limited by these terms. These terms are only used to distinguish one type of information from another. For example, a first can also be referred to as a second and, similarly, a second can also be referred to as a first without departing from the scope of one or more embodiments of the present description. The word "if," as used herein, may be interpreted as "at … …" or "when … …" or "in response to a determination," depending on the context.
In this specification, a method of training a damage assessment model is provided. One or more embodiments of the present disclosure also relate to a training apparatus for a damage assessment model, a damage assessment method and apparatus, a computing device, a computer-readable storage medium, and a computer program, which are described in detail in the following embodiments.
Fig. 1 shows a flowchart of a method for training a damage assessment model according to an embodiment of the present disclosure, which includes steps 102 to 108.
Step 102: obtaining multi-dimensional information to be evaluated of a party to be evaluated and a sample asset loss value of the party to be evaluated.
The party to be evaluated is the subject of the asset loss to be evaluated, for example, if the asset loss evaluation needs to be performed on a company a, the company a is the party to be evaluated, and if the asset loss evaluation is performed on an organization B, the organization B is the party to be evaluated.
The multidimensional information to be evaluated specifically refers to information of various dimensions related to the party to be evaluated, such as customer scale information of a company, supplier scale information, fund information, violation information, legal and regulation information, and the like.
The sample asset loss value refers to the asset loss value of the party to be evaluated due to a certain action, for example, a certain company loses $ 100 ten thousand due to financial counterfeiting, and the like.
In practical application, the types of the multidimensional information to be evaluated of the party to be evaluated are many, and the obtaining of the multidimensional information to be evaluated of the party to be evaluated comprises the following steps:
and obtaining the information of the to-be-evaluated body quantity and the to-be-evaluated violation information of the to-be-evaluated party.
The information of the body quantity to be evaluated refers to scale information corresponding to the party to be evaluated, and the information of the violation to be evaluated specifically refers to information that the party to be evaluated violates a relevant legal regulation.
Specifically, obtaining the information of the to-be-evaluated body size and the information of the to-be-evaluated violation of a rule of the to-be-evaluated party includes:
acquiring information of a party to be evaluated, customer information and fund information of the party to be evaluated, and forming the information of the party to be evaluated, the customer information and the fund information into information of the amount of the party to be evaluated;
and acquiring violation information and law and regulation information of the party to be evaluated, and combining the violation information and the law and regulation information into the violation information to be evaluated of the party to be evaluated.
Acquiring the information of the to-be-evaluated amount of the to-be-evaluated party, including acquiring the information of the to-be-evaluated party, client information, fund information and the like of the to-be-evaluated party, wherein the information of the to-be-evaluated party includes the company amount of the to-be-evaluated party, such as the number of company employees, the years of the company employees, the distribution of the company employees and the like; the customer information comprises the customer number of the party to be evaluated, customer size information, supplier number, supplier size information and the like; the fund information includes floating fund information, fixed asset information, loan information, etc. of the party to be assessed. And the collected information of the party to be evaluated, the client information, the fund information and the like form the information of the volume to be evaluated of the party to be evaluated.
Acquiring violation information to be evaluated of a party to be evaluated, wherein the violation information comprises violation behavior information, negative public opinion information and the like; the legal and legal information specifically refers to relevant compliant legal documents, and the like. And the collected violation behavior information, law and regulation information and the like form violation information to be evaluated of the party to be evaluated.
In practical application, obtaining a sample asset loss value of a party to be evaluated comprises:
and collecting the capital loss value, the market value loss value and the reputation loss value of the party to be evaluated, and converting the capital loss value, the market value loss value and the reputation loss value into a sample asset loss value according to a preset quantization rule.
When the to-be-evaluated party generates the asset loss, the asset loss may be in multiple aspects, for example, there may be direct economic loss, indirect market loss, reputation loss, and the like, and therefore, the loss values of multiple dimensions of the to-be-evaluated party, such as the capital loss value, the market loss value, the reputation loss value, and the like, are collected, and the loss values of the above dimensions are converted into sample asset loss values through specific quantitative conversion rules, for example, a company has a capital loss value of 50 ten thousand dollars due to financial counterfeiting, a market loss value of 2 percentage points of stock drop, a reputation loss value of three cooperative contracts loss, and then the conversion rules are added to convert the market loss value of 2 percentage points of stock drop into a loss of 30 ten thousand dollars, and a loss of 20 ten thousand dollars loss of three cooperative contracts loss, and then the asset loss value of 50 ten thousand dollars +30 ten thousand dollars +20 million dollars of the to-be-evaluated party, for a total of 100 million dollars.
It should be noted that the multidimensional information to be evaluated and the asset loss value of the sample provided in the present specification are correspondingly presented, and both of the multidimensional information to be evaluated and the asset loss value serve as training samples of the asset loss evaluation model.
In a specific embodiment provided in this specification, taking an evaluation-waiting party as a project platform C as an example, project platform information 1, customer scale information 2, supplier scale information 3, platform fund information 4, platform violation behavior 5, and legal regulation information 6 of the project platform C are collected, and meanwhile, a sample asset loss value of the project platform C is obtained as 230 ten thousand renminbi. And taking the collected information to be evaluated of each dimension and the sample asset loss value as sample data of the asset loss evaluation model.
Step 104: and inputting the multi-dimensional information to be evaluated into a resource loss evaluation model, and obtaining a predicted asset loss value output by the resource loss evaluation model.
The resource loss evaluation model is suitable for an artificial intelligence model for calculating the asset loss value according to the multi-dimensional information to be evaluated of the party to be evaluated, receives the multi-dimensional information to be evaluated of the party to be evaluated, extracts the characteristic value in each piece of the multi-dimensional information to be evaluated, carries out comprehensive calculation, and finally outputs the predicted asset loss value of the party to be evaluated.
Specifically, the multidimensional information to be evaluated comprises at least one piece of information to be evaluated;
inputting the multi-dimensional information to be evaluated into a resource loss evaluation model, and acquiring a predicted asset loss value output by the resource loss evaluation model, wherein the method comprises the following steps:
inputting the at least one piece of information to be evaluated into the loss evaluation model;
the resource loss evaluation model carries out embedding processing on the at least one piece of information to be evaluated to obtain at least one piece of vector to be evaluated;
obtaining a predicted capital loss value, a predicted market value loss value and a predicted reputation loss value of the party to be evaluated based on the at least one vector to be evaluated;
and forming a predicted asset loss value output by the asset loss evaluation model according to the predicted capital loss value, the predicted market value loss value and the predicted reputation loss value.
The multi-dimensional information to be evaluated generally comprises a plurality of pieces of information to be evaluated, such as information of a party to be evaluated, a client, fund, violation information, law and regulation information, of the party to be evaluated, the plurality of pieces of information to be evaluated are input into a resource loss evaluation model for processing, each piece of information to be evaluated is embedded in the resource loss evaluation model to obtain a corresponding vector to be evaluated, loss values of the party to be evaluated in multiple dimensions, such as a predicted fund loss value, a predicted market value loss value and a predicted commodity loss value, are obtained according to each vector to be evaluated, the loss values of the multiple dimensions are added to obtain a final predicted asset loss value, and the predicted asset loss value is used as a final output result of the resource loss evaluation model.
In another specific embodiment provided in this specification, the multidimensional information to be evaluated includes at least one piece of information to be evaluated, and each piece of information to be evaluated corresponds to an evaluation weight;
inputting the multi-dimensional information to be evaluated into a resource loss evaluation model, wherein the method comprises the following steps:
and inputting each information to be evaluated and the corresponding evaluation weight into the resource loss evaluation model.
In practical applications, different pieces of information to be evaluated have different degrees of importance, and therefore, each piece of information to be evaluated corresponds to one evaluation weight, and for the more important information to be evaluated, the corresponding evaluation weight is higher, for example, if violation information in the information to be evaluated is concerned, the evaluation weight is higher, and for the fund information, the evaluation weight is relatively lower, and the like.
Correspondingly, when the multi-dimensional information to be evaluated is input into the resource loss evaluation model, each piece of information to be evaluated and the corresponding evaluation weight are input into the resource loss evaluation model, and the resource loss evaluation model can predict the predicted asset loss value of the party to be evaluated according to each piece of information to be evaluated and the corresponding evaluation weight.
In a specific embodiment provided in this specification, following the above example, the collected project platform information 1, customer scale information 2, supplier scale information 3, platform capital information 4, platform violation behavior 5, legal and regulatory information 6, etc. are input to the damage assessment model, the damage assessment model predicts based on the above information, and outputs a predicted asset loss value of 200 ten thousand renminbi.
Step 106: calculating a model loss value based on the sample asset loss value and the predicted asset loss value.
The resource loss evaluation model can calculate a predicted asset loss value according to the multi-dimensional information to be evaluated, the resource loss evaluation model is not trained, the obtained measured asset loss value is not accurate enough, and the model loss value needs to be compared with a sample asset loss value in a training sample to calculate the model loss value.
The loss function is a function which maps the value of a random event or a random variable related to the random event into a non-negative real number to represent the risk or loss of the random event, the value calculated by the loss function is a loss value, and the loss function is usually used as a learning standard test and is associated with an optimization problem and is used for parameter estimation of a model in machine learning.
There are many loss functions for calculating the loss value, such as cross entropy loss function, maximum loss function, average loss function, etc., and in this specification, the specific manner of the loss function is not limited, and is subject to practical application.
In one embodiment provided herein, following the above example, the asset loss assessment model outputs a predicted asset loss value of 200 ten thousand RMB, while the sample asset loss value is 230 ten thousand RMB. And calculating a model Loss value Loss of the predicted asset Loss value and the sample asset Loss value according to the cross entropy Loss function.
Step 108: and adjusting model parameters of the resource loss evaluation model based on the model loss value, and continuing to train the resource loss evaluation model until a training stop condition is reached.
After the model loss value is obtained, the asset loss evaluation model can be adjusted according to the model loss value, specifically, the model loss value is subjected to back propagation to sequentially update model parameters in the asset loss evaluation model, and the model parameters are used for assisting the asset loss evaluation model to generate a corresponding prediction asset loss value according to the multi-dimensional information to be evaluated.
After the model parameters are adjusted, the above steps can be continuously repeated, and the training of the resource loss assessment model is continuously performed until a training stop condition is reached, wherein in practical application, the training stop condition of the resource loss assessment model includes:
the model loss value is smaller than a preset threshold value; or
The training round reaches the preset training round.
Specifically, in the process of training the resource loss assessment model, the training stop condition of the model may be set to be that the model loss value is smaller than the preset threshold, or the training stop condition may be set to be that the training round is the preset training round, for example, 10 rounds of training.
In a specific embodiment provided in this specification, following the above example, the model parameters of the resource Loss assessment model are adjusted according to the model Loss value Loss, and the resource Loss assessment model is continuously trained by using the next set of sample data until the model Loss value of the resource Loss assessment model is smaller than the preset threshold, so that the trained resource Loss assessment model is obtained, and the resource Loss assessment model can predict the resource Loss value of the party to be assessed according to the multidimensional information to be assessed of the party to be assessed.
The method for training the asset loss assessment model, provided by the specification, acquires multi-dimensional information to be assessed of a party to be assessed and a sample asset loss value of the party to be assessed; inputting the multi-dimensional information to be evaluated into a resource loss evaluation model, and obtaining a predicted asset loss value output by the resource loss evaluation model; calculating a model loss value from the sample asset loss value and the predicted asset loss value; and adjusting model parameters of the resource loss evaluation model based on the model loss value, and continuing to train the resource loss evaluation model until a training stop condition is reached. The training method for the resource loss assessment model provided in an embodiment of the specification realizes acquisition of multidimensional assessment information of a party to be assessed, inputs the multidimensional assessment information of the party to be assessed into the resource loss assessment model to be trained for processing, performs comprehensive calculation in the resource loss assessment model according to a characteristic value of assessment information of each dimension, outputs a predicted asset loss value, compares the predicted asset loss value with a sample asset loss value, and further continues training the model until a training stop condition is reached, so that an accurate and efficient resource loss assessment model can be trained.
Fig. 2 is a flowchart illustrating a method for assessing damage according to an embodiment of the present disclosure, including steps 202 to 206.
Step 202: and determining the party to be evaluated.
The party to be assessed is the subject of the loss of the asset to be assessed, and can be generally a company, an organization and the like.
In a specific embodiment provided in this specification, if the party to be evaluated is taken as the company T for explanation, it is determined that the party to be evaluated is the company T.
Step 204: and acquiring multi-dimensional information to be evaluated of the party to be evaluated.
In a specific embodiment provided in the present specification, the above example is adopted, and staff scale information, company scale information, customer scale information, company reserve fund information, company violation information, and legal regulation information of the company T are collected.
Step 206: and inputting the multi-dimensional information to be evaluated into a resource loss evaluation model for processing, and obtaining an asset loss value output by the resource loss evaluation model, wherein the resource loss evaluation model is obtained by training through the resource loss evaluation model training method.
In a specific embodiment provided in this specification, following the above example, the collected employee scale information, company scale information, customer scale information, company reserve fund information, company violation information, and legal regulation information are input into a trained investment and loss assessment model, and the investment and loss assessment model is obtained by training through the investment and loss assessment model.
And analyzing and calculating the information by the asset loss evaluation model to obtain a capital loss value of M1, a market loss value of M2 and a reputation loss value of M3, and adding the three types of loss values to obtain an asset loss value M output by the model, wherein M is M1+ M2+ M3.
The resource loss assessment method provided by one embodiment of the specification comprises the steps of determining a to-be-assessed party, obtaining multi-dimensional information to be assessed of the to-be-assessed party, inputting the multi-dimensional information to be assessed into a resource loss assessment model for processing, and obtaining an asset loss value output by the resource loss assessment model.
Corresponding to the above embodiment of the method for training the resource loss assessment model, the present specification further provides an embodiment of a training apparatus for the resource loss assessment model, and fig. 3 illustrates a schematic structural diagram of a training apparatus for the resource loss assessment model provided in an embodiment of the present specification. As shown in fig. 3, the apparatus includes:
an obtaining module 302 configured to obtain multidimensional information to be evaluated of a party to be evaluated and a sample asset loss value of the party to be evaluated;
the prediction module 304 is configured to input the multidimensional information to be evaluated into a resource loss evaluation model, and obtain a predicted asset loss value output by the resource loss evaluation model;
a calculation module 306 configured to calculate a model loss value based on the sample asset loss value and the predicted asset loss value;
a training module 308 configured to adjust model parameters of the asset assessment model based on the model loss value, and continue training the asset assessment model until a training stop condition is reached.
Optionally, the obtaining module 302 is further configured to:
and obtaining the information of the to-be-evaluated body quantity and the to-be-evaluated violation information of the to-be-evaluated party.
Optionally, the obtaining module 302 is further configured to:
acquiring information of a party to be evaluated, customer information and fund information of the party to be evaluated, and forming the information of the party to be evaluated, the customer information and the fund information into information of the amount of the party to be evaluated;
and acquiring violation information and law and regulation information of the party to be evaluated, and combining the violation information and the law and regulation information into the violation information to be evaluated of the party to be evaluated.
Optionally, the obtaining module 302 is further configured to:
and collecting the capital loss value, the market value loss value and the reputation loss value of the party to be evaluated, and converting the capital loss value, the market value loss value and the reputation loss value into a sample asset loss value according to a preset quantization rule.
Optionally, the multidimensional information to be evaluated includes at least one piece of information to be evaluated;
the prediction module 304, further configured to:
inputting the at least one piece of information to be evaluated into the loss evaluation model;
the resource loss evaluation model carries out embedding processing on the at least one piece of information to be evaluated to obtain at least one piece of vector to be evaluated;
obtaining a predicted fund loss value, a predicted market value loss value and a predicted reputation loss value of the party to be evaluated based on the at least one vector to be evaluated;
and forming a predicted asset loss value output by the asset loss evaluation model according to the predicted capital loss value, the predicted market value loss value and the predicted reputation loss value.
Optionally, the multidimensional information to be evaluated includes at least one piece of information to be evaluated, and each piece of information to be evaluated corresponds to an evaluation weight;
the prediction module 304, further configured to:
and inputting each information to be evaluated and the corresponding evaluation weight into the resource loss evaluation model.
Optionally, the training stop condition includes:
the model loss value is smaller than a preset threshold value; or
The training round reaches the preset training round.
The training device of the asset loss assessment model provided by the specification acquires multi-dimensional information to be assessed of a party to be assessed and a sample asset loss value of the party to be assessed; inputting the multi-dimensional information to be evaluated into a resource loss evaluation model, and obtaining a predicted asset loss value output by the resource loss evaluation model; calculating a model loss value from the sample asset loss value and the predicted asset loss value; and adjusting model parameters of the resource loss evaluation model based on the model loss value, and continuing to train the resource loss evaluation model until a training stop condition is reached. The training device for the resource loss assessment model provided in an embodiment of the present specification realizes acquisition of multidimensional assessment information of a party to be assessed, inputs the multidimensional assessment information of the party to be assessed into the resource loss assessment model to be trained for processing, performs comprehensive calculation in the resource loss assessment model according to a characteristic value of assessment information of each dimension, outputs a predicted asset loss value, compares the predicted asset loss value with a sample asset loss value, and further continues training the model until a training stop condition is reached, so that an accurate and efficient resource loss assessment model can be trained.
The above is an exemplary scheme of a training apparatus for a resource loss assessment model according to this embodiment. It should be noted that the technical solution of the training apparatus for the damage assessment model and the technical solution of the training method for the damage assessment model belong to the same concept, and the details of the technical solution of the training apparatus for the damage assessment model, which are not described in detail, can be referred to the description of the technical solution of the training method for the damage assessment model.
Corresponding to the above-mentioned asset loss assessment method embodiment, the present specification further provides an asset loss assessment apparatus embodiment, and fig. 4 shows a schematic structural diagram of an asset loss assessment apparatus provided in an embodiment of the present specification. As shown in fig. 4, the apparatus includes:
a determination module 402 configured to determine a party to be evaluated;
an obtaining module 404 configured to obtain multidimensional information to be evaluated of the party to be evaluated;
the prediction module 406 is configured to input the multidimensional information to be evaluated into a resource and loss evaluation model for processing, and obtain an asset loss value output by the resource and loss evaluation model, where the resource and loss evaluation model is obtained by training through any one of the above-mentioned training methods of the resource and loss evaluation model.
The resource loss evaluation device provided by one embodiment of the specification comprises a to-be-evaluated party, to-be-evaluated multidimensional information of the to-be-evaluated party is obtained, the to-be-evaluated multidimensional information is input into a resource loss evaluation model to be processed, and an asset loss value output by the resource loss evaluation model is obtained.
The above is an exemplary scheme of a loss evaluation apparatus according to this embodiment. It should be noted that the technical solution of the loss assessment apparatus and the technical solution of the loss assessment method described above belong to the same concept, and details of the technical solution of the loss assessment apparatus, which are not described in detail, can be referred to the description of the technical solution of the loss assessment method described above.
Fig. 5 illustrates a block diagram of a computing device 500 provided according to an embodiment of the present description. The components of the computing device 500 include, but are not limited to, a memory 510 and a processor 520. Processor 520 is coupled to memory 510 via bus 530, and database 550 is used to store data.
Computing device 500 also includes access device 540, access device 540 enabling computing device 500 to communicate via one or more networks 560. Examples of such networks include the Public Switched Telephone Network (PSTN), a Local Area Network (LAN), a Wide Area Network (WAN), a Personal Area Network (PAN), or a combination of communication networks such as the internet. The access device 540 may include one or more of any type of network interface, e.g., a Network Interface Card (NIC), wired or wireless, such as an IEEE802.11 Wireless Local Area Network (WLAN) wireless interface, a worldwide interoperability for microwave access (Wi-MAX) interface, an ethernet interface, a Universal Serial Bus (USB) interface, a cellular network interface, a bluetooth interface, a Near Field Communication (NFC) interface, and so forth.
In one embodiment of the present description, the above-described components of computing device 500, as well as other components not shown in FIG. 5, may also be connected to each other, such as by a bus. It should be understood that the block diagram of the computing device architecture shown in FIG. 5 is for purposes of example only and is not limiting as to the scope of the present description. Those skilled in the art may add or replace other components as desired.
Computing device 500 may be any type of stationary or mobile computing device, including a mobile computer or mobile computing device (e.g., tablet computer, personal digital assistant, laptop computer, notebook computer, netbook, etc.), mobile phone (e.g., smartphone), wearable computing device (e.g., smartwatch, smart glasses, etc.), or other type of mobile device, or a stationary computing device such as a desktop computer or PC. Computing device 500 may also be a mobile or stationary server.
Wherein, the processor 520, when executing the computer instructions, implements the steps of the method for training the resource assessment model or the method for assessing the resource.
The foregoing is a schematic diagram of a computing device of the present embodiment. It should be noted that the technical solution of the computing device and the above-mentioned technical solution of the method for training the resource assessment model or the method for assessing the resource loss belong to the same concept, and details of the technical solution of the computing device, which are not described in detail, can be referred to the description of the technical solution of the method for training the resource assessment model or the method for assessing the resource loss.
An embodiment of the present specification further provides a computer readable storage medium, which stores computer instructions, and when executed by a processor, the computer instructions implement the method for training the damage assessment model or the steps of the damage assessment method as described above.
The above is an illustrative scheme of a computer-readable storage medium of the present embodiment. It should be noted that the technical solution of the storage medium is the same as the technical solution of the aforementioned training method of the damage assessment model or the damage assessment method, and the details of the technical solution of the storage medium, which are not described in detail, can be referred to the description of the technical solution of the aforementioned training method of the damage assessment model or the damage assessment method.
An embodiment of the present specification further provides a computer program, wherein when the computer program is executed in a computer, the computer program causes the computer to execute the method for training the resource assessment model or the steps of the resource assessment method.
The above is an illustrative scheme of a computer program of the present embodiment. It should be noted that the technical solution of the computer program is the same as the technical solution of the aforementioned training method of the damage assessment model or the damage assessment method, and the details of the technical solution of the computer program, which are not described in detail, can be referred to the description of the technical solution of the aforementioned training method of the damage assessment model or the damage assessment method.
The foregoing description has been directed to specific embodiments of this disclosure. Other embodiments are within the scope of the following claims. In some cases, the actions or steps recited in the claims may be performed in a different order than in the embodiments and still achieve desirable results. In addition, the processes depicted in the accompanying figures do not necessarily require the particular order shown, or sequential order, to achieve desirable results. In some embodiments, multitasking and parallel processing may also be possible or may be advantageous.
The computer instructions comprise computer program code which may be in source code form, object code form, an executable file or some intermediate form, or the like. The computer-readable medium may include: any entity or device capable of carrying the computer program code, recording medium, usb disk, removable hard disk, magnetic disk, optical disk, computer Memory, Read-Only Memory (ROM), Random Access Memory (RAM), electrical carrier wave signals, telecommunications signals, software distribution medium, etc. It should be noted that the computer readable medium may contain content that is subject to appropriate increase or decrease as required by legislation and patent practice in jurisdictions, for example, in some jurisdictions, computer readable media does not include electrical carrier signals and telecommunications signals as is required by legislation and patent practice.
It should be noted that, for the sake of simplicity, the foregoing method embodiments are described as a series of acts, but those skilled in the art should understand that the present embodiment is not limited by the described acts, because some steps may be performed in other sequences or simultaneously according to the present embodiment. Further, those skilled in the art should also appreciate that the embodiments described in this specification are preferred embodiments and that acts and modules referred to are not necessarily required for an embodiment of the specification.
In the foregoing embodiments, the descriptions of the respective embodiments have respective emphasis, and for parts that are not described in detail in a certain embodiment, reference may be made to the related descriptions of other embodiments.
The preferred embodiments of the present specification disclosed above are intended only to aid in the description of the specification. Alternative embodiments are not exhaustive and do not limit the invention to the precise embodiments described. Obviously, many modifications and variations are possible in light of the teaching of the embodiments of the present disclosure. The embodiments were chosen and described in order to best explain the principles of the embodiments and the practical application, to thereby enable others skilled in the art to best understand and utilize the embodiments. The specification is limited only by the claims and their full scope and equivalents.
Claims (10)
1. A method for training a loss assessment model comprises the following steps:
obtaining multi-dimensional information to be evaluated of a party to be evaluated and a sample asset loss value of the party to be evaluated;
inputting the multi-dimensional information to be evaluated into a resource loss evaluation model, and obtaining a predicted asset loss value output by the resource loss evaluation model;
calculating a model loss value from the sample asset loss value and the predicted asset loss value;
and adjusting model parameters of the resource loss evaluation model based on the model loss value, and continuing to train the resource loss evaluation model until a training stop condition is reached.
2. The method for training the asset impairment assessment model according to claim 1, wherein the obtaining of the multidimensional information to be assessed of the party to be assessed includes:
and obtaining the information of the to-be-evaluated body quantity and the to-be-evaluated violation information of the to-be-evaluated party.
3. The method for training the resource loss assessment model of claim 2, wherein the obtaining of the information of the amount to be assessed and the information of the violation to be assessed comprises:
acquiring information of a party to be evaluated, customer information and fund information of the party to be evaluated, and forming the information of the party to be evaluated, the customer information and the fund information into information of the amount of the party to be evaluated;
and acquiring violation information and law and regulation information of the party to be evaluated, and combining the violation information and the law and regulation information into the violation information to be evaluated of the party to be evaluated.
4. The method for training the asset loss assessment model according to claim 1, wherein the step of obtaining the sample asset loss value of the party to be assessed comprises:
and collecting the capital loss value, the market value loss value and the reputation loss value of the party to be evaluated, and converting the capital loss value, the market value loss value and the reputation loss value into a sample asset loss value according to a preset quantization rule.
5. The method for training a damage assessment model according to claim 1, wherein the multidimensional information to be assessed includes at least one piece of information to be assessed;
inputting the multi-dimensional information to be evaluated into a resource loss evaluation model, and acquiring a predicted asset loss value output by the resource loss evaluation model, wherein the method comprises the following steps:
inputting the at least one piece of information to be evaluated into the loss evaluation model;
the resource loss evaluation model carries out embedding processing on the at least one piece of information to be evaluated to obtain at least one piece of vector to be evaluated;
obtaining a predicted capital loss value, a predicted market value loss value and a predicted reputation loss value of the party to be evaluated based on the at least one vector to be evaluated;
and forming a predicted asset loss value output by the asset loss evaluation model according to the predicted capital loss value, the predicted market value loss value and the predicted reputation loss value.
6. The method for training a resource loss assessment model according to claim 1, wherein the multidimensional information to be assessed includes at least one piece of information to be assessed, and each piece of information to be assessed corresponds to an assessment weight;
inputting the multi-dimensional information to be evaluated into a resource loss evaluation model, wherein the method comprises the following steps:
and inputting each information to be evaluated and the corresponding evaluation weight into the resource loss evaluation model.
7. The method of claim 1, wherein the training-stopping condition comprises:
the model loss value is smaller than a preset threshold value; or
The training round reaches the preset training round.
8. A method of asset impairment assessment, comprising:
determining a party to be evaluated;
acquiring multi-dimensional information to be evaluated of the party to be evaluated;
inputting the multidimensional information to be evaluated into a resource and loss evaluation model for processing, and obtaining an asset loss value output by the resource and loss evaluation model, wherein the resource and loss evaluation model is obtained by training according to the training method of any one of claims 1 to 7.
9. A training apparatus for a loss assessment model, comprising:
the system comprises an acquisition module, a storage module and a processing module, wherein the acquisition module is configured to acquire multi-dimensional information to be evaluated of a party to be evaluated and a sample asset loss value of the party to be evaluated;
the prediction module is configured to input the multi-dimensional information to be evaluated into a resource loss evaluation model and obtain a predicted asset loss value output by the resource loss evaluation model;
a calculation module configured to calculate a model loss value from the sample asset loss value and the predicted asset loss value;
a training module configured to adjust model parameters of the resource damage assessment model based on the model loss value, and continue training the resource damage assessment model until a training stop condition is reached.
10. An asset impairment assessment apparatus comprising:
a determination module configured to determine a party to be evaluated;
the acquisition module is configured to acquire multi-dimensional information to be evaluated of the party to be evaluated;
the prediction module is configured to input the multidimensional information to be evaluated into a resource loss evaluation model for processing, and obtain an asset loss value output by the resource loss evaluation model, wherein the resource loss evaluation model is obtained by training according to the training method of any one of claims 1 to 7.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202111386073.5A CN114091903A (en) | 2021-11-22 | 2021-11-22 | Training method and device of loss assessment model, and loss assessment method and device |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202111386073.5A CN114091903A (en) | 2021-11-22 | 2021-11-22 | Training method and device of loss assessment model, and loss assessment method and device |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN114091903A true CN114091903A (en) | 2022-02-25 |
Family
ID=80302761
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202111386073.5A Pending CN114091903A (en) | 2021-11-22 | 2021-11-22 | Training method and device of loss assessment model, and loss assessment method and device |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN114091903A (en) |
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN115841346A (en) * | 2023-02-23 | 2023-03-24 | 杭银消费金融股份有限公司 | Asset derating prediction method and system for business decisions |
Citations (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN110348615A (en) * | 2019-06-27 | 2019-10-18 | 西安工程大学 | Cable line fault probability forecasting method based on ant group optimization support vector machines |
| CN110991441A (en) * | 2019-12-13 | 2020-04-10 | 王文斌 | Asset assessment method and device based on image recognition and computer storage medium |
-
2021
- 2021-11-22 CN CN202111386073.5A patent/CN114091903A/en active Pending
Patent Citations (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN110348615A (en) * | 2019-06-27 | 2019-10-18 | 西安工程大学 | Cable line fault probability forecasting method based on ant group optimization support vector machines |
| CN110991441A (en) * | 2019-12-13 | 2020-04-10 | 王文斌 | Asset assessment method and device based on image recognition and computer storage medium |
Non-Patent Citations (1)
| Title |
|---|
| 王璐;肖先勇;汪颖;刘阳;: "基于深度神经网络的电压暂降经济损失评估模型", 电力自动化设备, no. 06, 30 June 2020 (2020-06-30), pages 156 - 161 * |
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN115841346A (en) * | 2023-02-23 | 2023-03-24 | 杭银消费金融股份有限公司 | Asset derating prediction method and system for business decisions |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN109816221B (en) | Project risk decision method, apparatus, computer device and storage medium | |
| CN111401788B (en) | Attribution method and device of service timing sequence index | |
| CN107451596B (en) | Network node classification method and device | |
| CN112561082A (en) | Method, device, equipment and storage medium for generating model | |
| CN113743991A (en) | Life cycle value prediction method and device | |
| CN113486302A (en) | Data processing method and device | |
| CN110555749B (en) | Credit behavior prediction method and device based on neural network | |
| CN114066405A (en) | Data processing method and device | |
| CN114091903A (en) | Training method and device of loss assessment model, and loss assessment method and device | |
| CN114202174A (en) | Electricity price risk grade early warning method and device and storage medium | |
| CN111222631A (en) | Service prediction method and device | |
| CN113569059A (en) | Target user identification method and device | |
| CN115271282A (en) | Customer value determination method and device based on fuzzy logic | |
| CN110796450B (en) | Trusted relationship processing method and device | |
| CN109884548B (en) | Method for predicting remaining life of lithium battery based on GASVM-AUKF algorithm | |
| CN110633971A (en) | Method and device for estimating loss | |
| CN109241249B (en) | Method and device for determining burst problem | |
| CN116757835A (en) | Method and device for monitoring transaction risk in credit card customer credit | |
| CN114091908A (en) | Power distribution network comprehensive evaluation method, device and equipment considering multi-mode energy storage station | |
| KR20220164149A (en) | Method for predicting depression occurrence using artificial intelligence model and computer readable record medium thereof | |
| CN112365189A (en) | Case distribution method and device | |
| Jadhav et al. | Data transformation: A preprocessing stage in machine learning regression problems | |
| CN111160662A (en) | Risk prediction method, electronic equipment and storage medium | |
| Ayyappa et al. | Forecasting Equity Prices using LSTM and BERT with Sentiment Analysis | |
| CN111340632A (en) | Resource control method and device |
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 |