CN109242361A - A kind of fire-fighting methods of risk assessment, device and terminal device - Google Patents
A kind of fire-fighting methods of risk assessment, device and terminal device Download PDFInfo
- Publication number
- CN109242361A CN109242361A CN201811286629.1A CN201811286629A CN109242361A CN 109242361 A CN109242361 A CN 109242361A CN 201811286629 A CN201811286629 A CN 201811286629A CN 109242361 A CN109242361 A CN 109242361A
- Authority
- CN
- China
- Prior art keywords
- fire
- data
- risk assessment
- statistical data
- characteristic variable
- 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
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
- G06Q10/0635—Risk analysis of enterprise or organisation activities
Landscapes
- Business, Economics & Management (AREA)
- Human Resources & Organizations (AREA)
- Engineering & Computer Science (AREA)
- Strategic Management (AREA)
- Entrepreneurship & Innovation (AREA)
- Economics (AREA)
- Operations Research (AREA)
- Game Theory and Decision Science (AREA)
- Development Economics (AREA)
- Marketing (AREA)
- Educational Administration (AREA)
- Quality & Reliability (AREA)
- Tourism & Hospitality (AREA)
- Physics & Mathematics (AREA)
- General Business, Economics & Management (AREA)
- General Physics & Mathematics (AREA)
- Theoretical Computer Science (AREA)
- Management, Administration, Business Operations System, And Electronic Commerce (AREA)
Abstract
The present invention is suitable for fire-fighting early warning technology field, provides a kind of fire-fighting methods of risk assessment, device and terminal device, wherein method includes: acquisition statistical data, obtains data sample according to statistical data;Risk assessment grade corresponding with social synthesis's statistical data of the same period is established according to Fire Data, the variable characterized by the statistical indicator of social synthesis's statistical data establishes training sample by desired output result of risk assessment grade;Characteristic variable is screened according to random forests algorithm and establishes index set;Training sample set is generated according to index set and establishes machine learning model, and pre-training is carried out to machine learning model by training sample set, obtains Optimized model;According to index set acquisition index score value and Optimized model is inputted, obtains the assessment result of Optimized model output.The present invention can identify dangerous material in monitored environment and obtain its corresponding risk assessment grade, and prevent an accident fire incident, improve fire prevention and treatment efficiency.
Description
Technical field
The invention belongs to fire-fighting early warning technology fields more particularly to a kind of fire-fighting methods of risk assessment, device and terminal to set
It is standby.
Background technique
Some researches show that security against fire is closely bound up with socio-economic development, such as the density of population is higher, and fire occurs
Probability is often higher, and harm caused by fire also tends to bigger.Therefore, capable field technique personnel are according to these density of population etc.
Statistical indicator establishes fire-fighting risk forecast model and carries out fire-fighting risk assessment.
However, the system for statistical indices for establishing existing fire-fighting risk forecast model is often by expert according to fire-fighting wind
What the result of dangerous factor analysis was established.Since fire-fighting risk Factor Analysis depends on subjective judgement, the choosing to statistical indicator
Objective basis may be lacked by selecting, and be easy to be influenced by subjective factor, caused in data sample that there are much noise interference.According to
The fire-fighting risk forecast model unstable result that data sample is established, is easy to appear large error.
Summary of the invention
In view of this, the embodiment of the invention provides a kind of fire-fighting methods of risk assessment, device and terminal device, to solve
Problems of the prior art.
The first aspect of the embodiment of the present invention provides a kind of fire-fighting methods of risk assessment, comprising:
Statistical data is obtained, and data sample is obtained according to the statistical data;Wherein, the statistical data includes society
Benchmark statistice and Fire Data;
Risk assessment corresponding with social synthesis's statistical data of the same period is established according to the Fire Data
Grade, using the statistical indicator of social synthesis's statistical data as characteristic variable, using the risk assessment grade as expectation
Output is as a result, and establish training sample according to the characteristic variable and the desired output result;
The characteristic variable is screened according to random forests algorithm, and is built according to the characteristic variable screened
Vertical index set;
Training sample set is generated according to the index set and establishes machine learning model, by the training sample set to institute
It states machine learning model and carries out pre-training, to obtain Optimized model;
The Optimized model is inputted according to the index set acquisition index score value, and by the index score value, to obtain
State the assessment result of Optimized model output.
Optionally, the acquisition statistical data, and the statistical data is pre-processed, to obtain data sample, packet
It includes:
Obtain statistical data;
To the statistical data carry out missing values analysis, outlier detection, repeated data analysis, the analysis of inconsistent value and
Value analysis containing additional character;
Missing values described in polishing, and reject the exceptional value, the repeated data, the inconsistent value and described contain
The value of additional character, to obtain data sample.
Optionally, risk assessment corresponding with social synthesis's statistical data etc. is established according to the Fire Data
Grade the step of include:
The Fire Data is analyzed according to the parameter in the Fire Data;Wherein, the fire
Parameter in statistical data includes fire frequency, the number of casualties, economic loss, influence of fire area;
Obtain the weight of each parameter in the Fire Data;
According to the weight of each parameter in the parameter and the Fire Data in the Fire Data, wind is obtained
Dangerous assessment index simultaneously establishes risk assessment grade.
Optionally, described that the characteristic variable is screened according to random forests algorithm, and according to the institute screened
It states characteristic variable and establishes index set, comprising:
The contribution rate of the characteristic variable is calculated by the Random Forest model parameter after optimization, and according to predetermined order side
Method is ranked up the contribution rate of the characteristic variable;
Contribution rate of accumulative total is obtained to meet the characteristic variable of preset ratio and establish index set.
Optionally, the Random Forest model parameter by after optimization, calculates the contribution rate of the characteristic variable, and root
According to predetermined order method to the contribution rate of the characteristic variable before, further includes:
Obtain Random Forest model parameter, feature division points selection criteria and feature selecting standard;Wherein, the feature choosing
The standard of selecting includes standard Gini coefficient and information gain standard;
Obtain decision tree information;Wherein, the decision tree information includes the decision tree quantity, the minimum of the decision tree
The minimum node number of leaf number and the decision tree;
By the decision tree information, the feature division points selection criteria and the feature selecting standard to described random
Forest model parameter optimizes, to obtain the Random Forest model parameter after optimization.
Optionally, the statistical indicator of social synthesis's statistical data of social synthesis's statistical data include the density of population,
Construction area, population from other places's ratio, knowledge on fire fighting popularity rate, income level of resident.
The second aspect of the embodiment of the present invention provides a kind of fire-fighting risk assessment device, comprising:
First obtains module, obtains data sample for obtaining statistical data, and according to the statistical data;Wherein, institute
Stating statistical data includes social synthesis's statistical data and Fire Data;
Module is established, for establishing and social synthesis's statistical data pair of the same period according to the Fire Data
The risk assessment grade answered, using the statistical indicator of the statistical data as characteristic variable, using the risk assessment grade as
Desired output is as a result, and establish training sample according to the characteristic variable and the desired output result;
Screening module, for being screened according to random forests algorithm to the characteristic variable, and according to screening
The characteristic variable establishes index set;
Pre-training module passes through institute for generating training sample set according to the index set and establishing machine learning model
It states training sample set and pre-training is carried out to the machine learning model, to obtain Optimized model;
Second obtains module, for according to the index set acquisition index score value, and will be described in index score value input
Optimized model, to obtain the assessment result of the Optimized model output.
The third aspect of the embodiment of the present invention provides a kind of terminal device, comprising: memory, processor and is stored in
In the memory and the computer program that can run on the processor, when the processor executes the computer program
It realizes such as the step of the above method.
The fourth aspect of the embodiment of the present invention provides a kind of computer readable storage medium, the computer-readable storage
Media storage has computer program, realizes when the computer program is executed by processor such as the step of the above method.
On the one hand the embodiment of the present invention establishes training sample according to risk assessment grade and social synthesis's statistical data, to spy
Sign variable is screened, and is simplified using the characteristic variable screened to training sample, can be according to risk assessment grade
Rejecting is associated with little characteristic variable with Urban Fires, reduces the invalid data in training sample, simultaneously effective reduces data
Dimension avoids dimension disaster, improves the training effectiveness of machine learning, at the same make to be trained according to training sample come machine learning
Model is more stable, predicts the fire risk based on machine learning model more accurate;On the other hand, root of the embodiment of the present invention
Factually border fire hazard calculation risk evaluation grade, and using risk assessment grade as the characteristic variable of screening training sample
Foundation, therefore the selection of characteristic variable is comparatively more objective.
Detailed description of the invention
It to describe the technical solutions in the embodiments of the present invention more clearly, below will be to embodiment or description of the prior art
Needed in attached drawing be briefly described, it should be apparent that, the accompanying drawings in the following description is only of the invention some
Embodiment for those of ordinary skill in the art without any creative labor, can also be according to these
Attached drawing obtains other attached drawings.
Fig. 1 is the flow diagram for the fire-fighting methods of risk assessment that the embodiment of the present invention one provides;
Fig. 2 is the flow diagram of fire-fighting methods of risk assessment provided by Embodiment 2 of the present invention;
Fig. 3 is the flow diagram for the fire-fighting methods of risk assessment that the embodiment of the present invention three provides;
Fig. 4 is the flow diagram for the fire-fighting methods of risk assessment that the embodiment of the present invention four provides;
Fig. 5 is the structural schematic diagram for the fire-fighting risk assessment device that the embodiment of the present invention five provides;
Fig. 6 is the schematic diagram for the terminal device that the embodiment of the present invention six provides.
Specific embodiment
In order to enable those skilled in the art to better understand the solution of the present invention, below in conjunction in the embodiment of the present invention
Attached drawing, technical solution in the embodiment of the present invention are explicitly described, it is clear that described embodiment is the present invention one
The embodiment divided, instead of all the embodiments.Based on the embodiments of the present invention, those of ordinary skill in the art are not doing
Every other embodiment obtained under the premise of creative work out, should fall within the scope of the present invention.
Description and claims of this specification and term " includes " and their any deformations in above-mentioned attached drawing, meaning
Figure, which is to cover, non-exclusive includes.Such as process, method or system comprising a series of steps or units, product or equipment do not have
It is defined in listed step or unit, but optionally further comprising the step of not listing or unit, or optionally also wrap
Include the other step or units intrinsic for these process, methods, product or equipment.In addition, term " first ", " second " and
" third " etc. is for distinguishing different objects, not for description particular order.
In order to illustrate technical solutions according to the invention, the following is a description of specific embodiments.
Embodiment one
As shown in Figure 1, the present embodiment provides a kind of fire-fighting methods of risk assessment, this method be can be applied to such as intelligent fire
The terminal devices such as prior-warning device, fire fighting alarm device, PC.Fire-fighting methods of risk assessment provided by the present embodiment, comprising:
S101, statistical data is obtained, and data sample is obtained according to the statistical data;Wherein, the statistical data packet
Include social synthesis's statistical data and Fire Data.
In a particular application, (in the present embodiment, can pass through social management information system) obtains about urban fire control area
The statistical data in domain, and data prediction is carried out to the statistical data, form data sample.Wherein, statistical data includes society
It can benchmark statistice and Fire Data.Social synthesis's statistical data includes statistical indicator.Statistical indicator includes but unlimited
In the density of population, construction area, population from other places's ratio, knowledge on fire fighting popularity rate, income level of resident.Data preprocessing method packet
It includes: statistical data progress missing values analysis, outlier detection, repeated data analysis, inconsistent value is analyzed and containing special
The value of symbol is analyzed.Polishing missing values, and excluding outlier, repeated data, inconsistent value and the value containing additional character.
S102, risk corresponding with social synthesis's statistical data of the same period is established according to the Fire Data
Evaluation grade, using the statistical indicator of social synthesis's statistical data as characteristic variable, using the risk assessment grade as
Desired output is as a result, and establish training sample according to the characteristic variable and the desired output result.
In a particular application, a specific period (such as in the past 2 years) can be chosen, it monthly (can also be by by the period
Season) it is divided into 24 periods.It is analyzed according to Fire Data of the parameter in Fire Data to every month,
The weight of each parameter in Fire Data is obtained by presupposition analysis method (such as analytic hierarchy process (AHP), expert's assignment method etc.),
By presetting the fire risk assessment index in the of that month Fire Data of calculation method calculating, establish and the society of the same period
The corresponding fire-fighting risk assessment step scale of meeting benchmark statistice, obtains of that month fire-fighting risk assessment grade, comprehensive with society
Close statistical data statistical indicator be characterized variable, using risk class as desired output as a result, establishing training sample.I.e. each instruction
Practicing sample includes of that month social synthesis's statistical data and fire-fighting risk assessment grade.Wherein, presupposition analysis method includes but not
It is limited to utilize analytic hierarchy process (AHP), expert's assignment method.Parameter in Fire Data includes but is not limited to fire frequency, wound
Die number, economic loss, influence of fire area.Default calculation method includes but is not limited to linear weighted function summation.
S103, the characteristic variable is screened according to random forests algorithm, and according to the feature screened
Variable establishes index set.
In a particular application, characteristic variable is screened, index set is established according to the characteristic variable after screening.In this reality
It applies in example, setting screens characteristic variable according to the contribution rate of characteristic variable by random forests algorithm.It needs to illustrate
It is that can be screened by other methods to characteristic variable.
S104, training sample set is generated according to the index set and establishes machine learning model, pass through the training sample
Collection carries out pre-training to the machine learning model, to obtain Optimized model.
In a particular application, training sample set is generated according to index set and establishes machine learning model, pass through training sample
Collection carries out pre-training to machine learning model, to obtain Optimized model.In the present embodiment, can set machine learning model as with
Machine forest model.Specifically, generating training sample set according to the index set will not be in the characteristic variable pair in index set
The data answered are rejected from training sample.
The implementation sample of Random Forest model:
Step 1. extracts the consistent data sample of n sample size with putting back to from all samples with repeat replication,
Training sample as training decision tree;
Step 2. extracts m characteristic variable in a random way from one of training sample;
Step 3. is chosen from m characteristic variable at the internal node of decision tree, according to Geordie impurity level minimum principle
The best feature xi of one classifying quality, is divided into Liang Ge branch for the node;
Step 4. repeats the above steps 3 operations to each internal node of decision tree, until the decision tree can
The Geordie impurity level of each node reaches minimum in Accurate classification training sample or decision tree;
Step 5. chooses next training sample, repeats step 2 to step 4, corresponding to all extraction training samples
Decision tree building finish;
The constructed decision tree come out of n training sample described in step 6. collectively constitutes Random Forest model, described random
Forest model building finishes.
S105, the Optimized model is inputted according to the index set acquisition index score value, and by the index score value, to obtain
Take the assessment result of the Optimized model output.
In a particular application, according to index set acquisition index score value, and index score value is inputted into Optimized model, it is excellent to obtain
Change the assessment result of model output.
In one embodiment, the statistical indicator of social synthesis's statistical data may include the density of population, building sides
Product, population from other places's ratio, knowledge on fire fighting popularity rate, income level of resident, total industrial output value etc..
The present embodiment establishes risk assessment grade by analysis to statistical data, processing, according to risk assessment grade and
Social synthesis's statistical data establishes training sample, screens to characteristic variable, using the characteristic variable screened to training
Sample is simplified, and can be rejected according to risk assessment grade and is associated with little characteristic variable with Urban Fires, and training sample is reduced
Invalid data in this, simultaneously effective reduces data dimension, avoids dimension disaster, improve the training effectiveness of machine learning, together
When make to be trained according to training sample come machine learning model it is more stable, keep the fire risk based on machine learning model pre-
It is more accurate to survey.
Embodiment two
As shown in Fig. 2, the present embodiment is the further explanation to the method and step in embodiment one.In the present embodiment,
Step S101, comprising:
S1011, statistical data is obtained.
In a particular application, the statistical data about urban fire control region is obtained by social management information system.
S1012, missing values analysis, outlier detection, repeated data analysis, inconsistent value are carried out to the statistical data
It analyzes and the value containing additional character is analyzed.
In a particular application, statistical data progress missing values analysis, outlier detection, repeated data are analyzed, is inconsistent
Value analysis and value analysis containing additional character, to be standardized to statistical data, normalized.
Missing values described in S1013, polishing, and reject the exceptional value, the repeated data, the inconsistent value and institute
The value containing additional character is stated, to obtain data sample.
In a particular application, polishing missing values, and excluding outlier, repeated data, inconsistent value and contain special symbol
Number value, or by the way that the special data in statistical data is analyzed and handled, to obtain standardized data sample.
The present embodiment is realized the standardization to statistical data, is returned by carrying out multinomial analysis processing operation to statistical data
One change processing, improves the accuracy rate of statistical data, effectively reduces error caused by due to anomaly statistics data.
Embodiment three
As shown in figure 3, the present embodiment is the further explanation to the method and step in embodiment one.In the present embodiment,
Step S102, comprising:
S1021, the Fire Data is analyzed according to the parameter in the Fire Data;Wherein, institute
Stating the parameter in Fire Data includes fire frequency, the number of casualties, economic loss, influence of fire area.
In a particular application, Fire Data is analyzed according to the multiple parameters in Fire Data, with right
The risk factor of fire is specifically assessed in Fire Data.Wherein, the parameter in Fire Data includes fire hair
Raw number, the number of casualties, economic loss, influence of fire area.
S1022, the weight for obtaining each parameter in the Fire Data.
In a particular application, the weight of each parameter in Fire Data is obtained by analytic hierarchy process (AHP).
S1023, according to the weight of each parameter in the parameter and the Fire Data in the Fire Data,
It obtains risk assessment index and establishes risk assessment grade.
In a particular application, logical according to the weight of each parameter in the parameter and Fire Data in Fire Data
The risk assessment index that default calculation method calculates event of fire in Fire Data is crossed, is established and the society of the same period
The corresponding fire-fighting risk assessment step scale of benchmark statistice, to establish risk assessment grade.Wherein, calculation method packet is preset
It includes but is not limited to linear weighted function summation.
The present embodiment obtains data sample by analysis, calculating to each parameter of event of fire in Fire Data
Corresponding risk assessment grade, and training sample is established, it lays a good foundation to improve efficiency and the accuracy rate of training, so as to basis
Practical fire hazard calculation risk evaluation grade, and using risk assessment grade as screening training sample characteristic variable according to
According to, therefore the selection of characteristic variable is comparatively more objective.The extraneous data in training sample is advantageously reduced to engineering
Practise the interference of model.
Example IV
As shown in figure 4, the present embodiment is the further explanation to the method and step in embodiment one.In the present embodiment,
Step S103, comprising:
S1031, the contribution rate that the characteristic variable is calculated by the Random Forest model parameter after optimization, and according to default
Sort method is ranked up the contribution rate of the characteristic variable.
In a particular application, characteristic variable is calculated by the Random Forest model parameter after optimizing in random forests algorithm
Contribution rate is ranked up according to contribution rate of the predetermined order method to characteristic variable.Wherein, predetermined order method include according to by
Small sequence is arrived greatly to be ranked up the contribution rate of characteristic variable or the contribution rate according to ascending sequence to characteristic variable
It is ranked up.
S1032, acquisition contribution rate of accumulative total meet the characteristic variable of preset ratio and establish index set.
In a particular application, it obtains in sequence, contribution rate of accumulative total meets the characteristic variable of preset ratio, and according to accumulative tribute
The characteristic variable that the rate of offering meets preset ratio establishes index set, wherein and preset ratio can specifically be set according to the actual situation,
For example, if set preset ratio as contribution rate it is high preceding 95%, in the sequence being ranked up according to descending sequence,
Obtain the characteristic variable that contribution rate of accumulative total is located at preceding 95%;In the sequence being ranked up according to ascending sequence, obtain
Contribution rate of accumulative total is located at rear 95% characteristic variable.
In one embodiment, before the step S1031, further includes:
S1033, Random Forest model parameter, feature division points selection criteria and feature selecting standard are obtained;Wherein, described
Feature selecting standard includes standard Gini coefficient and information gain standard.
S1034, decision tree information is obtained;Wherein, the decision tree information includes the decision tree quantity, the decision tree
Minimum leaf number and the decision tree minimum node number.
S1035, by the decision tree information, the feature division points selection criteria and the feature selecting standard to institute
It states Random Forest model parameter to optimize, to obtain the Random Forest model parameter after optimization.
In a particular application, Random Forest model parameter, feature division points selection criteria and feature selecting standard are obtained, into
And decision tree information is obtained, by decision tree feature division points selection criteria and feature selecting standard to Random Forest model parameter
It optimizes, to obtain the forest model parameter after optimization, realizes the screening operation to feature vector.Wherein, feature selecting mark
Standard includes but is not limited to standard Gini coefficient and information gain standard.Decision tree information includes but is not limited to decision tree quantity, determines
The minimum leaf number of plan tree and the minimum node number of decision tree.In the present embodiment, feature division points selection criteria can be random
Setting.
The present embodiment screens characteristic variable by Random Forest model, simplifies data sample, improves training
Efficiency and accuracy rate, further improve the risk assessment efficiency to dangerous material.
It should be understood that the size of the serial number of each step is not meant that the order of the execution order in above-described embodiment, each process
Execution sequence should be determined by its function and internal logic, the implementation process without coping with the embodiment of the present invention constitutes any limit
It is fixed.
Embodiment five
As shown in figure 5, the present embodiment provides a kind of fire-fighting risk assessment devices 100, for executing the side in embodiment one
Method step.Fire-fighting risk assessment device 100 provided in this embodiment, comprising:
First obtains module 101, obtains data sample for obtaining statistical data, and according to the statistical data;Wherein,
The statistical data includes social synthesis's statistical data and Fire Data;
Module 102 is established, for establishing and social synthesis's statistical number of the same period according to the Fire Data
According to corresponding risk assessment grade, using the statistical indicator of the statistical data as characteristic variable, by the risk assessment grade
As desired output as a result, and establishing training sample according to the characteristic variable and the desired output result;
Screening module 103, for being screened according to random forests algorithm to the characteristic variable, and according to screening
The characteristic variable establish index set;
Pre-training module 104 passes through for generating training sample set according to the index set and establishing machine learning model
The training sample set carries out pre-training to the machine learning model, to obtain Optimized model;
Second obtains module 105, for inputting institute according to the index set acquisition index score value, and by the index score value
Optimized model is stated, to obtain the assessment result of the Optimized model output.
In one embodiment, described first module 101 is obtained, comprising:
First acquisition unit, for obtaining statistical data;
First analytical unit, for carrying out missing values analysis, outlier detection, repeated data point to the statistical data
Analysis, inconsistent value analysis and the value analysis containing additional character;
Processing unit for missing values described in polishing, and rejects the exceptional value, repeated data, described inconsistent
Value and the value containing additional character, to obtain data sample.
In one embodiment, the module 102 of establishing includes:
Second analytical unit, for being divided according to the parameter in the Fire Data the Fire Data
Analysis;Wherein, the parameter in the Fire Data includes fire frequency, the number of casualties, economic loss, influence of fire face
Product;
Second acquisition unit, for obtaining the weight of each parameter in the Fire Data;
Third acquiring unit, for according to each in the parameter and the Fire Data in the Fire Data
The weight of parameter obtains risk assessment index and establishes risk assessment grade.
In one embodiment, the screening module 103 includes:
Computing unit, for calculating the contribution rate of the characteristic variable by the Random Forest model parameter after optimization, and
It is ranked up according to contribution rate of the predetermined order method to the characteristic variable;
4th acquiring unit meets the characteristic variable of preset ratio and establishes index for obtaining contribution rate of accumulative total
Collection.
In one embodiment, the screening module 103 further include:
5th acquiring unit, for obtaining Random Forest model parameter, feature division points selection criteria and feature selecting mark
It is quasi-;Wherein, the feature selecting standard includes standard Gini coefficient and information gain standard;
6th acquiring unit, for obtaining decision tree information;Wherein, the decision tree information includes the decision tree number
Amount, the minimum node number of the minimum leaf number of the decision tree and the decision tree;
Optimize unit, for passing through the decision tree information, the feature division points selection criteria and the feature selecting
Standard optimizes the Random Forest model parameter, to obtain the Random Forest model parameter after optimization.
In one embodiment, the statistical indicator of social synthesis's statistical data include the density of population, it is construction area, outer
Come population ratio, knowledge on fire fighting popularity rate, income level of resident, total industrial output value.
On the one hand the present embodiment establishes training sample according to risk assessment grade and social synthesis's statistical data, become to feature
Amount is screened, and is simplified, can be rejected according to risk assessment grade to training sample using the characteristic variable screened
It is associated with little characteristic variable with Urban Fires, reduces the invalid data in training sample, simultaneously effective reduces data dimension,
Avoid dimension disaster, improve the training effectiveness of machine learning, at the same make to be trained according to training sample come machine learning model
It is more stable, predict the fire risk based on machine learning model more accurate;On the other hand, the embodiment of the present invention is according to reality
Border fire hazard calculation risk evaluation grade, and using risk assessment grade as screening training sample characteristic variable according to
According to, therefore the selection of characteristic variable is comparatively more objective.
Embodiment six
Fig. 6 is the schematic diagram for the terminal device that the present embodiment six provides.As shown in fig. 6, the terminal device 6 of the embodiment wraps
It includes: processor 60, memory 61 and being stored in the computer that can be run in the memory 61 and on the processor 60
Program 62, such as fire-fighting risk assessment procedures.The processor 60 realizes above-mentioned each disappear when executing the computer program 62
Step in anti-methods of risk assessment embodiment, such as step S101 to S105 shown in FIG. 1.Alternatively, the processor 60 is held
The function of each module/unit in above-mentioned each Installation practice, such as module shown in Fig. 5 are realized when the row computer program 62
101 to 105 function.
Illustratively, the computer program 62 can be divided into one or more module/units, it is one or
Multiple module/units are stored in the memory 61, and are executed by the processor 60, to complete the present invention.Described one
A or multiple module/units can be the series of computation machine program instruction section that can complete specific function, which is used for
Implementation procedure of the computer program 62 in the terminal device 6 is described.For example, the computer program 62 can be divided
It is cut into the first acquisition module, establishes module, screening module, pre-training module and the second acquisition module, each module concrete function is such as
Under:
First obtains module, obtains data sample for obtaining statistical data, and according to the statistical data;Wherein, institute
Stating statistical data includes social synthesis's statistical data and Fire Data;
Module is established, for establishing and social synthesis's statistical data pair of the same period according to the Fire Data
The risk assessment grade answered, using the statistical indicator of the statistical data as characteristic variable, using the risk assessment grade as
Desired output is as a result, and establish training sample according to the characteristic variable and the desired output result;
Screening module, for being screened according to random forests algorithm to the characteristic variable, and according to screening
The characteristic variable establishes index set;
Pre-training module passes through institute for generating training sample set according to the index set and establishing machine learning model
It states training sample set and pre-training is carried out to the machine learning model, to obtain Optimized model;
Second obtains module, for according to the index set acquisition index score value, and will be described in index score value input
Optimized model, to obtain the assessment result of the Optimized model output.
The terminal device 6 can be the calculating such as desktop PC, notebook, palm PC and cloud server and set
It is standby.The terminal device may include, but be not limited only to, processor 60, memory 61.It will be understood by those skilled in the art that Fig. 6
The only example of terminal device 6 does not constitute the restriction to terminal device 6, may include than illustrating more or fewer portions
Part perhaps combines certain components or different components, such as the terminal device can also include input-output equipment, net
Network access device, bus etc..
Alleged processor 60 can be central processing unit (Central Processing Unit, CPU), can also be
Other general processors, digital signal processor (Digital Signal Processor, DSP), specific integrated circuit
(Application Specific Integrated Circuit, ASIC), ready-made programmable gate array (Field-
Programmable Gate Array, FPGA) either other programmable logic device, discrete gate or transistor logic,
Discrete hardware components etc..General processor can be microprocessor or the processor is also possible to any conventional processor
Deng.
The memory 61 can be the internal storage unit of the terminal device 6, such as the hard disk or interior of terminal device 6
It deposits.The memory 61 is also possible to the External memory equipment of the terminal device 6, such as be equipped on the terminal device 6
Plug-in type hard disk, intelligent memory card (Smart Media Card, SMC), safe digital card (Secure Digital, SD) dodge
Deposit card (Flash Card) etc..Further, the memory 61 can also both include the storage inside list of the terminal device 6
Member also includes External memory equipment.The memory 61 is for storing needed for the computer program and the terminal device
Other programs and data.The memory 61 can be also used for temporarily storing the data that has exported or will export.
It is apparent to those skilled in the art that for convenience of description and succinctly, only with above-mentioned each function
Can unit, module division progress for example, in practical application, can according to need and by above-mentioned function distribution by different
Functional unit, module are completed, i.e., the internal structure of described device is divided into different functional unit or module, more than completing
The all or part of function of description.Each functional unit in embodiment, module can integrate in one processing unit, can also
To be that each unit physically exists alone, can also be integrated in one unit with two or more units, it is above-mentioned integrated
Unit both can take the form of hardware realization, can also realize in the form of software functional units.In addition, each function list
Member, the specific name of module are also only for convenience of distinguishing each other, the protection scope being not intended to limit this application.Above system
The specific work process of middle unit, module, can refer to corresponding processes in the foregoing method embodiment, and details are not described herein.
In the above-described embodiments, it all emphasizes particularly on different fields to the description of each embodiment, is not described in detail or remembers in some embodiment
The part of load may refer to the associated description of other embodiments.
Those of ordinary skill in the art may be aware that list described in conjunction with the examples disclosed in the embodiments of the present disclosure
Member and algorithm steps can be realized with the combination of electronic hardware or computer software and electronic hardware.These functions are actually
It is implemented in hardware or software, the specific application and design constraint depending on technical solution.Professional technician
Each specific application can be used different methods to achieve the described function, but this realization is it is not considered that exceed
The scope of the present invention.
In embodiment provided by the present invention, it should be understood that disclosed device/terminal device and method, it can be with
It realizes by another way.For example, device described above/terminal device embodiment is only schematical, for example, institute
The division of module or unit is stated, only a kind of logical function partition, there may be another division manner in actual implementation, such as
Multiple units or components can be combined or can be integrated into another system, or some features can be ignored or not executed.Separately
A bit, shown or discussed mutual coupling or direct-coupling or communication connection can be through some interfaces, device
Or the INDIRECT COUPLING or communication connection of unit, it can be electrical property, mechanical or other forms.
The unit as illustrated by the separation member may or may not be physically separated, aobvious as unit
The component shown may or may not be physical unit, it can and it is in one place, or may be distributed over multiple
In network unit.It can select some or all of unit therein according to the actual needs to realize the mesh of this embodiment scheme
's.
It, can also be in addition, the functional units in various embodiments of the present invention may be integrated into one processing unit
It is that each unit physically exists alone, can also be integrated in one unit with two or more units.Above-mentioned integrated list
Member both can take the form of hardware realization, can also realize in the form of software functional units.
If the integrated module/unit be realized in the form of SFU software functional unit and as independent product sale or
In use, can store in a computer readable storage medium.Based on this understanding, the present invention realizes above-mentioned implementation
All or part of the process in example method, can also instruct relevant hardware to complete, the meter by computer program
Calculation machine program can be stored in a computer readable storage medium, the computer program when being executed by processor, it can be achieved that on
The step of stating each embodiment of the method.Wherein, the computer program includes computer program code, the computer program generation
Code can be source code form, object identification code form, executable file or certain intermediate forms etc..The computer-readable medium
It may include: any entity or device, recording medium, USB flash disk, mobile hard disk, magnetic that can carry the computer program code
Dish, CD, computer storage, read-only memory (ROM, Read-Only Memory), random access memory (RAM,
Random Access Memory), electric carrier signal, telecommunication signal and software distribution medium etc..It should be noted that described
The content that computer-readable medium includes can carry out increasing appropriate according to the requirement made laws in jurisdiction with patent practice
Subtract, such as does not include electric carrier signal and electricity according to legislation and patent practice, computer-readable medium in certain jurisdictions
Believe signal.
Embodiment described above is merely illustrative of the technical solution of the present invention, rather than its limitations;Although referring to aforementioned reality
Applying example, invention is explained in detail, those skilled in the art should understand that: it still can be to aforementioned each
Technical solution documented by embodiment is modified or equivalent replacement of some of the technical features;And these are modified
Or replacement, the spirit and scope for technical solution of various embodiments of the present invention that it does not separate the essence of the corresponding technical solution should all
It is included within protection scope of the present invention.
Claims (10)
1. a kind of fire-fighting methods of risk assessment characterized by comprising
Statistical data is obtained, and data sample is obtained according to the statistical data;Wherein, the statistical data includes social synthesis
Statistical data and Fire Data;
Risk assessment grade corresponding with social synthesis's statistical data of the same period is established according to the Fire Data,
Using the statistical indicator of social synthesis's statistical data as characteristic variable, using the risk assessment grade as desired output knot
Fruit, and training sample is established according to the characteristic variable and the desired output result;
The characteristic variable is screened according to random forests algorithm, and is referred to according to the characteristic variable foundation screened
Mark collection;
Training sample set is generated according to the index set and establishes machine learning model, by the training sample set to the machine
Device learning model carries out pre-training, to obtain Optimized model;
The Optimized model is inputted according to the index set acquisition index score value, and by the index score value, it is described excellent to obtain
Change the assessment result of model output.
2. fire-fighting methods of risk assessment as described in claim 1, which is characterized in that the acquisition statistical data, and to described
Statistical data is pre-processed, to obtain data sample, comprising:
Obtain statistical data;
Missing values analysis, outlier detection, repeated data analysis, inconsistent value analysis are carried out to the statistical data and contained
The value of additional character is analyzed;
Missing values described in polishing, and reject the exceptional value, the repeated data, the inconsistent value and described containing special
The value of symbol, to obtain data sample.
3. fire-fighting methods of risk assessment as described in claim 1, which is characterized in that according to the Fire Data establish with
The corresponding risk assessment grade of social synthesis's statistical data includes:
The Fire Data is analyzed according to the parameter in the Fire Data;Wherein, the fire statistics
Parameter in data includes fire frequency, the number of casualties, economic loss, influence of fire area;
Obtain the weight of each parameter in the Fire Data;
According to the weight of each parameter in the parameter and the Fire Data in the Fire Data, obtains risk and comment
Estimate index and establishes risk assessment grade.
4. fire-fighting methods of risk assessment as described in claim 1, which is characterized in that it is described according to random forests algorithm to described
Characteristic variable is screened, and establishes index set according to the characteristic variable screened, comprising:
The contribution rate of the characteristic variable is calculated by the Random Forest model parameter after optimization, and according to predetermined order method pair
The contribution rate of the characteristic variable is ranked up;
Contribution rate of accumulative total is obtained to meet the characteristic variable of preset ratio and establish index set.
5. fire-fighting methods of risk assessment as claimed in claim 4, which is characterized in that the random forest mould by after optimization
Shape parameter calculates the contribution rate of the characteristic variable, and is arranged according to contribution rate of the predetermined order method to the characteristic variable
Before sequence, further includes:
Obtain Random Forest model parameter, feature division points selection criteria and feature selecting standard;Wherein, the feature selecting mark
Standard includes standard Gini coefficient and information gain standard;
Obtain decision tree information;Wherein, the decision tree information includes the minimum leaf of the decision tree quantity, the decision tree
Several and the decision tree minimum node number;
By the decision tree information, the feature division points selection criteria and the feature selecting standard to the random forest
Model parameter optimizes, to obtain the Random Forest model parameter after optimization.
6. fire-fighting methods of risk assessment as described in claim 1, which is characterized in that the statistics of social synthesis's statistical data
Index includes the density of population, construction area, population from other places's ratio, knowledge on fire fighting popularity rate, income level of resident.
7. a kind of fire-fighting risk assessment device characterized by comprising
First obtains module, obtains data sample for obtaining statistical data, and according to the statistical data;Wherein, the system
It counts including social synthesis's statistical data and Fire Data;
Module is established, it is corresponding with social synthesis's statistical data of the same period for being established according to the Fire Data
Risk assessment grade, using the statistical indicator of the statistical data as characteristic variable, using the risk assessment grade as expectation
Output is as a result, and establish training sample according to the characteristic variable and the desired output result;
Screening module, for being screened according to random forests algorithm to the characteristic variable, and according to screening
Characteristic variable establishes index set;
Pre-training module passes through the instruction for generating training sample set according to the index set and establishing machine learning model
Practice sample set and pre-training is carried out to the machine learning model, to obtain Optimized model;
Second obtains module, for inputting the optimization according to the index set acquisition index score value, and by the index score value
Model, to obtain the assessment result of the Optimized model output.
8. fire-fighting risk assessment device as claimed in claim 7, which is characterized in that described first obtains module, comprising:
First acquisition unit, for obtaining statistical data;
Second analytical unit, for analyzing, no statistical data progress missing values analysis, outlier detection, repeated data
Consistent value analysis and the value containing additional character are analyzed;
Processing unit for missing values described in polishing, and rejects the exceptional value, the repeated data, the inconsistent value
And the value containing additional character, to obtain data sample.
9. a kind of terminal device, including memory, processor and storage are in the memory and can be on the processor
The computer program of operation, which is characterized in that the processor realizes such as claim 1 to 6 when executing the computer program
The step of any one the method.
10. a kind of computer readable storage medium, the computer-readable recording medium storage has computer program, and feature exists
In when the computer program is executed by processor the step of any one of such as claim 1 to 6 of realization the method.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201811286629.1A CN109242361A (en) | 2018-10-31 | 2018-10-31 | A kind of fire-fighting methods of risk assessment, device and terminal device |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201811286629.1A CN109242361A (en) | 2018-10-31 | 2018-10-31 | A kind of fire-fighting methods of risk assessment, device and terminal device |
Publications (1)
Publication Number | Publication Date |
---|---|
CN109242361A true CN109242361A (en) | 2019-01-18 |
Family
ID=65079985
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN201811286629.1A Pending CN109242361A (en) | 2018-10-31 | 2018-10-31 | A kind of fire-fighting methods of risk assessment, device and terminal device |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN109242361A (en) |
Cited By (13)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN110459031A (en) * | 2019-08-06 | 2019-11-15 | 福建工程学院 | A kind of anti-disaster method of fire-fighting, system and storage medium having warning function |
CN110503566A (en) * | 2019-07-08 | 2019-11-26 | 中国平安人寿保险股份有限公司 | Air control method for establishing model, device, computer equipment and storage medium |
CN111353386A (en) * | 2020-02-04 | 2020-06-30 | 重庆特斯联智慧科技股份有限公司 | Fire-fighting risk intelligent evaluation method and system based on deep learning |
CN111353702A (en) * | 2020-02-28 | 2020-06-30 | 中国工商银行股份有限公司 | Change operation risk calculation method and device |
CN112348296A (en) * | 2019-08-07 | 2021-02-09 | 中移信息技术有限公司 | Telecommunication data acquisition method, device, equipment and storage medium |
CN112529327A (en) * | 2020-12-21 | 2021-03-19 | 北京建筑大学 | Method for constructing fire risk prediction grade model of buildings in commercial areas |
CN113313417A (en) * | 2021-06-23 | 2021-08-27 | 北京鼎泰智源科技有限公司 | Complaint risk signal grading method and device based on decision tree model |
CN113553754A (en) * | 2020-04-23 | 2021-10-26 | 中国石油化工股份有限公司 | Memory, fire risk prediction model construction method, system and device |
CN113569482A (en) * | 2021-07-29 | 2021-10-29 | 石家庄铁道大学 | Method and device for evaluating service performance of tunnel, terminal and storage medium |
CN113723835A (en) * | 2021-09-02 | 2021-11-30 | 国网河北省电力有限公司电力科学研究院 | Thermal power plant water use evaluation method and terminal equipment |
CN114399816A (en) * | 2021-12-28 | 2022-04-26 | 北方工业大学 | Community fire risk sensing method and device |
CN115471993A (en) * | 2022-07-06 | 2022-12-13 | 江苏科技大学 | Fire alarm management level evaluation method and system based on IMODE (inertial measurement System) hierarchical evolution algorithm |
CN116503026A (en) * | 2023-06-26 | 2023-07-28 | 广东省科技基础条件平台中心 | Operation and maintenance risk assessment method, system and storage medium for science and technology items |
Citations (2)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
US20100138253A1 (en) * | 2008-12-02 | 2010-06-03 | Chun-Chang Chao | All aspect quantification system for the risk rating of operating errors for an advanced boiling water reactor |
CN108376310A (en) * | 2018-02-06 | 2018-08-07 | 深圳前海大观信息技术有限公司 | Building fire risk class appraisal procedure |
-
2018
- 2018-10-31 CN CN201811286629.1A patent/CN109242361A/en active Pending
Patent Citations (2)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
US20100138253A1 (en) * | 2008-12-02 | 2010-06-03 | Chun-Chang Chao | All aspect quantification system for the risk rating of operating errors for an advanced boiling water reactor |
CN108376310A (en) * | 2018-02-06 | 2018-08-07 | 深圳前海大观信息技术有限公司 | Building fire risk class appraisal procedure |
Non-Patent Citations (1)
Title |
---|
党杰: ""城市火灾风险评估指标体系研究"", 《中国优秀硕士学位论文全文数据库 工程科技Ⅱ辑》 * |
Cited By (21)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN110503566A (en) * | 2019-07-08 | 2019-11-26 | 中国平安人寿保险股份有限公司 | Air control method for establishing model, device, computer equipment and storage medium |
CN110503566B (en) * | 2019-07-08 | 2024-02-09 | 中国平安人寿保险股份有限公司 | Wind control model building method and device, computer equipment and storage medium |
CN110459031A (en) * | 2019-08-06 | 2019-11-15 | 福建工程学院 | A kind of anti-disaster method of fire-fighting, system and storage medium having warning function |
CN112348296A (en) * | 2019-08-07 | 2021-02-09 | 中移信息技术有限公司 | Telecommunication data acquisition method, device, equipment and storage medium |
CN112348296B (en) * | 2019-08-07 | 2023-12-22 | 中移信息技术有限公司 | Telecommunication data acquisition method, device, equipment and storage medium |
CN111353386B (en) * | 2020-02-04 | 2023-01-17 | 重庆特斯联智慧科技股份有限公司 | Fire-fighting risk intelligent assessment method and system based on deep learning |
CN111353386A (en) * | 2020-02-04 | 2020-06-30 | 重庆特斯联智慧科技股份有限公司 | Fire-fighting risk intelligent evaluation method and system based on deep learning |
CN111353702A (en) * | 2020-02-28 | 2020-06-30 | 中国工商银行股份有限公司 | Change operation risk calculation method and device |
CN113553754A (en) * | 2020-04-23 | 2021-10-26 | 中国石油化工股份有限公司 | Memory, fire risk prediction model construction method, system and device |
CN112529327A (en) * | 2020-12-21 | 2021-03-19 | 北京建筑大学 | Method for constructing fire risk prediction grade model of buildings in commercial areas |
CN113313417B (en) * | 2021-06-23 | 2024-01-26 | 北京鼎泰智源科技有限公司 | Method and device for classifying complaint risk signals based on decision tree model |
CN113313417A (en) * | 2021-06-23 | 2021-08-27 | 北京鼎泰智源科技有限公司 | Complaint risk signal grading method and device based on decision tree model |
CN113569482A (en) * | 2021-07-29 | 2021-10-29 | 石家庄铁道大学 | Method and device for evaluating service performance of tunnel, terminal and storage medium |
CN113569482B (en) * | 2021-07-29 | 2024-02-06 | 石家庄铁道大学 | Tunnel service performance evaluation method, device, terminal and storage medium |
CN113723835A (en) * | 2021-09-02 | 2021-11-30 | 国网河北省电力有限公司电力科学研究院 | Thermal power plant water use evaluation method and terminal equipment |
CN113723835B (en) * | 2021-09-02 | 2024-02-06 | 国网河北省电力有限公司电力科学研究院 | Water consumption evaluation method and terminal equipment for thermal power plant |
CN114399816A (en) * | 2021-12-28 | 2022-04-26 | 北方工业大学 | Community fire risk sensing method and device |
CN115471993A (en) * | 2022-07-06 | 2022-12-13 | 江苏科技大学 | Fire alarm management level evaluation method and system based on IMODE (inertial measurement System) hierarchical evolution algorithm |
CN115471993B (en) * | 2022-07-06 | 2023-09-26 | 江苏科技大学 | Fire alarm management level evaluation method and system based on IMODE hierarchical evolutionary algorithm |
CN116503026A (en) * | 2023-06-26 | 2023-07-28 | 广东省科技基础条件平台中心 | Operation and maintenance risk assessment method, system and storage medium for science and technology items |
CN116503026B (en) * | 2023-06-26 | 2024-02-09 | 广东省科技基础条件平台中心 | Operation and maintenance risk assessment method, system and storage medium for science and technology items |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
CN109242361A (en) | A kind of fire-fighting methods of risk assessment, device and terminal device | |
US11120552B2 (en) | Crop grading via deep learning | |
CN108427669A (en) | Abnormal behaviour monitoring method and system | |
CN103812961B (en) | Identify and specify the method and apparatus of classification IP address, defence method and system | |
CN111901171B (en) | Anomaly detection and attribution method, apparatus, device, and computer-readable storage medium | |
CN109686036A (en) | A kind of fire monitoring method, device and edge calculations device | |
CN108875525A (en) | Behavior prediction method, apparatus, system and storage medium | |
CN107767021A (en) | A kind of risk control method and equipment | |
KR102161256B1 (en) | Stocks selection apparatus for constructing stock portfolio and method thereof | |
CN106302522A (en) | A kind of network safety situations based on neutral net and big data analyze method and system | |
CN109544399B (en) | Power transmission equipment state evaluation method and device based on multi-source heterogeneous data | |
CN108898476A (en) | A kind of loan customer credit-graded approach and device | |
CN108717534A (en) | Operator's functional status assessment technology based on functional near-infrared spectrum technique | |
CN107203866A (en) | The processing method and device of order | |
CN110232499A (en) | A kind of power distribution network information physical side method for prewarning risk and system | |
CN103049483B (en) | The recognition system of webpage danger | |
CN110163251A (en) | A kind of Optimum Identification Method of fire hazard rating, device and terminal device | |
CN110348490A (en) | A kind of soil quality prediction technique and device based on algorithm of support vector machine | |
CN111986027A (en) | Abnormal transaction processing method and device based on artificial intelligence | |
CN109255480A (en) | Between servant lead prediction technique, device, computer equipment and storage medium | |
Gonaygunta | Machine learning algorithms for detection of cyber threats using logistic regression | |
CN109409923A (en) | Distribution method, computer readable storage medium and the terminal device of sales region | |
CN109657916A (en) | A kind of Fire risk assessment method, device and server | |
CN109523141A (en) | A kind of fire-fighting region deployment method, apparatus and terminal device | |
CN109377436A (en) | A kind of accurate monitoring and managing method of environment and device, terminal device and storage medium |
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 | ||
RJ01 | Rejection of invention patent application after publication |
Application publication date: 20190118 |
|
RJ01 | Rejection of invention patent application after publication |