CN109447127A - Data processing method and device - Google Patents
Data processing method and device Download PDFInfo
- Publication number
- CN109447127A CN109447127A CN201811149626.3A CN201811149626A CN109447127A CN 109447127 A CN109447127 A CN 109447127A CN 201811149626 A CN201811149626 A CN 201811149626A CN 109447127 A CN109447127 A CN 109447127A
- Authority
- CN
- China
- Prior art keywords
- driving behavior
- mentioned
- reliability
- attribute
- behavior data
- 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
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/24—Classification techniques
- G06F18/241—Classification techniques relating to the classification model, e.g. parametric or non-parametric approaches
Landscapes
- Engineering & Computer Science (AREA)
- Data Mining & Analysis (AREA)
- Theoretical Computer Science (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Bioinformatics & Cheminformatics (AREA)
- Bioinformatics & Computational Biology (AREA)
- Artificial Intelligence (AREA)
- Evolutionary Biology (AREA)
- Evolutionary Computation (AREA)
- Physics & Mathematics (AREA)
- General Engineering & Computer Science (AREA)
- General Physics & Mathematics (AREA)
- Life Sciences & Earth Sciences (AREA)
- Information Retrieval, Db Structures And Fs Structures Therefor (AREA)
Abstract
This application provides a kind of data processing method and devices, this method comprises: the driving behavior data of object to be sorted are obtained, the driving behavior data of driving behavior data and the second attribute in above-mentioned driving behavior data including the first attribute；The first reliability and the second reliability are determined according to the driving behavior data of the driving behavior data of above-mentioned first attribute and above-mentioned second attribute respectively；Wherein, above-mentioned first reliability is the reliability that the object to be sorted belongs to the first driving behavior classification, and above-mentioned second reliability is the reliability that above-mentioned object to be sorted belongs to the second driving behavior classification；Reliability maximum value is determined from above-mentioned first reliability and above-mentioned second reliability, and the corresponding driving behavior classification of above-mentioned reliability maximum value is determined as to the driving behavior classification of above-mentioned object to be sorted.Implement the application, the accuracy of determining driving behavior classification can be improved.
Description
Technical field
This application involves field of computer technology more particularly to a kind of data processing method and devices.
Background technique
With the rapid development of automobile internet, more and more users possess private car, while also more and more vehicles
It is travelled on road.
However, often there is the driving habit of some danger during car owner's driving.According to driving habit
Difference judge the driving behavior classification of car owner, corresponding service can be provided to different classes of car owner.For example, can be with
Recommend corresponding insurance set meal etc. than relatively hazardous car owner to driving habit.
How correctly to judge that the driving behavior classification of car owner is urgently to be resolved as a result,.
Summary of the invention
The application proposes that a kind of data processing method and device are mentioned for determining the driving behavior classification of object to be sorted
Height determines the accuracy of driving behavior classification.
In a first aspect, the embodiment of the present application provides a kind of data processing method, comprising:
The driving behavior data of object to be sorted are obtained, include the driving row of the first attribute in above-mentioned driving behavior data
For the driving behavior data of data and the second attribute；
Is determined according to the driving behavior data of the driving behavior data of above-mentioned first attribute and above-mentioned second attribute respectively
One reliability and the second reliability；Wherein, above-mentioned first reliability is the letter that above-mentioned object to be sorted belongs to the first driving behavior classification
Degree, above-mentioned second reliability are the reliability that above-mentioned object to be sorted belongs to the second driving behavior classification；
Reliability maximum value is determined from above-mentioned first reliability and above-mentioned second reliability, above-mentioned reliability maximum value is corresponding
Driving behavior classification be determined as the driving behavior classification of above-mentioned object to be sorted.
In the embodiment of the present application, the driving behavior data of object to be sorted are obtained, above-mentioned driving behavior data include at least
The driving behavior data of two attributes calculate above-mentioned object to be sorted according to above-mentioned driving behavior data and belong to driving behavior
The reliability of classification has determined the driving behavior classification of object to be sorted according to reliability, has improved the standard of determining driving behavior classification
Exactness.
With reference to first aspect, in one possible implementation, above-mentioned respectively according to the driving row of above-mentioned first attribute
Driving behavior data for data and above-mentioned second attribute determine the first reliability and the second reliability, comprising:
The driving behavior data of the driving behavior data and above-mentioned second attribute that calculate separately above-mentioned first attribute correspond to
The probability of above-mentioned first driving behavior classification, obtains the first Making by Probability Sets；
And calculate separately the driving behavior data of above-mentioned first attribute and the driving behavior data pair of above-mentioned second attribute
The probability that above-mentioned second driving behavior classification should be arrived, obtains the second Making by Probability Sets；
Above-mentioned first reliability is calculated according to above-mentioned first Making by Probability Sets, and is calculated according to above-mentioned second Making by Probability Sets
State the second reliability.
In the embodiment of the present application, the first Making by Probability Sets and the second Making by Probability Sets are calculated, according to above-mentioned first probability set
It closes and the second Making by Probability Sets calculates the first reliability and the second reliability, the accuracy for calculating reliability can be improved, and then improve true
Determine the accuracy of driving behavior classification.
Optionally, the driving row of above-mentioned the driving behavior data for calculating separately above-mentioned first attribute and above-mentioned second attribute
The probability that above-mentioned first driving behavior classification is corresponded to for data, obtains the first Making by Probability Sets；And calculate separately above-mentioned first
The driving behavior data of attribute and the driving behavior data of above-mentioned second attribute correspond to the general of above-mentioned second driving behavior classification
Rate obtains the second Making by Probability Sets, comprising:
The driving behavior data of the driving behavior data of above-mentioned first attribute and above-mentioned second attribute are input to respectively pre-
If in Bayesian rough set model, the positive domain set of output first and the second positive domain set；
The first support set is calculated according to the above-mentioned first positive domain set, and is gathered according to the above-mentioned second positive domain
Calculate the second support set；
The first confidence gain function set is calculated according to the above-mentioned first positive domain set, and just according to above-mentioned second
Domain set calculates the second confidence gain function set；
Above-mentioned first probability set is calculated according to above-mentioned first support set and above-mentioned first confidence gain function set
It closes, and above-mentioned second Making by Probability Sets is calculated according to above-mentioned second support set and above-mentioned second confidence gain function set.
Optionally, above-mentioned that first probability is calculated according to the first support set and the first confidence gain function set
Set, and the second Making by Probability Sets is calculated according to the second support set and the second confidence gain function set, comprising:
Above-mentioned first support set and above-mentioned first confidence gain function set are normalized respectively, obtained
To the first confidence gain function set of the first support set of normalization and normalization；
And place is normalized in above-mentioned second support set and above-mentioned second confidence gain function set respectively
Reason obtains the second support set of normalization and normalizes the second confidence gain function set；
According to the weighting of above-mentioned normalization the first support set and above-mentioned normalization the first confidence gain function set
With obtain the first Making by Probability Sets；
And according to above-mentioned normalization the second support set and above-mentioned normalization the second confidence gain function set plus
Quan He obtains the second Making by Probability Sets.
In the embodiment of the present application, the first probability is calculated according to the first support and the weighted sum of the first confidence gain function
Set, and the second Making by Probability Sets is calculated according to the second support and the weighted sum of the second confidence gain function, meter can be reduced
The error of Making by Probability Sets is calculated, and then improves the accuracy for determining driving behavior classification.
Optionally, above-mentioned that above-mentioned first reliability is calculated according to above-mentioned first Making by Probability Sets, and according to above-mentioned second probability
Set calculates above-mentioned second reliability, comprising:
Fusion treatment is carried out to above-mentioned first Making by Probability Sets by combining evidences rule, obtains above-mentioned first reliability；
And fusion treatment is carried out to above-mentioned second Making by Probability Sets by above-mentioned combining evidences rule, obtain above-mentioned second
Reliability.
In the embodiment of the present application, each Making by Probability Sets indicates that the driving behavior data of different attribute correspond to a certain driving
The set of the probability of behavior classification carries out fusion treatment to above-mentioned each Making by Probability Sets by evidence fusion rule and obtains reliability,
The probability that a certain driving behavior classification can be corresponded to the driving behavior data of comprehensive consideration different attribute is improved determination and driven
Sail the accuracy of behavior classification.
With reference to first aspect, in one possible implementation, believe described from first reliability with described second
Reliability maximum value is determined in degree, and the corresponding driving behavior classification of the reliability maximum value is determined as the object to be sorted
Driving behavior classification after, the method also includes:
The corresponding driving behavior classification of the reliability maximum value is output in display；
Alternatively, in the corresponding driving behavior classification of the reliability maximum value in the case where preset class of risk range,
Output safety prompt information.
Second aspect, the embodiment of the present application provide a kind of data processing equipment, comprising:
Acquiring unit includes first in above-mentioned driving behavior data for obtaining the driving behavior data of object to be sorted
The driving behavior data of attribute and the driving behavior data of the second attribute；
First determination unit, for respectively according to the driving behavior data of above-mentioned first attribute and above-mentioned second attribute
Driving behavior data determine the first reliability and the second reliability；Wherein, above-mentioned first reliability is that above-mentioned object to be sorted belongs to first
The reliability of driving behavior classification, above-mentioned second reliability are the reliability that above-mentioned object to be sorted belongs to the second driving behavior classification；
Second determination unit, for determining reliability maximum value from first reliability and second reliability, by institute
State the driving behavior classification that the corresponding driving behavior classification of reliability maximum value is determined as the object to be sorted.
In conjunction with second aspect, in one possible implementation, above-mentioned first determination unit includes:
First computation subunit, for calculate separately above-mentioned first attribute driving behavior data and above-mentioned second attribute
Driving behavior data correspond to the probability of above-mentioned first driving behavior classification, obtain the first Making by Probability Sets；And it calculates separately
The driving behavior data of above-mentioned first attribute and the driving behavior data of above-mentioned second attribute correspond to above-mentioned second driving behavior
The probability of classification obtains the second Making by Probability Sets；
Second computation subunit is used to calculate above-mentioned first reliability according to above-mentioned first Making by Probability Sets, and according to above-mentioned
Second Making by Probability Sets calculates above-mentioned second reliability.
Optionally, above-mentioned second computation subunit is specifically used for through combining evidences rule to above-mentioned first Making by Probability Sets
Fusion treatment is carried out, above-mentioned first reliability is obtained；And above-mentioned second Making by Probability Sets is carried out by above-mentioned combining evidences rule
Fusion treatment obtains above-mentioned second reliability.
Optionally, above-mentioned first computation subunit, specifically for respectively by the driving behavior data of above-mentioned first attribute and
The driving behavior data of above-mentioned second attribute are input in default Bayesian rough set model, the first positive domain of output set and the
Two positive domain set；The first support set is calculated according to the above-mentioned first positive domain set, and total according to the above-mentioned second positive domain collection
Calculate the second support set；The first confidence gain function set is calculated according to the above-mentioned first positive domain set, and according to above-mentioned the
Two positive domain set calculate the second confidence gain function set；According to above-mentioned first support set and above-mentioned first confidence gain
Function set calculates above-mentioned first Making by Probability Sets, and according to above-mentioned second support set and above-mentioned second confidence gain letter
Manifold is total to be counted in stating the second Making by Probability Sets.
Optionally, above-mentioned first computation subunit is specifically used for respectively by above-mentioned first support set and above-mentioned first
Confidence gain function set is normalized, and obtains the first support set of normalization and normalizes the first confidence gain
Function set；And place is normalized in above-mentioned second support set and above-mentioned second confidence gain function set respectively
Reason obtains the second support set of normalization and normalizes the second confidence gain function set；According to above-mentioned normalization first
The weighted sum of degree of holding set and above-mentioned normalization the first confidence gain function set, obtains the first Making by Probability Sets；And according to upper
The weighted sum for stating normalization the second support set and above-mentioned normalization the second confidence gain function set, obtains the second probability
Set.
In the embodiment of the present application, above-mentioned first computation subunit is by the first support and the first confidence gain function
Weighted sum is general as second as the first Making by Probability Sets, and using the second support and the weighted sum of the second confidence gain function
Rate set can reduce calculating error, improve the accuracy for determining driving behavior classification.
Optionally, above-mentioned second computation subunit is specifically used for through combining evidences rule to above-mentioned first Making by Probability Sets
Fusion treatment is carried out, above-mentioned first reliability is obtained；And above-mentioned second Making by Probability Sets is carried out by above-mentioned combining evidences rule
Fusion treatment obtains above-mentioned second reliability.
The third aspect, the embodiment of the present application also provides a kind of data processing equipments, including processor and memory；On
Processor and above-mentioned memory is stated to be connected with each other by bus；Wherein, above-mentioned memory is above-mentioned for storing computer program
Computer program includes program instruction, and above-mentioned processor is configured for calling above procedure instruction, executes above-mentioned first aspect
Method.
Fourth aspect, the embodiment of the present application provide a kind of computer readable storage medium, above-mentioned computer-readable storage
Media storage has computer program, and above-mentioned computer program includes program instruction, and above procedure instruction, which is worked as, to be executed by processor
When, the method that makes above-mentioned processor execute above-mentioned first aspect.
5th aspect, the embodiment of the present application provides a kind of computer program product comprising program instruction, when it is being counted
When being run on calculation machine, so that computer executes method described in above-mentioned first aspect.
Detailed description of the invention
Technical solution in ord to more clearly illustrate embodiments of the present application or in background technique, below will be to the application reality
Attached drawing needed in example or background technique is applied to be illustrated.
Fig. 1 is a kind of flow diagram of data processing method provided by the embodiments of the present application；
Fig. 2 is the flow diagram of another data processing method provided by the embodiments of the present application；
Fig. 3 is a kind of concrete application scene schematic diagram of data processing method provided by the embodiments of the present application；
Fig. 4 is a kind of structural schematic diagram of data processing equipment provided by the embodiments of the present application；
Fig. 5 is a kind of structural schematic diagram of first determination unit provided by the embodiments of the present application；
Fig. 6 is the structural schematic diagram of another data processing equipment provided by the embodiments of the present application.
Specific embodiment
The description and claims of this application and term " first " in above-mentioned attached drawing, " second " etc. are for area
Not different objects, is not use to describe a particular order.In addition, term " includes " and " having " and their any deformations,
It is intended to cover and non-exclusive includes.Such as it contains the process, method of a series of steps or units, system, product or sets
It is standby to be not limited to listed step or unit, but optionally further comprising the step of not listing or unit, or optionally
It further include other step or units intrinsic for these process, methods or equipment.
This application provides a kind of data processing method and device, it can be used for determining the driving behavior class of object to be sorted
Not, the accuracy for determining driving behavior classification is improved.The embodiment of the present application is described below in conjunction with attached drawing.
Fig. 1 is a kind of flow diagram of data processing method provided by the embodiments of the present application.This method can be applied to
Data processing equipment, above-mentioned data processing equipment can be the terminal comprising processor, memory and bus, and above-mentioned terminal can
To include the handheld device with wireless communication function, mobile unit, wearable device, calculate equipment or be connected to wireless tune
Other processing equipments of modulator-demodulator and various forms of user equipmenies (user equipment, UE), mobile station
(mobile station, MS).Optionally, the data processing equipment can also for server etc., the embodiment of the present application for
The concrete form of the data processing equipment does not make uniqueness restriction.
As shown in Figure 1, this method comprises:
101, the driving behavior data of object to be sorted are obtained, include driving for the first attribute in above-mentioned driving behavior data
Sail the driving behavior data of behavioral data and the second attribute.
In the embodiment of the present application, which can indicate different car owners or different users.Specifically, for example
It is more effective to distinguish different objects to be sorted, and driving behavior data corresponding with the object to be sorted, therefore, one
In a little implementations, which can also carry identity.For example, it needs while locating in data processing equipment
In the case where the driving behavior data for managing two or more objects to be sorted, which is being obtained wait divide
When the driving behavior data of class object, which can carry identity.
In the present embodiment, which can be automobile generated data in the process of running.It is vivider
, above-mentioned driving behavior data can be regarded as above-mentioned object to be sorted generated data during manipulating automobile.Specifically
, the driving behavior data of driving behavior data and the second attribute in the driving behavior data including the first attribute, this first
Produced by the driving behavior data of attribute and the driving behavior data of second attribute can be understood as automobile in the process of running
Different attribute data.Or the driving behavior data of above-mentioned multiple attributes can reflect the driving habit of object to be sorted
Used or driving ability.
For example, the driving behavior data of first attribute can be regarded as what automobile was generated when observing traffic rules and regulations
Data, the driving behavior data of second attribute can be regarded as the data that automobile is generated when not observing traffic rules and regulations.
Again for example, the driving behavior data of above-mentioned multiple attributes may include that sideslip drives frequency, and road occupying drives frequency
Rate, emergency turn frequency, the one or more in general steering frequency and furious driving frequency.Wherein, above-mentioned sideslip drives
At the appointed time there is the number of sideslip driving behavior in the driving procedure in section in the above-mentioned object to be sorted of frequency representation, above-mentioned
Road occupying, which drives, there is the number of road occupying driving in driving procedure of the above-mentioned object to be sorted of frequency representation at the appointed time in section,
At the appointed time there is time of emergency turn in the driving procedure in section in the above-mentioned above-mentioned object to be sorted of emergency turn frequency representation
Number, above-mentioned general steering frequency indicate time of normal direction of rotation in driving procedure of the above-mentioned object to be sorted at the appointed time in section
At the appointed time there is furious driving in the driving procedure in section in number, the above-mentioned above-mentioned object to be sorted of furious driving frequency representation
The number of behavior；Wherein, above-mentioned designated time period can be set according to the actual situation.
It is understood that above-mentioned first attribute and the second attribute are only a kind of example, in the concrete realization, it is also possible to including
Driving behavior data and the driving behavior data of the 4th attribute of three attributes etc., the embodiment of the present application is not construed as limiting.
It is understood that the attribute of the above driving behavior data is only a kind of example, in the concrete realization, it is also possible to including it
His attribute, is no longer described in detail one by one here.
The embodiment of the present application, can be with each driving behavior of comprehensive consideration by the driving behavior data of the multiple attributes of acquisition
Data treat the influence of object of classification, help to improve the accuracy of calculating.
Optionally, the driving behavior data of above-mentioned acquisition object to be sorted may include:
The driving behavior of above-mentioned object to be sorted is monitored, above-mentioned driving behavior data are captured；
Alternatively, obtaining the driving behavior data of object to be sorted from storage device.
The above-mentioned above-mentioned driving behavior data of capture may include the above-mentioned object to be sorted that acquisition speed sensor is captured
Furious driving number in driving procedure, and obtain shadow of the object to be sorted captured by camera in driving procedure
Picture, and determine by schema extraction and algorithm for pattern recognition the number of above-mentioned object to be sorted emergency turn in driving procedure.
Above-mentioned storage device may include independent database, memory entrained by storage server and other terminal devices, on
State the driving behavior data that storage device stores object to be sorted.
Specifically, the data processing equipment can obtain the driving behavior of object to be sorted at predetermined intervals
Data；Alternatively, the data processing equipment can also obtain driving school's behavioral data of the object to be sorted with scheduled frequency；
Or the data processing equipment can also be in the case where receiving acquisition instruction, to obtain the driving of the object to be sorted
Behavioral data.Wherein, which can be the acquisition instruction for user's input that the data processing equipment receives, alternatively,
After the acquisition instruction can also be for other equipment triggering, acquired in the data processing equipment.It is specially as the other equipment
What equipment, the embodiment of the present application are not construed as limiting, if the other equipment can be the mobile phone connecting with the data processing equipment,
Or bracelet etc..
102, the driving behavior data respectively according to the driving behavior data of above-mentioned first attribute and above-mentioned second attribute are true
Fixed first reliability and the second reliability；Wherein, above-mentioned first reliability is that above-mentioned object to be sorted belongs to the first driving behavior classification
Reliability, above-mentioned second reliability are the reliability that above-mentioned object to be sorted belongs to the second driving behavior classification.
Above-mentioned reliability indicates that above-mentioned object to be sorted corresponds to the probability of driving behavior classification.Above-mentioned driving behavior classification
Represent the degree of danger of driving behavior habit.
Specifically, above-mentioned driving behavior classification may include the type of complying with the rules and the type that varies from a rule, the above-mentioned type that complies with the rules
Indicate that the driving behavior of above-mentioned object to be sorted meets traffic rules, the above-mentioned type of varying from a rule indicates above-mentioned object to be sorted
Traffic rules are violated in driving behavior.It further, is to distinguish different driving school's behavior types, therefore above-mentioned driving in more detail
Behavior classification can also include safety and stability type, dare for one in type, impulsive style, violation operation type and dangerous driving type or
Person is multinomial.
Specifically, available disclosed driving behavior data, determine above-mentioned driving row according to above-mentioned driving behavior data
For pre-set interval corresponding to classification, driving behavior classification is quantified with realizing, improves the accuracy of calculating.For example, can
To obtain the driving behavior data announced on network by web crawlers technology, above-mentioned driving behavior data include driving in violation of rules and regulations
Data and the driving data that complies with the rules；Above-mentioned distribution is divided into different by the distribution for calculating above-mentioned driving behavior data
Section；According to each section definition driving behavior classification.Above-mentioned driving behavior classification may include safety and stability type, dare for type,
One or more in impulsive style, violation operation type and dangerous driving type.
In the embodiment of the present application, driving according to the driving behavior data of above-mentioned first attribute and above-mentioned second attribute respectively
It sails behavioral data and determines the first reliability and the second reliability, it will be appreciated that are as follows: according to driving behavior data of above-mentioned first attribute and upper
The driving behavior data for stating the second attribute determine the first reliability, and according to the driving behavior data of above-mentioned first attribute and upper
The driving behavior data for stating the second attribute determine the second reliability.
103, reliability maximum value is determined from above-mentioned first reliability and above-mentioned second reliability, by above-mentioned reliability maximum value
Corresponding driving behavior classification is determined as the driving behavior classification of above-mentioned object to be sorted.
Specifically, including that the first driving behavior classification, the second driving behavior classification and third drive in driving behavior classification
In the case where behavior classification, available first reliability of the data processing equipment, the second reliability and third reliability.The number as a result,
A reliability maximum value can be determined from first reliability, the second reliability and third reliability according to processing unit, by above-mentioned letter
Driving behavior classification corresponding to degree maximum value is determined as the driving behavior classification of object to be sorted, by above-mentioned reliability maximum value institute
Corresponding driving behavior classification is determined as the driving behavior classification of object to be sorted.
Optionally, after determining above-mentioned driving behavior classification, above-mentioned driving behavior classification information can be exported to aobvious
Show device, above-mentioned driving behavior classification information is shown to user by aforementioned display device.
It optionally, can also be in the driving behavior classification in preset danger after obtaining above-mentioned driving behavior classification
In the case where class scope, output safety prompt information, above-mentioned safety instruction information is for prompting object to be sorted to pay attention to traffic
Rule.
It, can be with that is, after data processing equipment determines the corresponding driving behavior classification of reliability maximum value
Export the corresponding driving behavior classification of the reliability maximum value.
Implement the embodiment of the present application, the driving behavior data of the multiple attributes of object to be sorted is obtained, according to the driving row
The reliability that above-mentioned object to be sorted belongs to above-mentioned driving behavior classification is calculated for data, determines object to be sorted according to reliability
The accuracy of determining driving behavior classification can be improved in driving behavior classification.
Referring to Fig. 2, Fig. 2 is the flow diagram of another data processing method provided by the embodiments of the present application, the party
Method includes:
201, the driving behavior data of object to be sorted are obtained, include driving for the first attribute in above-mentioned driving behavior data
Sail the driving behavior data of behavioral data and the second attribute.
202, the driving behavior data of above-mentioned first attribute and the driving behavior data pair of above-mentioned second attribute are calculated separately
The first Making by Probability Sets should be obtained to the probability of the first driving behavior classification；And calculate separately the driving row of above-mentioned first attribute
The probability that the second driving behavior classification is corresponded to for the driving behavior data of data and above-mentioned second attribute, obtains the second probability
Set.
Specifically, the driving row of above-mentioned the driving behavior data for calculating separately above-mentioned first attribute and above-mentioned second attribute
The probability that the first driving behavior classification is corresponded to for data, obtains the first Making by Probability Sets；And calculate separately above-mentioned first attribute
Driving behavior data and the driving behavior data of above-mentioned second attribute correspond to the probability of above-mentioned second driving behavior classification,
Obtain the second Making by Probability Sets, comprising:
The driving behavior data of the driving behavior data of above-mentioned first attribute and above-mentioned second attribute are input to respectively pre-
If in Bayesian rough set model, the positive domain set of output first and the second positive domain set；
The first support set is calculated according to the above-mentioned first positive domain set, and is gathered according to the above-mentioned second positive domain
Calculate the second support set；
The first confidence gain function set is calculated according to the above-mentioned first positive domain set, and just according to above-mentioned second
Domain set calculates the second confidence gain function set；
Above-mentioned first probability set is calculated according to above-mentioned first support set and above-mentioned first confidence gain function set
It closes, and above-mentioned second Making by Probability Sets is calculated according to above-mentioned second support set and above-mentioned second confidence gain function set.
Specifically, above-mentioned first positive domain collection is combined into the driving of the driving behavior data and the second attribute of above-mentioned first attribute
Behavioral data respectively corresponds the set to the positive domain of the first driving behavior classification, and above-mentioned second positive domain collection is combined into above-mentioned first and belongs to
The driving behavior data of property and the driving behavior data of the second attribute respectively correspond the collection to the positive domain of the second driving behavior classification
It closes, wherein positive domain is the parameter of above-mentioned preset Bayesian rough set model.Correspondingly, above-mentioned first support collection is combined into
The driving behavior data of the driving behavior data and the second attribute of stating the first attribute are respectively corresponded to the first driving behavior classification
The set of support, above-mentioned second support collection are combined into the driving behavior data of above-mentioned first attribute and the driving of the second attribute
Behavioral data respectively corresponds the set of the support to the second driving behavior classification.Correspondingly, above-mentioned first confidence gain function
Collection is combined into the driving behavior data of above-mentioned first attribute and the driving behavior data of the second attribute respectively correspond to first and drive row
For the set of the confidence gain function of classification, above-mentioned second confidence gain function collection is combined into the driving behavior of above-mentioned first attribute
Data and the driving behavior data of the second attribute respectively correspond the set of the confidence gain function to the second driving behavior classification.
Specifically, above-mentioned preset Bayesian rough set model is thick to traditional Bayes based on confidence gain function
Rough collection model improves obtained Bayesian rough set model, below to above-mentioned preset Bayesian rough set model into
Row is explained.
Traditional Bayesian rough set model can handle the case where two-value decision better, but work as decision classification
When being multiple, traditional rough set model can be such that the same object to be sorted is divided into in different classes of, violate practical feelings
Condition.The embodiment of the present application improves above-mentioned traditional Bayesian rough set model using confidence gain function, is preset
Bayesian rough set model.Above-mentioned preset Bayesian rough set model includes positive domainNegative domain
And Boundary RegionThree parameters, wherein d is the subset of driving behavior category set D, and C is driving behavior data
Set, is the calculation formula of three parameters below:
Wherein, U is object of classification set, and U includes object to be sorted and history object of classification, and E is U at equivalence relation C
Division, g (d | E) is confidence gain function, above-mentioned confidence gain function formula are as follows:
Above-mentioned confidence gain function indicates the probability increase of event d generation or the degree of reduction after introducing event E.Wherein,The radix of card (d) expression set d.
Specifically, above-mentioned respectively by the driving behavior of the driving behavior data of above-mentioned first attribute and above-mentioned second attribute
Data are input in default Bayesian rough set model, and the positive domain set of output first and the second positive domain set can specifically include
Following steps:
1) history object of classification, the driving behavior data of above-mentioned history object of classification and history object of classification are obtained
Driving behavior classification；
2) history information table is established according to the driving behavior data of above-mentioned history object of classification and above-mentioned history object of classification
Lattice, historical information table are as follows:
Table 1
| U' | c1 | ... | cn | D |
| e1 | ||||
| e2 | ||||
| e3 | ||||
| ... | ||||
| en |
U' in historical information table is history object of classification, and c1~cn is the driving behavior data of history object of classification,
D is driving behavior classification, and e1~en is history object of classification, and the driving behavior classification of the history object of classification is known.
3) above-mentioned object to be sorted and above-mentioned driving behavior data are added in above-mentioned historical information table, are determined
Plan information form.
4) object of classification, driving behavior data and the driving behavior classification in above-mentioned decision information table are inputted above-mentioned
In the calculation formula in positive domain, the first positive domain set and the second positive domain set are obtained.Wherein, which includes to be sorted right
As with history object of classification.
Optionally, above-mentioned that first support set is calculated according to the above-mentioned first positive domain set, and just according to above-mentioned second
Domain set calculates the second support set, comprising:
Above-mentioned first positive domain set and the above-mentioned second positive domain set are inputted in support formula respectively, export first
Degree of holding set and the second support set.
Above-mentioned support formula are as follows:
Wherein, dl is first of driving behavior classification, and ci is the driving behavior data of ith attribute, above-mentioned support table
Show and obtains the specific gravity that the object of classification definitely classified accounts for all object of classification in above-mentioned positive domain.
Optionally, above-mentioned that first confidence gain function set is calculated according to the above-mentioned first positive domain set, and according to
Above-mentioned second positive domain set calculates the second confidence gain function set, comprising: respectively just by the above-mentioned first positive domain set and second
Domain set inputs in above-mentioned confidence gain function formula, exports above-mentioned first confidence gain function set and the second confidence gain
Function set；Wherein, above-mentioned confidence gain function formula are as follows:
Wherein, d=U/ (D=dl),Wherein U/ (D=dl) indicates set U in driving behavior classification dl
Under division.
Specifically, above-mentioned above-mentioned according to above-mentioned first support set and the calculating of above-mentioned first confidence gain function set
First Making by Probability Sets, and above-mentioned the is calculated according to above-mentioned second support set and above-mentioned second confidence gain function set
Two Making by Probability Sets, comprising:
Above-mentioned first support set and above-mentioned first confidence gain function set are normalized respectively, obtained
To the first confidence gain function set of the first support set of normalization and normalization；And respectively by above-mentioned second support
Set and above-mentioned second confidence gain function set are normalized, and obtain the second support set of normalization and normalizing
Change the second confidence gain function set；
According to the weighting of the first confidence gain function set of the first support set of the normalization and the normalization
With obtain the first Making by Probability Sets；And according to the second confidence gain of the second support set of the normalization and the normalization
The weighted sum of function set obtains the second Making by Probability Sets.
The normalization formula of above-mentioned first support set and above-mentioned second support set are as follows:
Wherein,Indicate that the driving behavior data of the i-th attribute belong to the normalizing of l driving behavior classification (i.e. dl)
Change support,Indicate that above-mentioned driving behavior data are not belonging to the normalization support of any driving behavior classification, Θ table
Show the case where being not belonging to any driving behavior classification.
The normalization formula of above-mentioned first confidence gain function set and above-mentioned second confidence gain function set are as follows:
Wherein,To normalize confidence gain function.
Optionally, above-mentioned according to the first confidence gain function of the first support set of the normalization and the normalization
The weighted sum of set obtains the first Making by Probability Sets；And according to the second support set of the normalization and the normalization the
The weighted sum of two confidence gain function set, obtains the second Making by Probability Sets, comprising: by above-mentioned first support set and first
Confidence gain function set is weighted and averaged processing with identical weight, obtains the first Making by Probability Sets；And by above-mentioned second
Support set and the second confidence gain function set are weighted and averaged processing with identical weight, obtain the second probability set
It closes.The expression of above-mentioned weighted average processing are as follows:
M=(m(1)+m(2))/2
Wherein m is Making by Probability Sets.
In the embodiment of the present application, using the first support and the weighted sum of the first confidence gain function as the first probability
Set, and using the second support and the weighted sum of the second confidence gain function as the second Making by Probability Sets, can reduce calculating
Error improves the accuracy of calculating.
203, fusion treatment is carried out to above-mentioned first Making by Probability Sets by combining evidences rule, obtains above-mentioned first reliability；
And fusion treatment is carried out to above-mentioned second Making by Probability Sets by above-mentioned combining evidences rule, obtain above-mentioned second reliability.
Above-mentioned combining evidences rule belongs to a part in evidence theory, and above-mentioned evidence theory is information fusion technology
Middle one kind, the information for obtaining multisensor merge, and obtain fuse information, above-mentioned fuse information is often than single
Sensor information obtained is more accurate and has more globality.The specific manifestation form of above-mentioned combining evidences rule are as follows:
ForWherein, Θ is to include limited multiple Making by Probability Sets m1, m2, m3..., mnIdentification frame, on
State the combining evidences formula of limited multiple Making by Probability Sets are as follows:
Wherein,Φ is empty set, and M (dl) is that object to be sorted belongs to driving
The reliability of behavior classification dl.
Above-mentioned at least two Making by Probability Sets is inputted respectively in above-mentioned combining evidences rule, at least two reliabilities are exported.
In the embodiment of the present application, the first Making by Probability Sets and the second Making by Probability Sets are melted by evidence fusion theory
Conjunction processing, obtain the first reliability and the second reliability, can and globality more representative with the calculating of reliability, can be improved really
Determine the accuracy of driving behavior classification.
204, reliability maximum value is determined from above-mentioned first reliability and above-mentioned second reliability, by above-mentioned reliability maximum value
Corresponding driving behavior classification is determined as the driving behavior classification of above-mentioned object to be sorted.
It after step 203, obtains reliability M (d1), M (d2) ..., M (d3) take M (d1), M (d2) ..., M (dn)
In maximum value M (dt) corresponding to driving behavior classification of the dt as object to be sorted.
In the embodiment of the present application, by preset Bayesian rough set model and combining evidences rule to above-mentioned driving
Behavioral data is analyzed, it is determined that the driving behavior classification of above-mentioned object to be sorted improves determining driving behavior classification
Accuracy.
Referring to Fig. 3, Fig. 3 is a kind of concrete scene schematic diagram of data processing method provided by the embodiments of the present application.It answers
The understanding, the following contents describe the realization process that above-mentioned data processing method is directed to concrete scene, and the following contents should not
As the limitation to the application.As shown in figure 3, above-mentioned data processing method can include:
301, the driving behavior data of object to be sorted are obtained；
The driving behavior data of above-mentioned object to be sorted include furious driving frequency c1, and sideslip drives frequency c2, urgent to turn
To frequency c3 and general steering frequency c4.According to the different by above-mentioned driving behavior data divided rank of frequency, calculated with facilitating
And statistics.Wherein, if frequency F≤10, grade 1 is set by above-mentioned driving behavior data；It, will if frequency 10 < F≤50
Above-mentioned driving behavior data are set as grade 2；If frequency F > 50, grade 3 is set by above-mentioned driving behavior data.
302, driving behavior classification is defined；
Above-mentioned driving behavior classification includes safety and stability type d1, dares for type d2, impulsive style d3, violation operation type d4 and danger
Dangerous driving-type d5, wherein above-mentioned safety and stability type d1 indicates that the driving habit of car owner is safer, above-mentioned dangerous driving type d5
Indicate that the degree of danger of the driving habit of car owner is high, with the increase of serial number, the degree of danger of car owner's driving habit gradually on
It rises.
303, decision information table is established；
Obtain the driving behavior data of history object of classification and history object of classification, wherein above-mentioned history object of classification
Driving behavior classification be known.
According to the driving behavior data of history object of classification and history object of classification, historical information table is established；It will be wait divide
The driving behavior data of class object and object to be sorted and to be sorted are added in historical information table, obtain decision information table
The form of lattice, above-mentioned decision information table is as follows:
Table 2
304, above-mentioned decision information table is inputted in preset Bayesian rough set model, exports positive domain set；
Set d1 for the driving behavior classification of above-mentioned object to be sorted, i.e., by it is above-mentioned "? " place is set as d1, then will be upper
State the corresponding driving behavior data c1~cn of safety and stability type d1, safety and stability type d1 and the safety and stability type in decision table
The corresponding object of classification of d1 is input in preset Bayesian rough set model, availableWithThe calculation formula in positive domain please refers to 202.Wherein,Indicate that furious driving frequency c1 corresponds to the positive domain of safety and stability type d1, correspondingly,Indicate ci pairs
The positive domain of safety and stability type d1 should be arrived.
Similarly, it can acquire WithWherein,Indicate ci correspond to dare for
The positive domain of type d2,Indicate that ci corresponds to the positive domain of impulsive style d3,It indicates that ci is corresponded to grasp in violation of rules and regulations
Make the positive domain of type d4,Indicate that ci (such as table 2, i takes 1~4) corresponds to the positive domain of dangerous driving type d5.
305, support set is calculated according to above-mentioned positive domain set；
By positive domainIt inputs in support formula respectively, exports support WithAbove-mentioned support formula please refers to 202.Wherein,Indicate that furious driving frequency c1 corresponds to safety and stability type
The support of d1, correspondingly,Indicate that ci (c1 to c4 in such as table 2) corresponds to the support of safety and stability type d1.
Similarly, it can acquireWithWherein,Indicate ci pairs
The support dared for type d2 should be arrived,Indicate that ci corresponds to the support of impulsive style d3,It indicates that ci is corresponded to grasp in violation of rules and regulations
Make the support of type d4,Indicate that ci corresponds to the support of dangerous driving type d5.
306, confidence gain function set is calculated according to above-mentioned positive domain set；
By positive domainIt inputs in confidence gain function formula respectively, acquires confidence gain letter
Number maxgc1(d1)、maxgc2(d1)、maxgc3(d1) and maxgc4(d1).Wherein, the calculation formula of above-mentioned confidence gain function
Are as follows:
Wherein, d=U/ (D=dl),U/ (D=dl) indicates set U at driving behavior classification dl
Division.Wherein, maxgc1(d1) indicate that furious driving frequency c1 corresponds to the confidence gain function of safety and stability type d1, accordingly
, maxgci(d1) indicate that ci corresponds to the confidence gain function of safety and stability type d1.
Similarly, maxg can be acquiredc1(d2)~maxgc4(d2)、maxgc1(d3)~maxgc4(d3)、 maxgc1(d4)~
maxgc4(d4) and maxgc1(d5)~maxgc4(d5).Wherein, maxgci(d2) it indicates that ci is corresponded to dare to increase for the confidence of type d2
Beneficial function, maxgci(d3) indicate that ci corresponds to the confidence gain function of impulsive style d3, maxgci(d4) expression ci is corresponded to separated
Advise the confidence gain function of manipulation type d4, maxgci(d5) indicate that ci corresponds to the confidence gain function of dangerous driving type d5.
307, every driving behavior data difference is calculated according to above-mentioned confidence gain function set and above-mentioned support set
Correspond to the Making by Probability Sets of every driving behavior classification；
Above-mentioned support is normalized, normalization support is obtained；Wherein, the formula of normalized are as follows:
Wherein,For the support after normalization, indicate that ci corresponds to the support of dl,Indicate above-mentioned
Driving behavior data are not belonging to the normalization support of any driving behavior classification, and Θ expression is not belonging to any driving behavior class
Other situation.
Above-mentioned confidence gain function is normalized, normalization confidence gain function is obtained；Wherein, it normalizes
Handle formula are as follows:
Wherein,To normalize confidence gain function, indicate that ci corresponds to the confidence gain function of dl.
Above-mentioned first support set and the first confidence gain function set are weighted and averaged place with identical weight
Reason obtains Making by Probability Sets, the expression of above-mentioned weighted average processing are as follows:
Wherein, mi(dl) indicate that ci corresponds to the probability of dl (such as table 2, l takes 1~5).By above-mentioned steps, acquisition it is general
Rate set includes: m1(d1)、m2(d1)、m3(d1)、m4(d1)、m1(d2)、m2(d2)、m3(d2)、m4(d2)、 m1(d3)、m2
(d3)、m3(d3)、m4(d3)、m1(d4)、m2(d4)、m3(d4)、m4(d4)、m1(d5)、m2(d5)、 m3(d5)、m4(d5)、m1
(Θ)、m2(Θ)、m3(Θ) and m4(Θ)。
308, above-mentioned Making by Probability Sets is inputted in combining evidences rule, obtains above-mentioned object to be sorted and belongs to each driving
The reliability of behavior classification；
Above-mentioned combining evidences rule are as follows: forWherein, Θ is to include limited multiple Making by Probability Sets m1, m2,
m3..., mnIdentification frame, the combining evidences formula of above-mentioned limited multiple Making by Probability Sets are as follows:
Wherein,Φ is empty set, and M (dl) belongs to dl for object to be sorted and drives
Sail the reliability of behavior classification.
By m1(d1)、m2(d1)、m3(d1) and m4(d1) it inputs in above-mentioned combining evidences rule, it is above-mentioned to be sorted right to export
Reliability M (d1) as belonging to driving behavior classification d1.
M (d2), M (d3), M (d4), M (d5) and M (Θ) can similarly be acquired.
309, reliability maximum value is determined from each reliability, by the corresponding driving behavior classification of above-mentioned reliability maximum value
It is determined as the driving behavior classification of above-mentioned object to be sorted.
It is maximized M (dt) from M (d2), M (d2), M (d3), M (d4), M (d5) and M (Θ), the maximum value M (dt) institute
Corresponding dt is the driving behavior classification of object to be sorted.
In the embodiment of the present application, the driving behavior data for obtaining four attributes of object to be sorted, according to above-mentioned driving row
The reliability that above-mentioned object to be sorted is belonging respectively to five driving behavior classifications is calculated for data, has been determined according to reliability wait divide
The driving behavior classification of class object improves the accuracy of determining driving behavior classification.
Referring to Fig. 4, Fig. 4 is a kind of structural schematic diagram of data processing equipment provided by the embodiments of the present application, above-mentioned dress
It sets and includes:
Acquiring unit 401 includes in above-mentioned driving behavior data for obtaining the driving behavior data of object to be sorted
The driving behavior data of first attribute and the driving behavior data of the second attribute；
First determination unit 402, for respectively according to the driving behavior data of above-mentioned first attribute and above-mentioned second attribute
Driving behavior data determine the first reliability and the second reliability；Wherein, above-mentioned first reliability is that above-mentioned object to be sorted belongs to the
The reliability of one driving behavior classification, above-mentioned second reliability are the reliability that above-mentioned object to be sorted belongs to the second driving behavior classification；
Second determination unit 403, for determining reliability maximum value from first reliability and second reliability,
The corresponding driving behavior classification of the reliability maximum value is determined as to the driving behavior classification of the object to be sorted.
As shown in figure 5, above-mentioned first determination unit 402 includes:
First computation subunit 4021, for calculating separately the driving behavior data and described second of first attribute
The driving behavior data of attribute correspond to the probability of the first driving behavior classification, obtain the first Making by Probability Sets；And respectively
The driving behavior data of the driving behavior data and second attribute that calculate first attribute correspond to described second and drive
The probability of behavior classification obtains the second Making by Probability Sets；
Second computation subunit 4022 is used for according to first Making by Probability Sets calculating, first reliability, and according to
Second Making by Probability Sets calculates second reliability.
Above-mentioned second computation subunit 4022 is specifically used for carrying out above-mentioned first Making by Probability Sets by combining evidences rule
Fusion treatment obtains above-mentioned first reliability；And above-mentioned second Making by Probability Sets is merged by above-mentioned combining evidences rule
Processing, obtains above-mentioned second reliability.
Above-mentioned first computation subunit 4021 is specifically used for driving behavior data of above-mentioned first attribute and above-mentioned respectively
The driving behavior data of second attribute are input in default Bayesian rough set model, and the positive domain set of output first and second is just
Domain set；The first support set is calculated according to the above-mentioned first positive domain set, and total according to the above-mentioned second positive domain collection
Calculate the second support set；The first confidence gain function set is calculated according to the above-mentioned first positive domain set, and according to upper
It states the second positive domain set and calculates the second confidence gain function set；According to above-mentioned first support set and above-mentioned first confidence
Gain function set calculates above-mentioned first Making by Probability Sets, and is increased according to above-mentioned second support set and above-mentioned second confidence
Beneficial function set calculates above-mentioned second Making by Probability Sets.
Above-mentioned first computation subunit 4021 specifically is also used to respectively set above-mentioned first support set and above-mentioned first
Letter gain function set is normalized, and obtains the first support set of normalization and normalizes the first confidence gain letter
Manifold is closed；And above-mentioned second support set and above-mentioned second confidence gain function set are normalized respectively,
It obtains the second support set of normalization and normalizes the second confidence gain function set.
Above-mentioned first computation subunit 4021 is specifically also used to according to the first support set of the normalization and described returns
One changes the weighted sum of the first confidence gain function set, obtains the first Making by Probability Sets；And it is supported according to the normalization second
The weighted sum of the second confidence gain function set of degree set and the normalization, obtains the second Making by Probability Sets.
In the embodiment of the present application, above-mentioned first computation subunit 4021 is by the first support and the first confidence gain letter
Several weighted sums is as the first Making by Probability Sets, and using the weighted sum of the second support and the second confidence gain function as the
Two Making by Probability Sets can reduce calculating error, and then improve the accuracy for determining driving behavior classification.
It is understood that the specific implementation of Fig. 4 and data processing equipment shown in fig. 5 reference may also be made to Fig. 1, Fig. 2 and Fig. 3
Shown in method, be no longer described in detail one by one here.
Implement the embodiment of the present application, can determine the driving behavior classification of object to be sorted, improves and determine driving behavior class
Other accuracy.
Referring to Fig. 6, Fig. 6 is the structural schematic diagram of another data processing equipment provided by the embodiments of the present application.The dress
Setting includes: at least one processor 601, such as central processing unit (central processing unit, CPU), and at least one
A memory 602, at least one transceiver 603 and at least one bus 604.Wherein, above-mentioned bus 604 can be one group simultaneously
Capable data line, for realizing the interconnection of above-mentioned processor 601, above-mentioned memory 602 and above-mentioned transceiver 603；It is above-mentioned
Memory 602 can be high-speed random access memory (random access memory, RAM), be also possible to non-volatile
Memory (non-volatile memory), for example, at least a read-only memory (read only memory, ROM).
Optionally, above-mentioned transceiver 603, can be used for obtaining driving behavior data, and to terminal device or server
Send driving behavior classification information and safety instruction information.
Optionally, which can also include display, which can be used for showing reliability maximum value pair
Driving behavior classification answered etc., the embodiment of the present application is not construed as limiting.
In one possible implementation, program instruction, above-mentioned processor 601 be can store in above-mentioned memory 602
It can be used for method shown in caller instruction execution Fig. 1, Fig. 2 and Fig. 3.
For another example, which can also be used to execute side shown in acquiring unit, the first determination unit and the second determination unit
Method.Alternatively, in some implementations, transceiver can also be used for executing method shown in acquiring unit etc., and the application is implemented
Example is not construed as limiting.
Those of ordinary skill in the art will appreciate that all or part of the steps in the various methods of above-described embodiment is can
It is completed with instructing relevant hardware by program, which can be stored in a computer readable storage medium, storage
Medium include read-only memory (read only memory, ROM), random access memory (random access memory,
RAM), programmable read only memory (programmable read only memory, PROM), erasable programmable is read-only deposits
Reservoir (erasable programmable read only memory, EPROM), disposable programmable read-only memory (one-
Time programmable read-only memory, OTPROM), the electronics formula of erasing can make carbon copies read-only memory
(electrically-erasable programmable read-only memory, EEPROM), CD-ROM (compact
Disc read-only memory, CD-ROM) or other disc memories, magnetic disk storage, magnetic tape storage or can
For carrying or any other computer-readable medium of storing data.
A kind of data processing method and device disclosed in the embodiment of the present application are described in detail above, herein
Applying specific case, the principle and implementation of this application are described, and the explanation of above example is only intended to sides
Assistant solves the present processes and its core concept；At the same time, for those skilled in the art, the think of according to the application
Think, there will be changes in the specific implementation manner and application range.In conclusion the content of the present specification should not be construed as
Limitation to the application.
Claims (10)
1. a kind of data processing method characterized by comprising
The driving behavior data of object to be sorted are obtained, include the driving behavior data of the first attribute in the driving behavior data
With the driving behavior data of the second attribute；
The first letter is determined according to the driving behavior data of the driving behavior data of first attribute and second attribute respectively
Degree and the second reliability；Wherein, first reliability is the reliability that the object to be sorted belongs to the first driving behavior classification, described
Second reliability is the reliability that the object to be sorted belongs to the second driving behavior classification；
Reliability maximum value is determined from first reliability and second reliability, by the corresponding driving of the reliability maximum value
Behavior classification is determined as the driving behavior classification of the object to be sorted.
2. the method according to claim 1, wherein described respectively according to the driving behavior number of first attribute
The first reliability and the second reliability are determined according to the driving behavior data with second attribute, comprising:
The driving behavior data of the driving behavior data and second attribute that calculate separately first attribute correspond to described
The probability of first driving behavior classification, obtains the first Making by Probability Sets；
And calculate separately the driving behavior data of first attribute and the driving behavior data of second attribute correspond to
The probability of the second driving behavior classification, obtains the second Making by Probability Sets；
First reliability is calculated according to first Making by Probability Sets, and calculates described second according to second Making by Probability Sets
Reliability.
3. according to the method described in claim 2, it is characterized in that, the driving behavior number for calculating separately first attribute
The probability that the first driving behavior classification is corresponded to according to the driving behavior data with second attribute, obtains the first probability set
It closes；And calculate separately the driving behavior data of first attribute and the driving behavior data of second attribute correspond to institute
The probability for stating the second driving behavior classification, obtains the second Making by Probability Sets, comprising:
The driving behavior data of the driving behavior data of first attribute and second attribute are input to default shellfish respectively
In this rough set model of leaf, the positive domain set of output first and the second positive domain set；
The first support set is calculated according to the described first positive domain set, and calculates second according to the described second positive domain set
Degree of holding set；
The first confidence gain function set is calculated according to the described first positive domain set, and is calculated according to the described second positive domain set
Second confidence gain function set；
First Making by Probability Sets is calculated according to the first support set and the first confidence gain function set, and
Second Making by Probability Sets is calculated according to the second support set and the second confidence gain function set.
4. according to the method described in claim 3, it is characterized in that, described according to the first support set and the first confidence gain
The first Making by Probability Sets is calculated in function set, and is calculated according to the second support set and the second confidence gain function set
Obtain the second Making by Probability Sets, comprising:
The first support set and the first confidence gain function set are normalized respectively, obtain normalizing
Change the first confidence gain function set of the first support set and normalization；
And the second support set and the second confidence gain function set are normalized respectively, it obtains
Normalize the second confidence gain function set of the second support set and normalization；
According to the weighted sum of the first confidence gain function set of the first support set of the normalization and the normalization, obtain
First Making by Probability Sets；
And the weighted sum according to the second confidence gain function set of the second support set of the normalization and the normalization,
Obtain the second Making by Probability Sets.
5. according to method described in claim 2 to 4 any one, which is characterized in that described according to first Making by Probability Sets
First reliability is calculated, and second reliability is calculated according to second Making by Probability Sets, comprising:
Fusion treatment is carried out to first Making by Probability Sets by combining evidences rule, obtains the first reliability；
And fusion treatment is carried out to second Making by Probability Sets by the combining evidences rule, obtain the second reliability.
6. a kind of data processing equipment characterized by comprising
Acquiring unit includes the first attribute in the driving behavior data for obtaining the driving behavior data of object to be sorted
Driving behavior data and the second attribute driving behavior data；
First determination unit, for respectively according to the driving row of the driving behavior data of first attribute and second attribute
The first reliability and the second reliability are determined for data；Wherein, first reliability is that the object to be sorted belongs to the first driving row
For the reliability of classification, second reliability is the reliability that the object to be sorted belongs to the second driving behavior classification；
Second determination unit, for determining reliability maximum value from first reliability and second reliability, by the letter
The corresponding driving behavior classification of degree maximum value is determined as the driving behavior classification of the object to be sorted.
7. method according to claim 6, which is characterized in that first determination unit includes:
First computation subunit, for calculating separately the driving behavior data of first attribute and the driving of second attribute
Behavioral data corresponds to the probability of the first driving behavior classification, obtains the first Making by Probability Sets；And calculate separately described
The driving behavior data of one attribute and the driving behavior data of second attribute correspond to the second driving behavior classification
Probability obtains the second Making by Probability Sets；
Second computation subunit is used to calculate first reliability according to first Making by Probability Sets, and according to described second
Making by Probability Sets calculates second reliability.
8. method according to claim 7, which is characterized in that
Second computation subunit, specifically for being carried out at fusion by combining evidences rule to first Making by Probability Sets
Reason, obtains first reliability；And fusion treatment is carried out to second Making by Probability Sets by the combining evidences rule, it obtains
To second reliability.
9. a kind of data processing equipment, which is characterized in that including processor and memory；The processor and the memory are logical
Cross bus interconnection；Wherein, for the memory for storing computer program, the computer program includes program instruction,
The processor is configured for calling described program instruction, executes such as method described in any one of claim 1 to 5.
10. a kind of computer readable storage medium, which is characterized in that the computer-readable recording medium storage has computer journey
Sequence, the computer program include program instruction, and described program instructs when being executed by a processor, execute the processor such as
Method described in any one of claim 1 to 5.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201811149626.3A CN109447127A (en) | 2018-09-29 | 2018-09-29 | Data processing method and device |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201811149626.3A CN109447127A (en) | 2018-09-29 | 2018-09-29 | Data processing method and device |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN109447127A true CN109447127A (en) | 2019-03-08 |
Family
ID=65545579
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201811149626.3A Pending CN109447127A (en) | 2018-09-29 | 2018-09-29 | Data processing method and device |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN109447127A (en) |
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN110082116A (en) * | 2019-03-18 | 2019-08-02 | 深圳市元征科技股份有限公司 | A kind of evaluation method, evaluating apparatus and the storage medium of vehicular four wheels location data |
Citations (19)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| JP2008126908A (en) * | 2006-11-22 | 2008-06-05 | Denso Corp | Driving behavior estimation method and device |
| CN101545026A (en) * | 2009-05-08 | 2009-09-30 | 大连理工大学 | Weight determining method of influence factor of oxygen blasting amount of converter steel smelting |
| CN102289660A (en) * | 2011-07-26 | 2011-12-21 | 华南理工大学 | Method for detecting illegal driving behavior based on hand gesture tracking |
| CN103818327A (en) * | 2013-11-22 | 2014-05-28 | 深圳先进技术研究院 | Method and device for analyzing driving behaviors |
| CN104484912A (en) * | 2014-11-24 | 2015-04-01 | 深圳市华宝电子科技有限公司 | Analysis method and analysis device of driving behaviors, and server |
| CN104732785A (en) * | 2015-01-09 | 2015-06-24 | 杭州好好开车科技有限公司 | Driving behavior analyzing and reminding method and system |
| CN105044556A (en) * | 2015-07-06 | 2015-11-11 | 河南理工大学 | Atom sparse evidence fusion-based adaptive fault line selection method of small current grounding faults |
| CN105374211A (en) * | 2015-12-09 | 2016-03-02 | 敏驰信息科技（上海）有限公司 | System and method for calculating driving risks and assisting automobile insurance pricing based on multi-source data |
| CN105679020A (en) * | 2016-01-29 | 2016-06-15 | 深圳市美好幸福生活安全系统有限公司 | Driving behavior analysis device and method |
| US20170102708A1 (en) * | 2015-10-08 | 2017-04-13 | Denso Corporation | Drive assist apparatus |
| CN107665520A (en) * | 2016-07-27 | 2018-02-06 | 上海博泰悦臻网络技术服务有限公司 | A kind of driving monitoring method and device based on vehicle operation data |
| CN107665579A (en) * | 2016-07-27 | 2018-02-06 | 上海博泰悦臻网络技术服务有限公司 | A kind of user's driving behavior monitoring method and device |
| CN107945311A (en) * | 2017-11-29 | 2018-04-20 | 深圳市轱辘车联数据技术有限公司 | A kind of method for early warning of dangerous driving behavior, device, storage medium and server |
| CN108108766A (en) * | 2017-12-28 | 2018-06-01 | 东南大学 | Driving behavior recognition methods and system based on Fusion |
| CN108205828A (en) * | 2016-12-20 | 2018-06-26 | 大陆汽车电子（连云港）有限公司 | Driver's driving habit analysis system and method |
| CN108288312A (en) * | 2017-03-06 | 2018-07-17 | 腾讯科技（深圳）有限公司 | Driving behavior determines method and device |
| CN108407814A (en) * | 2018-03-08 | 2018-08-17 | 辽宁工业大学 | A kind of driver characteristics discrimination method |
| CN108428343A (en) * | 2018-05-17 | 2018-08-21 | 长沙理工大学 | A kind of more vehicle driving behavior analysis and danger early warning method and system |
| CN108446824A (en) * | 2018-02-08 | 2018-08-24 | 深圳市赛格导航科技股份有限公司 | A kind of methods of risk assessment of driving behavior, device, equipment and storage medium |
-
2018
- 2018-09-29 CN CN201811149626.3A patent/CN109447127A/en active Pending
Patent Citations (19)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| JP2008126908A (en) * | 2006-11-22 | 2008-06-05 | Denso Corp | Driving behavior estimation method and device |
| CN101545026A (en) * | 2009-05-08 | 2009-09-30 | 大连理工大学 | Weight determining method of influence factor of oxygen blasting amount of converter steel smelting |
| CN102289660A (en) * | 2011-07-26 | 2011-12-21 | 华南理工大学 | Method for detecting illegal driving behavior based on hand gesture tracking |
| CN103818327A (en) * | 2013-11-22 | 2014-05-28 | 深圳先进技术研究院 | Method and device for analyzing driving behaviors |
| CN104484912A (en) * | 2014-11-24 | 2015-04-01 | 深圳市华宝电子科技有限公司 | Analysis method and analysis device of driving behaviors, and server |
| CN104732785A (en) * | 2015-01-09 | 2015-06-24 | 杭州好好开车科技有限公司 | Driving behavior analyzing and reminding method and system |
| CN105044556A (en) * | 2015-07-06 | 2015-11-11 | 河南理工大学 | Atom sparse evidence fusion-based adaptive fault line selection method of small current grounding faults |
| US20170102708A1 (en) * | 2015-10-08 | 2017-04-13 | Denso Corporation | Drive assist apparatus |
| CN105374211A (en) * | 2015-12-09 | 2016-03-02 | 敏驰信息科技（上海）有限公司 | System and method for calculating driving risks and assisting automobile insurance pricing based on multi-source data |
| CN105679020A (en) * | 2016-01-29 | 2016-06-15 | 深圳市美好幸福生活安全系统有限公司 | Driving behavior analysis device and method |
| CN107665520A (en) * | 2016-07-27 | 2018-02-06 | 上海博泰悦臻网络技术服务有限公司 | A kind of driving monitoring method and device based on vehicle operation data |
| CN107665579A (en) * | 2016-07-27 | 2018-02-06 | 上海博泰悦臻网络技术服务有限公司 | A kind of user's driving behavior monitoring method and device |
| CN108205828A (en) * | 2016-12-20 | 2018-06-26 | 大陆汽车电子（连云港）有限公司 | Driver's driving habit analysis system and method |
| CN108288312A (en) * | 2017-03-06 | 2018-07-17 | 腾讯科技（深圳）有限公司 | Driving behavior determines method and device |
| CN107945311A (en) * | 2017-11-29 | 2018-04-20 | 深圳市轱辘车联数据技术有限公司 | A kind of method for early warning of dangerous driving behavior, device, storage medium and server |
| CN108108766A (en) * | 2017-12-28 | 2018-06-01 | 东南大学 | Driving behavior recognition methods and system based on Fusion |
| CN108446824A (en) * | 2018-02-08 | 2018-08-24 | 深圳市赛格导航科技股份有限公司 | A kind of methods of risk assessment of driving behavior, device, equipment and storage medium |
| CN108407814A (en) * | 2018-03-08 | 2018-08-17 | 辽宁工业大学 | A kind of driver characteristics discrimination method |
| CN108428343A (en) * | 2018-05-17 | 2018-08-21 | 长沙理工大学 | A kind of more vehicle driving behavior analysis and danger early warning method and system |
Non-Patent Citations (3)
| Title |
|---|
| 施志坚 等: "基于改进信息融合的航空发动机状态评估方法", 《航空计算技术》 * |
| 王新胜 等: "基于贝叶斯模型的驾驶行为识别与预测", 《通信学报》 * |
| 陈可 等: "基于改进贝叶斯粗糙集和证据理论的决策信息融合方法", 《计算机应用研究》 * |
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN110082116A (en) * | 2019-03-18 | 2019-08-02 | 深圳市元征科技股份有限公司 | A kind of evaluation method, evaluating apparatus and the storage medium of vehicular four wheels location data |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| WO2019154115A1 (en) | Resource transferring monitoring method and device | |
| CN107563757B (en) | Data risk identification method and device | |
| CN108665366A (en) | Determine method, terminal device and the computer readable storage medium of consumer's risk grade | |
| CN110830448B (en) | Target event flow abnormity detection method and device, electronic equipment and medium | |
| EP3038025A1 (en) | Retention risk determiner | |
| CN108446824A (en) | A kind of methods of risk assessment of driving behavior, device, equipment and storage medium | |
| CN108898476A (en) | A kind of loan customer credit-graded approach and device | |
| CN108711085A (en) | A kind of response method and its equipment of transaction request | |
| CN111062808B (en) | Credit card limit evaluation method, credit card limit evaluation device, computer equipment and storage medium | |
| CN109614414B (en) | User information determining method and device | |
| CN111062444A (en) | Credit risk prediction method, system, terminal and storage medium | |
| CN106897282B (en) | User group classification method and device | |
| CN110276178A (en) | A kind of risk control method, device and the equipment of identity-based verifying | |
| CN114090985A (en) | Method and device for setting user operation permission in block chain and electronic equipment | |
| CN113946590A (en) | Method, device and equipment for updating integral data and storage medium | |
| CN109308564A (en) | The recognition methods of crowd's performance ratings, device, storage medium and computer equipment | |
| CN108197795A (en) | The account recognition methods of malice group, device, terminal and storage medium | |
| CN109409648A (en) | Claims Resolution air control method, apparatus, computer equipment and storage medium | |
| CN109447127A (en) | Data processing method and device | |
| CN113869700A (en) | Performance index prediction method and device, electronic equipment and storage medium | |
| CN111325572B (en) | Data processing method and device | |
| CN115577983B (en) | Enterprise task matching method based on block chain, server and storage medium | |
| CN111476668A (en) | Identification method and device of credible relationship, storage medium and computer equipment | |
| CN108259363A (en) | A kind of method and device of staged service traffics control | |
| CN109544348A (en) | Assets security screening technique, equipment and computer readable 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 | ||
| RJ01 | Rejection of invention patent application after publication |
Application publication date: 20190308 |