CN113642600A - Driving behavior feature extraction method based on mRMR algorithm and principal component analysis - Google Patents

Driving behavior feature extraction method based on mRMR algorithm and principal component analysis Download PDF

Info

Publication number
CN113642600A
CN113642600A CN202110724482.5A CN202110724482A CN113642600A CN 113642600 A CN113642600 A CN 113642600A CN 202110724482 A CN202110724482 A CN 202110724482A CN 113642600 A CN113642600 A CN 113642600A
Authority
CN
China
Prior art keywords
data
vehicle
driving behavior
principal component
index
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.)
Granted
Application number
CN202110724482.5A
Other languages
Chinese (zh)
Other versions
CN113642600B (en
Inventor
何水龙
邹智宏
李超
冯海波
许恩永
姚柳成
王方圆
Current Assignee (The listed assignees may be inaccurate. Google has not performed a legal analysis and makes no representation or warranty as to the accuracy of the list.)
Guilin University of Electronic Technology
Dongfeng Liuzhou Motor Co Ltd
Original Assignee
Guilin University of Electronic Technology
Dongfeng Liuzhou Motor Co Ltd
Priority date (The priority date is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the date listed.)
Filing date
Publication date
Application filed by Guilin University of Electronic Technology, Dongfeng Liuzhou Motor Co Ltd filed Critical Guilin University of Electronic Technology
Priority to CN202110724482.5A priority Critical patent/CN113642600B/en
Publication of CN113642600A publication Critical patent/CN113642600A/en
Application granted granted Critical
Publication of CN113642600B publication Critical patent/CN113642600B/en
Active legal-status Critical Current
Anticipated expiration legal-status Critical

Links

Images

Classifications

    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06F—ELECTRIC DIGITAL DATA PROCESSING
    • G06F18/00—Pattern recognition
    • G06F18/20—Analysing
    • G06F18/21—Design or setup of recognition systems or techniques; Extraction of features in feature space; Blind source separation
    • G06F18/213—Feature extraction, e.g. by transforming the feature space; Summarisation; Mappings, e.g. subspace methods
    • G06F18/2135—Feature extraction, e.g. by transforming the feature space; Summarisation; Mappings, e.g. subspace methods based on approximation criteria, e.g. principal component analysis
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06F—ELECTRIC DIGITAL DATA PROCESSING
    • G06F17/00—Digital computing or data processing equipment or methods, specially adapted for specific functions
    • G06F17/10—Complex mathematical operations
    • G06F17/16—Matrix or vector computation, e.g. matrix-matrix or matrix-vector multiplication, matrix factorization
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06F—ELECTRIC DIGITAL DATA PROCESSING
    • G06F17/00—Digital computing or data processing equipment or methods, specially adapted for specific functions
    • G06F17/10—Complex mathematical operations
    • G06F17/18—Complex mathematical operations for evaluating statistical data, e.g. average values, frequency distributions, probability functions, regression analysis
    • 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
    • G06Q50/00—Information and communication technology [ICT] specially adapted for implementation of business processes of specific business sectors, e.g. utilities or tourism
    • G06Q50/40—Business processes related to the transportation industry

Landscapes

  • Engineering & Computer Science (AREA)
  • Physics & Mathematics (AREA)
  • General Physics & Mathematics (AREA)
  • Data Mining & Analysis (AREA)
  • Theoretical Computer Science (AREA)
  • Mathematical Physics (AREA)
  • Pure & Applied Mathematics (AREA)
  • Computational Mathematics (AREA)
  • Mathematical Optimization (AREA)
  • Mathematical Analysis (AREA)
  • General Engineering & Computer Science (AREA)
  • Databases & Information Systems (AREA)
  • Business, Economics & Management (AREA)
  • Bioinformatics & Cheminformatics (AREA)
  • Bioinformatics & Computational Biology (AREA)
  • Software Systems (AREA)
  • Evolutionary Biology (AREA)
  • Computer Vision & Pattern Recognition (AREA)
  • Life Sciences & Earth Sciences (AREA)
  • Algebra (AREA)
  • Human Resources & Organizations (AREA)
  • Economics (AREA)
  • Evolutionary Computation (AREA)
  • Computing Systems (AREA)
  • Probability & Statistics with Applications (AREA)
  • Health & Medical Sciences (AREA)
  • Operations Research (AREA)
  • General Health & Medical Sciences (AREA)
  • Artificial Intelligence (AREA)
  • Marketing (AREA)
  • Primary Health Care (AREA)
  • Strategic Management (AREA)
  • Tourism & Hospitality (AREA)
  • General Business, Economics & Management (AREA)
  • Traffic Control Systems (AREA)
  • Control Of Transmission Device (AREA)

Abstract

The invention discloses a driving behavior feature extraction method based on an mRMR algorithm and principal component analysis, which comprises the steps of acquiring Internet of vehicles data of operating vehicles on line based on a vehicle-mounted data acquisition terminal data management system; preprocessing the Internet of vehicles data, and performing data cleaning work and index data calculation; calculating each index data by mutual information, and calculating mutual information by using the index edge probability density and the index joint probability density; sequentially calculating the correlation and redundancy among the index data by using a forward sorting method to finish the sorting and selection of the importance of the mRMR characteristics; and extracting data information in the index by combining a principal component analysis method, and analyzing to obtain driving behavior information in the Internet of vehicles data. The invention reduces the redundancy among data indexes in the driving behavior analysis, and simultaneously reduces the data dimension, thereby improving the data use efficiency and providing an effective tool for better utilizing the vehicle networking data and extracting the driving behavior characteristics.

Description

Driving behavior feature extraction method based on mRMR algorithm and principal component analysis
Technical Field
The invention relates to the technical field of data analysis of the Internet of vehicles, in particular to a driving behavior feature extraction method based on an mRMR algorithm and principal component analysis.
Background
With the progress of internet technology, more and more vehicles are connected into the car networking system, and the networking and intelligent degree of the car is improved accordingly. However, how to make more efficient use of these internet of vehicles data is a major focus of current research. In particular, when driving behavior application analysis is performed using the related data, the index data often has characteristics that the information contained in the problem to be studied is unclear and redundancy is large.
Therefore, the method takes the problems of high dimension and redundancy in the use of the Internet of vehicles as a starting point, selects a characteristic optimization method to reduce the dimension of the Internet of vehicles and removes the data redundancy.
Disclosure of Invention
This section is for the purpose of summarizing some aspects of embodiments of the invention and to briefly introduce some preferred embodiments. In this section, as well as in the abstract and the title of the invention of this application, simplifications or omissions may be made to avoid obscuring the purpose of the section, the abstract and the title, and such simplifications or omissions are not intended to limit the scope of the invention.
The present invention has been made in view of the above-mentioned conventional problems.
Therefore, the technical problem solved by the invention is as follows: the problem of redundancy in the use of car networking data is solved.
In order to solve the technical problems, the invention provides the following technical scheme: the method comprises the steps that vehicle networking data of an operating vehicle are acquired on line based on a vehicle-mounted data acquisition terminal and a data management system; preprocessing the Internet of vehicles data, and performing data cleaning work and index data calculation; calculating each index data by mutual information, and calculating mutual information by using the index edge probability density and the index joint probability density; sequentially calculating the correlation and redundancy among the index data by using a forward sorting method to finish the sorting and selection of the importance of the mRMR characteristics; and extracting data information in the index by combining a principal component analysis method, and analyzing to obtain driving behavior information in the Internet of vehicles data.
As a preferable scheme of the driving behavior feature extraction method based on the mRMR algorithm and principal component analysis of the present invention, wherein: the vehicle networking data comprises a vehicle identification number, driving time, GPS longitude of a vehicle, GPS latitude of the position of the vehicle, GPS altitude of the position of the vehicle, total fuel consumption of an ECU of the vehicle, accumulated total fuel consumption of the vehicle, meter mileage of the vehicle, ECU speed of the vehicle, engine rotating speed of the vehicle, acceleration of the vehicle, engine torque load rate of the vehicle and engine load rate of the vehicle.
As a preferable scheme of the driving behavior feature extraction method based on the mRMR algorithm and principal component analysis of the present invention, wherein: the data cleaning work comprises time jump inspection, data abnormal value processing and data missing processing; the index data calculation comprises index calculation of relevant parameters of the engine class and the altitude, wherein the indexes comprise an engine load rate mean value, an engine torque load rate mean value, an engine rotating speed mean value, an altitude mean value, an engine load rate standard deviation, an engine torque load rate standard deviation, an engine rotating speed standard deviation and an altitude standard deviation; and carrying out index calculation on the driving behavior related parameters, wherein the indexes comprise a speed mean value, an acceleration mean value, an accelerator opening mean value, a gearbox rotating speed mean value, a speed standard deviation, an acceleration standard deviation, an accelerator opening standard deviation and a gearbox rotating speed standard deviation.
As a preferable scheme of the driving behavior feature extraction method based on the mRMR algorithm and principal component analysis of the present invention, wherein: calculating the mutual information includes calculating the mutual information by,
Figure BDA0003137307690000021
wherein, I (X, Y) is mutual information quantity between characteristic variables X and Y, X and Y are data variables, p (X) and p (Y) are edge probability distribution functions of X and Y respectively, and p (X, Y) represents a joint probability density function of X and Y.
As a preferable scheme of the driving behavior feature extraction method based on the mRMR algorithm and principal component analysis of the present invention, wherein: performing the mRMR feature importance ranking and selecting the average information correlation of the index data and the target data to be calculated, including,
Figure BDA0003137307690000022
wherein S represents a subset of the characteristic variables, c represents the target variable, fiD (S, c) is the average of the mutual information.
As a preferable scheme of the driving behavior feature extraction method based on the mRMR algorithm and principal component analysis of the present invention, wherein: and sequentially calculating the correlation and redundancy among the index data by utilizing a forward sorting method, including,
Figure BDA0003137307690000023
where R (S) is the minimum redundancy measure for the feature subset.
As a preferable scheme of the driving behavior feature extraction method based on the mRMR algorithm and principal component analysis of the present invention, wherein: calculating the mRMR characteristic values and the ranking of the target data and the index data comprises,
Figure BDA0003137307690000031
wherein, the mRMR is a measure integrating maximum mutual information and minimum redundancy.
As a preferable scheme of the driving behavior feature extraction method based on the mRMR algorithm and principal component analysis of the present invention, wherein: extracting data information in the index comprises performing KMO ball-type inspection and significance inspection on the index data; when the inspection data meet the requirements, extracting principal component characteristics, including calculating a correlation coefficient matrix and corresponding eigenvectors and eigenvalues thereof and expressing the correlation coefficient matrix as a principal component;
Figure BDA0003137307690000032
Figure BDA0003137307690000033
where k, n represent the dimension of the sample matrix, xijIs an element in the sample matrix, rho is the standard covariance matrix, ziIs extracted i principal components, lijAre the elements contained in the eigenvector of the data matrix.
As a preferable scheme of the driving behavior feature extraction method based on the mRMR algorithm and principal component analysis of the present invention, wherein: the driving behavior information includes integrating the features extracted by the mRMR algorithm with the features extracted by the principal component analysis.
The invention has the beneficial effects that: the invention reduces the redundancy among data indexes in the driving behavior analysis, and simultaneously reduces the data dimension, thereby improving the data use efficiency and providing an effective tool for better utilizing the vehicle networking data and extracting the driving behavior characteristics.
Drawings
In order to more clearly illustrate the technical solutions of the embodiments of the present invention, the drawings needed to be used in the description of the embodiments will be briefly introduced below, and it is obvious that the drawings in the following description are only some embodiments of the present invention, and it is obvious for those skilled in the art to obtain other drawings based on these drawings without inventive exercise. Wherein:
fig. 1 is a schematic flow chart of a driving behavior feature extraction method based on an mRMR algorithm and principal component analysis according to an embodiment of the present invention;
fig. 2 is a schematic flow chart of an mRMR algorithm of a driving behavior feature extraction method based on an mRMR algorithm and principal component analysis according to an embodiment of the present invention;
fig. 3 is a schematic frame diagram of a driving behavior feature extraction method based on an mRMR algorithm and principal component analysis according to an embodiment of the present invention.
Detailed Description
In order to make the aforementioned objects, features and advantages of the present invention comprehensible, specific embodiments accompanied with figures are described in detail below, and it is apparent that the described embodiments are a part of the embodiments of the present invention, not all of the embodiments. All other embodiments, which can be obtained by a person skilled in the art without making creative efforts based on the embodiments of the present invention, shall fall within the protection scope of the present invention.
In the following description, numerous specific details are set forth in order to provide a thorough understanding of the present invention, but the present invention may be practiced in other ways than those specifically described and will be readily apparent to those of ordinary skill in the art without departing from the spirit of the present invention, and therefore the present invention is not limited to the specific embodiments disclosed below.
Furthermore, reference herein to "one embodiment" or "an embodiment" means that a particular feature, structure, or characteristic described in connection with the embodiment is included in at least one implementation of the invention. The appearances of the phrase "in one embodiment" in various places in the specification are not necessarily all referring to the same embodiment, nor are separate or alternative embodiments mutually exclusive of other embodiments.
The present invention will be described in detail with reference to the drawings, wherein the cross-sectional views illustrating the structure of the device are not enlarged partially in general scale for convenience of illustration, and the drawings are only exemplary and should not be construed as limiting the scope of the present invention. In addition, the three-dimensional dimensions of length, width and depth should be included in the actual fabrication.
Meanwhile, in the description of the present invention, it should be noted that the terms "upper, lower, inner and outer" and the like indicate orientations or positional relationships based on the orientations or positional relationships shown in the drawings, and are only for convenience of describing the present invention and simplifying the description, but do not indicate or imply that the referred device or element must have a specific orientation, be constructed in a specific orientation and operate, and thus, cannot be construed as limiting the present invention. Furthermore, the terms first, second, or third are used for descriptive purposes only and are not to be construed as indicating or implying relative importance.
The terms "mounted, connected and connected" in the present invention are to be understood broadly, unless otherwise explicitly specified or limited, for example: can be fixedly connected, detachably connected or integrally connected; they may be mechanically, electrically, or directly connected, or indirectly connected through intervening media, or may be interconnected between two elements. The specific meanings of the above terms in the present invention can be understood in specific cases to those skilled in the art.
Example 1
Referring to fig. 1, 2 and 3, a driving behavior feature extraction method based on an mRMR algorithm and principal component analysis is provided as a first embodiment of the present invention, and specifically includes:
s1: and acquiring the Internet of vehicles data of the operating vehicles on line based on the vehicle-mounted data acquisition terminal data management system. It should be noted that the car networking data includes:
the vehicle identification number, the driving time, the GPS longitude of the vehicle, the GPS latitude of the position of the vehicle, the GPS altitude of the position of the vehicle, the total fuel consumption of an ECU (electronic control unit) of the vehicle, the accumulated total fuel consumption of the vehicle, the meter mileage of the vehicle, the ECU speed of the vehicle, the engine speed of the vehicle, the acceleration of the vehicle, the engine torque load rate of the vehicle and the engine load rate of the vehicle.
S2: and preprocessing the data of the Internet of vehicles, and performing data cleaning work and index data calculation. It should be noted that, the data cleaning includes:
time jump checking, data abnormal value processing and data missing processing;
the index data calculation comprises index calculation of relevant parameters of the engines and the altitude, wherein the indexes comprise an engine load rate mean value, an engine torque load rate mean value, an engine rotating speed mean value, an altitude mean value, an engine load rate standard deviation, an engine torque load rate standard deviation, an engine rotating speed standard deviation and an altitude standard deviation;
and carrying out index calculation on the driving behavior related parameters, wherein the indexes comprise a speed average value, an acceleration average value, an accelerator opening average value, a gearbox rotating speed average value, a speed standard deviation, an acceleration standard deviation, an accelerator opening standard deviation and a gearbox rotating speed standard deviation.
S3: and calculating each index data by using the mutual information, and calculating the mutual information by using the index edge probability density and the index joint probability density. It should be further noted that the calculating the mutual information includes:
Figure BDA0003137307690000061
wherein, I (X, Y) is mutual information quantity between characteristic variables X and Y, X and Y are data variables, p (X) and p (Y) are edge probability distribution functions of X and Y respectively, and p (X, Y) represents a joint probability density function of X and Y.
S4: and sequentially calculating the correlation and redundancy among the index data by using a forward sorting method to finish the sorting and selection of the importance of the mRMR characteristics. It should be further noted that, in this step, performing mRMR feature importance ranking and selecting the average information correlation between the index data to be calculated and the target data includes:
Figure BDA0003137307690000062
wherein S represents a subset of the characteristic variables, c represents the target variable, fiD (S, c) is the average of the mutual information.
Sequentially calculating the correlation and redundancy among index data by using a forward sorting method, wherein the method comprises the following steps:
Figure BDA0003137307690000063
where R (S) is the minimum redundancy measure for the feature subset.
The step of calculating the mRMR characteristic values and the sequence of the index data and the target data comprises the following steps:
Figure BDA0003137307690000064
wherein, the mRMR is a measure integrating maximum mutual information and minimum redundancy.
S5: and extracting data information in the index by combining a principal component analysis method, and analyzing to obtain driving behavior information in the Internet of vehicles data. It should be noted again that the data information in the extracted index includes:
performing KMO spherical inspection and significance inspection on the index data;
when the inspection data meet the requirements, performing main city inverse feature extraction, including calculating a correlation coefficient matrix and corresponding feature vectors and feature values thereof and expressing the correlation coefficient matrix as a main component;
Figure BDA0003137307690000071
Figure BDA0003137307690000072
where k, n represent the dimension of the sample matrix, xijIs an element in the sample matrix, rho is the standard covariance matrix, ziIs extracted i principal components, lijElements contained in the characteristic vector of the data matrix;
the driving behavior information includes features extracted by integrating the mRMR algorithm with features extracted by principal component analysis.
Example 2
Preferably, the present embodiment is different from the first embodiment in that:
(1) and acquiring the Internet of vehicles data of the operating vehicles on line based on the vehicle-mounted data acquisition terminal data management system.
The focus is on whether the following categories of data items are acquired: the vehicle identification number, the driving time, the GPS longitude of the vehicle, the GPS latitude of the position of the vehicle, the GPS altitude of the position of the vehicle, the total fuel consumption of an ECU (electronic control unit) of the vehicle, the accumulated total fuel consumption of the vehicle, the meter mileage of the vehicle, the ECU speed of the vehicle, the engine speed of the vehicle, the acceleration of the vehicle, the engine torque load rate of the vehicle and the engine load rate of the vehicle.
(2) And performing data cleaning work and index data calculation on the acquired Internet of vehicles data.
The method comprises the steps of time jump inspection, data abnormal value processing and data missing processing, and index calculation is carried out on relevant parameters of the engine and the altitude, and indexes mainly comprise an engine load rate mean value, an engine torque load rate mean value, an engine rotating speed mean value, an altitude mean value, an engine load rate standard deviation, an engine torque load rate standard deviation, an engine rotating speed standard deviation and an altitude standard deviation.
And carrying out index calculation on the driving behavior related parameters, wherein the indexes mainly comprise a speed average value, an acceleration average value, an accelerator opening average value, a gearbox rotating speed average value, a speed standard deviation, an acceleration standard deviation, an accelerator opening standard deviation and a gearbox rotating speed standard deviation.
And performing index edge probability density, index joint probability density and mutual information calculation on each calculated index data by using the following formula.
Figure BDA0003137307690000081
And sequentially calculating the correlation and redundancy among the indexes by using a forward sorting method so as to sort and select the importance of the mRMR characteristics.
The results of calculating the redundancy of relevance and the ranking of the mRMR importance are shown in the following table.
Table 1: mRMR importance ranking results
Figure BDA0003137307690000082
And (5) sorting the results of the importance of the mRMR, wherein the redundancy of the set containing the first six indexes and other indexes is minimum, and the last 10 indexes are selected for further extracting the features.
And further extracting data information in the index by using a principal component analysis method, firstly performing KMO spherical inspection and significance inspection on the data, and extracting principal component characteristics when the data is inspected to meet requirements. And is expressed as a principal component. The results are shown in Table 2.
Table 2: KMO ball-type test and significance test
Figure BDA0003137307690000083
Figure BDA0003137307690000091
KMO is 0.853 to 0.6, and the significance test F is less than 0.005, which indicates that the data meets the requirement of principal component analysis.
The correlation matrix is given in the calculation as shown in table 3.
Table 3: and (5) performing correlation analysis results after feature extraction.
Figure BDA0003137307690000092
Further, the contribution rate of the principal component in the data and the cumulative contribution rate thereof are calculated as shown in table 4 below, and when the number of the principal components is four, the extraction information rate reaches 88.02% (the information rate is greater than 80% by default, namely, the information rate can be used for principal component extraction), and can be used as the extraction characteristic of the original data index.
TABLE 4 Total contribution rate of extracted principal component and cumulative contribution rate
Figure BDA0003137307690000093
And synthesizing the results to extract the first four main components to replace the driving behavior information in the original data.
It should be noted that the above-mentioned embodiments are only for illustrating the technical solutions of the present invention and not for limiting, and although the present invention has been described in detail with reference to the preferred embodiments, it should be understood by those skilled in the art that modifications or equivalent substitutions may be made on the technical solutions of the present invention without departing from the spirit and scope of the technical solutions of the present invention, which should be covered by the claims of the present invention.

Claims (9)

1. A driving behavior feature extraction method based on an mRMR algorithm and principal component analysis is characterized by comprising the following steps: comprises the steps of (a) preparing a mixture of a plurality of raw materials,
acquiring Internet of vehicles data of an operating vehicle on line based on a vehicle-mounted data acquisition terminal data management system;
preprocessing the Internet of vehicles data, and performing data cleaning work and index data calculation;
calculating each index data by mutual information, and calculating mutual information by using the index edge probability density and the index joint probability density;
sequentially calculating the correlation and redundancy among the index data by using a forward sorting method to finish the sorting and selection of the importance of the mRMR characteristics;
and extracting data information in the index by combining a principal component analysis method, and analyzing to obtain driving behavior information in the Internet of vehicles data.
2. The driving behavior feature extraction method based on mRMR algorithm and principal component analysis according to claim 1, characterized in that: the internet of vehicles data includes at least one of,
the vehicle identification number, the driving time, the GPS longitude of the vehicle, the GPS latitude of the position of the vehicle, the GPS altitude of the position of the vehicle, the total fuel consumption of an ECU (electronic control unit) of the vehicle, the accumulated total fuel consumption of the vehicle, the meter mileage of the vehicle, the ECU speed of the vehicle, the engine speed of the vehicle, the acceleration of the vehicle, the engine torque load rate of the vehicle and the engine load rate of the vehicle.
3. The driving behavior feature extraction method based on the mRMR algorithm and principal component analysis according to claim 1 or 2, characterized in that: the data cleaning work comprises time jump inspection, data abnormal value processing and data missing processing;
the index data calculation comprises index calculation of relevant parameters of the engine class and the altitude, wherein the indexes comprise an engine load rate mean value, an engine torque load rate mean value, an engine rotating speed mean value, an altitude mean value, an engine load rate standard deviation, an engine torque load rate standard deviation, an engine rotating speed standard deviation and an altitude standard deviation;
and carrying out index calculation on the driving behavior related parameters, wherein the indexes comprise a speed mean value, an acceleration mean value, an accelerator opening mean value, a gearbox rotating speed mean value, a speed standard deviation, an acceleration standard deviation, an accelerator opening standard deviation and a gearbox rotating speed standard deviation.
4. The driving behavior feature extraction method based on mRMR algorithm and principal component analysis according to claim 3, characterized in that: calculating the mutual information includes calculating the mutual information by,
Figure FDA0003137307680000011
wherein, I (X, Y) is mutual information quantity between characteristic variables X and Y, X and Y are data variables, p (X) and p (Y) are edge probability distribution functions of X and Y respectively, and p (X, Y) represents a joint probability density function of X and Y.
5. The method for extracting driving behavior feature based on mRMR algorithm and principal component analysis according to claim 4, wherein: performing the mRMR feature importance ranking and selecting the average information correlation of the index data and the target data to be calculated, including,
Figure FDA0003137307680000021
wherein S represents a subset of the characteristic variables, c represents the target variable, fiD (S, c) is the average of the mutual information.
6. The driving behavior feature extraction method based on mRMR algorithm and principal component analysis according to claim 5, characterized in that: and sequentially calculating the correlation and redundancy among the index data by utilizing a forward sorting method, including,
Figure FDA0003137307680000022
where R (S) is the minimum redundancy measure for the feature subset.
7. The driving behavior feature extraction method based on mRMR algorithm and principal component analysis according to claim 6, characterized in that: calculating the mRMR characteristic values and the ranking of the target data and the index data comprises,
Figure FDA0003137307680000023
wherein, the mRMR is a measure integrating maximum mutual information and minimum redundancy.
8. The driving behavior feature extraction method based on mRMR algorithm and principal component analysis according to claim 7, characterized in that: the extracting of the data information in the indicator includes,
performing KMO ball type inspection and significance inspection on the index data;
when the inspection data meet the requirements, extracting principal component characteristics, including calculating a correlation coefficient matrix and corresponding eigenvectors and eigenvalues thereof and expressing the correlation coefficient matrix as a principal component;
Figure FDA0003137307680000031
Figure FDA0003137307680000032
where k, n represent the dimension of the sample matrix, xijIs an element in the sample matrix, rho is the standard covariance matrix, ziIs extracted i principal components, lijAre the elements contained in the eigenvector of the data matrix.
9. The driving behavior feature extraction method based on mRMR algorithm and principal component analysis according to claim 8, characterized in that: the driving behavior information includes integrating the features extracted by the mRMR algorithm with the features extracted by the principal component analysis.
CN202110724482.5A 2021-06-29 2021-06-29 Driving behavior feature extraction method based on mRMR algorithm and principal component analysis Active CN113642600B (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
CN202110724482.5A CN113642600B (en) 2021-06-29 2021-06-29 Driving behavior feature extraction method based on mRMR algorithm and principal component analysis

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
CN202110724482.5A CN113642600B (en) 2021-06-29 2021-06-29 Driving behavior feature extraction method based on mRMR algorithm and principal component analysis

Publications (2)

Publication Number Publication Date
CN113642600A true CN113642600A (en) 2021-11-12
CN113642600B CN113642600B (en) 2024-02-02

Family

ID=78416249

Family Applications (1)

Application Number Title Priority Date Filing Date
CN202110724482.5A Active CN113642600B (en) 2021-06-29 2021-06-29 Driving behavior feature extraction method based on mRMR algorithm and principal component analysis

Country Status (1)

Country Link
CN (1) CN113642600B (en)

Cited By (1)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN115018395A (en) * 2022-08-08 2022-09-06 南京佑友软件技术有限公司 Power quality analysis method

Citations (5)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US20020128751A1 (en) * 2001-01-21 2002-09-12 Johan Engstrom System and method for real-time recognition of driving patters
CN108280415A (en) * 2018-01-17 2018-07-13 武汉理工大学 Driving behavior recognition methods based on intelligent mobile terminal
CN110175557A (en) * 2019-05-24 2019-08-27 北京交通大学 Driver's driving behavior method for identifying and classifying and system based on driving characteristics group
CN110390272A (en) * 2019-06-30 2019-10-29 天津大学 A kind of EEG signal feature dimension reduction method based on weighted principal component analyzing
CN110633729A (en) * 2019-08-06 2019-12-31 清华大学苏州汽车研究院（相城） Driving risk hierarchical clustering method for intelligent networking vehicle group test

Patent Citations (5)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US20020128751A1 (en) * 2001-01-21 2002-09-12 Johan Engstrom System and method for real-time recognition of driving patters
CN108280415A (en) * 2018-01-17 2018-07-13 武汉理工大学 Driving behavior recognition methods based on intelligent mobile terminal
CN110175557A (en) * 2019-05-24 2019-08-27 北京交通大学 Driver's driving behavior method for identifying and classifying and system based on driving characteristics group
CN110390272A (en) * 2019-06-30 2019-10-29 天津大学 A kind of EEG signal feature dimension reduction method based on weighted principal component analyzing
CN110633729A (en) * 2019-08-06 2019-12-31 清华大学苏州汽车研究院（相城） Driving risk hierarchical clustering method for intelligent networking vehicle group test

Non-Patent Citations (1)

* Cited by examiner, † Cited by third party
Title
易灵芝;刘智磊;龙辛;: "基于互信息冗余性分析的神经网络风电功率预测", 湘潭大学自然科学学报, no. 02 *

Cited By (1)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN115018395A (en) * 2022-08-08 2022-09-06 南京佑友软件技术有限公司 Power quality analysis method

Also Published As

Publication number Publication date
CN113642600B (en) 2024-02-02

Similar Documents

Publication Publication Date Title
CN109033332B (en) Driving behavior analysis method, medium and system
CN109635830B (en) Screening method for effective data for estimating automobile quality
CN109050535B (en) Rapid terrain condition identification method based on vehicle attitude
CN110705774A (en) Vehicle energy consumption analysis prediction method and system
CN114860793A (en) Data processing method, data processing device, storage medium and electronic equipment
CN110414561A (en) A kind of construction method of the natural scene data set suitable for machine vision
CN112668630A (en) Lightweight image classification method, system and equipment based on model pruning
CN114240719A (en) Air quality missing data filling method and system based on multiple stepwise regression
CN113642600A (en) Driving behavior feature extraction method based on mRMR algorithm and principal component analysis
KR20090131014A (en) Method for evaluating technology and service and forming service-oriented technology roadmap on the basis of patent information
CN110188592B (en) Urine formed component cell image classification model construction method and classification method
CN109238740B (en) Comprehensive evaluation method for automobile sliding resistance based on whole automobile mass and volume
CN111710360B (en) Method, system, device and medium for predicting protein sequence
CN112232885A (en) Multi-mode information fusion-based warehouse rental price prediction method
Rajesh Effective morphological transformation and sub-pixel classification of clustered images
CN113610350B (en) Complex working condition fault diagnosis method, equipment, storage medium and device
CN111026935B (en) Cross-modal retrieval reordering method based on adaptive measurement fusion
CN114691911A (en) Cross-view angle geographic image retrieval method based on information bottleneck variation distillation
CN110059353B (en) Automatic modal parameter identification, simplification and practical method
CN114997458A (en) Vehicle insurance claim rate prediction method based on principal component analysis and linear regression
CN109559053B (en) Vegetation reachability measurement method based on path distance
Inglis et al. Some observations on, and developments in, the analysis of multivariate survey data
CN117076293B (en) Software performance visual evaluation method based on lean sample hierarchy credible clustering thermodynamic diagram
CN107368672A (en) A kind of diesel engine stage division and device
Wang et al. The application of principal component cluster analysis in environment classification for Chinese cities

Legal Events

Date Code Title Description
PB01 Publication
PB01 Publication
SE01 Entry into force of request for substantive examination
SE01 Entry into force of request for substantive examination
GR01 Patent grant
GR01 Patent grant