CN112766373A - Driving behavior analysis method based on Internet of vehicles - Google Patents
Driving behavior analysis method based on Internet of vehicles Download PDFInfo
- Publication number
- CN112766373A CN112766373A CN202110070391.4A CN202110070391A CN112766373A CN 112766373 A CN112766373 A CN 112766373A CN 202110070391 A CN202110070391 A CN 202110070391A CN 112766373 A CN112766373 A CN 112766373A
- Authority
- CN
- China
- Prior art keywords
- data
- vehicles
- internet
- driving
- driving behavior
- 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
Images
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/23—Clustering techniques
- G06F18/232—Non-hierarchical techniques
- G06F18/2321—Non-hierarchical techniques using statistics or function optimisation, e.g. modelling of probability density functions
- G06F18/23213—Non-hierarchical techniques using statistics or function optimisation, e.g. modelling of probability density functions with fixed number of clusters, e.g. K-means clustering
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/24—Classification techniques
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N3/00—Computing arrangements based on biological models
- G06N3/02—Neural networks
- G06N3/04—Architecture, e.g. interconnection topology
- G06N3/044—Recurrent networks, e.g. Hopfield networks
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N3/00—Computing arrangements based on biological models
- G06N3/02—Neural networks
- G06N3/08—Learning methods
- G06N3/084—Backpropagation, e.g. using gradient descent
-
- G—PHYSICS
- G07—CHECKING-DEVICES
- G07C—TIME OR ATTENDANCE REGISTERS; REGISTERING OR INDICATING THE WORKING OF MACHINES; GENERATING RANDOM NUMBERS; VOTING OR LOTTERY APPARATUS; ARRANGEMENTS, SYSTEMS OR APPARATUS FOR CHECKING NOT PROVIDED FOR ELSEWHERE
- G07C5/00—Registering or indicating the working of vehicles
- G07C5/08—Registering or indicating performance data other than driving, working, idle, or waiting time, with or without registering driving, working, idle or waiting time
- G07C5/0808—Diagnosing performance data
Abstract
The invention discloses a driving behavior analysis method based on the Internet of vehicles, which comprises the following steps: s1: acquiring original information of a driving record of a driver in the Internet of vehicles to obtain original database data; s2: cleaning the original information of the driving record by combining the Isolationforest algorithm and the SOM algorithm to obtain sample data; s3: extracting and storing the characteristic parameters of the sample data according to the data attributes and the characteristic classification; s4: performing K-Means clustering analysis on the characteristic parameters to obtain a clustering center; s5: adopting bp neural network classification to judge whether the driving behavior is in a safe range; according to the driving behavior analysis method based on the Internet of vehicles, the driving behaviors of drivers of the vehicles can be analyzed by excavating driving data such as vehicle speed, rapid acceleration and the like, the driving behaviors of the drivers are favorably normalized, and meanwhile, the intelligent management of the vehicles is optimized and promoted.
Description
Technical Field
The invention relates to the technical field of traffic safety, in particular to a driving behavior analysis method based on the Internet of vehicles.
Background
With the increasing of the quantity of vehicles kept in China, a series of problems are brought to traffic safety and society. The concern of car networking and management departments on driving safety is also increasing. And the car networking technology is an effective way to solve the problems. The car networking system is characterized in that advanced sensing technology, network technology, computing technology, control technology and intelligent technology are utilized to comprehensively sense roads and traffic, interaction of large-range and large-capacity data among a plurality of systems is achieved, traffic whole-course control is conducted on each car, traffic whole-time-space control is conducted on each road, and therefore network and application which are mainly traffic efficiency and traffic safety are provided.
The method has mass data in the environment of the Internet of vehicles, and thus has special significance and value for mining and analyzing the driving behaviors and characteristics of the vehicles. The driving behavior of a vehicle driver can be researched by excavating driving data such as vehicle speed, rapid acceleration and the like, the driving behavior of the driver can be normalized, and meanwhile, the intelligent management of the vehicle can be optimized and promoted. However, how to extract and utilize these driving data to regulate the driving behavior of the vehicle driver is still lacking.
Disclosure of Invention
The invention aims to provide a driving behavior analysis method based on the Internet of vehicles, which can analyze the driving behavior of a vehicle driver by mining driving data such as vehicle speed, rapid acceleration and the like, is beneficial to standardizing the driving behavior of the driver, and simultaneously optimizes and promotes intelligent management of the vehicle.
In order to achieve the purpose, the invention adopts the following technical scheme:
a driving behavior analysis method based on Internet of vehicles comprises the following steps:
s1: acquiring original information of a driving record of a driver in the Internet of vehicles to obtain original database data;
s2: cleaning the original information of the driving record by combining the Isolation Forest algorithm and the SOM algorithm to obtain sample data;
s3: extracting and storing the characteristic parameters of the sample data according to the data attributes and the characteristic classification;
s4: performing K-Means clustering analysis on the characteristic parameters to obtain a clustering center;
s5: and (4) classifying by adopting a bp neural network, and judging whether the driving behavior is in a safe range.
Preferably, the specific method of step S2 is:
s21: randomly selecting a sample point from the training data by adopting an Isolationforest algorithm as a subsample, and putting the sample point into a root node of a tree;
s22: randomly appointing a dimension attribute, and randomly generating a cutting point p in the current node data, wherein the cutting point p is generated between the maximum value and the minimum value of the appointed dimension in the current node data;
s23: generating a hyperplane by using a cutting point p, and dividing the data space of the current node into 2 subspaces: placing the data smaller than the cut point p in the specified dimension on the left child of the current node, and placing the data larger than or equal to the cut point p on the right child of the current node;
s24: recursion steps S22 and S23 in the child nodes, new child nodes are continuously constructed until only one piece of data which can not be cut any more in the child nodes or the child nodes reach the limited height, t pieces of iTrees are obtained, and the iForest training is finished;
s25: evaluating the test data by using the generated iForest, removing abnormal data to obtain primary screening data and storing the primary screening data in a database;
s26: performing secondary screening and abnormal data cleaning on the primary screened data by adopting an SOM algorithm, and outputting normal data and storing the normal data in a database if abnormal data do not exist; if the abnormal data still exists, the steps S21 to S26 are repeatedly executed until the abnormal data is completely eliminated.
Preferably, the specific method for evaluating test data by iForest in step S25 is as follows:
s251: traversing each iTree by each test data respectively;
s252: calculating the number of layers of each test data falling on each tree finally, and calculating the height average value of each test data in the forest respectively;
s253: and obtaining the average path length of each piece of test data, wherein the test data with the average path length lower than the threshold value are abnormal data, and eliminating the abnormal data.
Preferably, the normal data output in step S26 is further subjected to redundancy check processing, and the specific method of the redundancy check processing is as follows: the Python data cleaning uses a duplicate method to repeatedly judge normal data, returns a sequence with the same number of lines as the original data, if the data lines are not repeated, corresponds to false, otherwise corresponds to true; and returning to true to clean again by using any method and returning to true once only one true exists in the sequence.
Preferably, the normal data after the redundancy check processing is further subjected to data correction, where the data correction includes data missing supplement, data confusion correction, and data deduplication, and finally robust and feasible sample data is obtained.
Preferably, the characteristic parameters of step S3 include the proportion of fatigue driving, overspeed, rapid acceleration and rapid deceleration, and rapid lane change;
the fatigue driving is analyzed by time series data collected in a vehicle-mounted terminal, the in-transit time of a vehicle is divided into driving time and stopping time according to data characteristics, and the time is analyzed in a segmented interval mode, if a driver continuously drives for 4 hours, the driver does not stop and rest for 20 minutes or more, and therefore the fatigue driving frequency on the day is judged to be increased by 1 time;
the overspeed is limited according to the high speed of different vehicle types, and is processed according to the overspeed if the overspeed exceeds the 10 percent of the specified rule of the intersection;
the proportion of the rapid acceleration to the rapid deceleration is firstly calculated to obtain the times of the rapid acceleration and the rapid deceleration of each vehicle, and then the times are divided by the data volume of the vehicles to obtain the percentage;
the lane changing time of the quick lane changing vehicle is fixed on the whole, the minimum lane changing duration is 1.0s, the maximum lane changing duration is 16.5s, and the vehicle is shown to run in a stable state in the range.
Preferably, the K-Means clustering method in step S4 specifically includes:
s41: determining a k value, and obtaining k sets by the sample data through a clustering method;
s42: randomly selecting k data points from the sample data as a centroid;
s43: for each point in the sample data, calculating the distance between the point and each centroid, and dividing the point into a set to which the centroid belongs when the point is close to which centroid;
s44: after all sample data are grouped together, k sets are in total, the centroid of each set is recalculated, and the steps S41 to S43 are repeatedly executed;
s45: until the distance between the newly calculated centroid and the original centroid is less than the threshold, the clustering has reached the desired result and the algorithm terminates.
Preferably, the bp neural network in step S5 is composed of an input layer, a hidden layer and an output layer; the specific method of step S5 is:
s51: 5 label dimensions of rapid acceleration, rapid deceleration, overspeed, rapid lane change and fatigue driving are used as input, and corresponding feature vectors are input; the hidden layer consists of 10 neurons; the output layer corresponds to the classified result, and the classified result is the category corresponding to the driver;
s52: and normalizing a typical sample clustering result to be used as a training sample, distributing the sample data, setting the maximum training times and a target error so as to train the bp neural network to obtain a bp neural network classifier of the driving behavior of the driver, and judging whether the driving behavior is safe or not by using the bp neural network classifier.
Preferably, the training parameters of the bp neural network in step S52 are: sample data is distributed according to the ratio of 8:2, the maximum training frequency is set to be 5000, and the target error is 0.001.
After the technical scheme is adopted, the invention has the following beneficial effects:
the invention relates to a driving behavior analysis method based on the Internet of vehicles, which is based on Internet of vehicles data, can quickly and effectively screen and eliminate abnormal detection values by using an Isolationforest algorithm, and can be visually corrected and supplemented by combining an SOM algorithm, original data can be cleaned through cross matching of the two algorithms to accurately and effectively eliminate abnormal values, normal available sample data can be screened, driving behaviors of vehicle drivers can be analyzed by mining driving data such as fatigue driving, overspeed, urgent acceleration, urgent deceleration, urgent lane change and the like, the driving behavior of the drivers can be favorably normalized, and intelligent management of vehicles can be optimized and promoted.
Drawings
FIG. 1 is a block flow diagram of the present invention;
FIG. 2 is a block flow diagram illustrating a specific method of step S2 according to the present invention;
FIG. 3 is a block flow diagram illustrating a specific method for evaluating test data by iForest in step S25 according to the present invention;
fig. 4 is a flow chart of cleaning the original information of the driving record according to the present invention.
Detailed Description
In order to make the objects, technical solutions and advantages of the present invention more apparent, the present invention is described in further detail below with reference to the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are merely illustrative of the invention and are not intended to limit the invention.
Examples
Referring to fig. 1 to 4, a driving behavior analysis method based on internet of vehicles includes the following steps:
s1: acquiring original information of a driving record of a driver in the Internet of vehicles to obtain original database data;
s2: cleaning the original information of the driving record by combining the Isolation Forest algorithm and the SOM algorithm to obtain sample data;
the specific method of step S2 is:
s21: randomly selecting a sample point from the training data by adopting an Isolationforest algorithm as a subsample, and putting the sample point into a root node of a tree; isolationforest is an efficient anomaly detection algorithm, and if some data indexes reach leaf nodes quickly, namely the distance d from a leaf to a root is short, the leaf nodes are considered to be likely anomaly points;
s22: randomly appointing a dimension attribute, and randomly generating a cutting point p in the current node data, wherein the cutting point p is generated between the maximum value and the minimum value of the appointed dimension in the current node data;
s23: generating a hyperplane by using a cutting point p, and dividing the data space of the current node into 2 subspaces: placing the data smaller than the cut point p in the specified dimension on the left child of the current node, and placing the data larger than or equal to the cut point p on the right child of the current node;
s24: recursion steps S22 and S23 in the child nodes, new child nodes are continuously constructed until only one piece of data which can not be cut any more in the child nodes or the child nodes reach the limited height, t pieces of iTrees are obtained, and the iForest training is finished;
s25: evaluating the test data by using the generated iForest, removing abnormal data to obtain primary screening data and storing the primary screening data in a database;
the specific method for evaluating the test data by the iForest in the step S25 is as follows:
s251: traversing each iTree by each test data respectively;
s252: calculating the number of layers of each test data which finally fall on each tree, namely the number of layers is the height of the test data in the tree, and respectively calculating the average height value of each test data in the forest;
s253: obtaining the average path length of each piece of test data, wherein the test data with the average path length lower than a threshold value are abnormal data, and eliminating the abnormal data;
s26: performing secondary screening and abnormal data cleaning on the primary screened data by adopting an SOM algorithm, and outputting normal data and storing the normal data in a database if abnormal data do not exist; if the abnormal data still exists, repeating the steps S21 to S26 until the abnormal data is completely eliminated;
the normal data output in step S26 is further subjected to redundancy check processing, and the specific method of the redundancy check processing is as follows: the Python data cleaning uses a duplicate method to repeatedly judge normal data, returns a sequence with the same number of lines as the original data, if the data lines are not repeated, corresponds to false, otherwise corresponds to true; then using any method, returning to true for cleaning again as long as one true exists in the sequence;
the normal data after the redundancy check processing is further subjected to data correction, wherein the data correction comprises data missing supplement, data disorder correction and data repeated deletion, and finally robust and feasible sample data is obtained;
s3: extracting and storing the characteristic parameters of the sample data according to the data attributes and the characteristic classification;
the characteristic parameters of the step S3 comprise the proportion of fatigue driving, overspeed, rapid acceleration and rapid deceleration and rapid lane change;
the fatigue driving is analyzed by time series data collected in a vehicle-mounted terminal, the in-transit time of a vehicle is divided into driving time and stopping time according to data characteristics, and the time is analyzed in a segmented interval mode, if a driver continuously drives for 4 hours, the driver does not stop and rest for 20 minutes or more, and therefore the fatigue driving frequency on the day is judged to be increased by 1 time;
the overspeed is limited according to the high speed of different vehicle types, and is processed according to the overspeed if the overspeed exceeds the 10 percent of the specified rule of the intersection;
the proportion of the rapid acceleration to the rapid deceleration is firstly calculated to obtain the times of the rapid acceleration and the rapid deceleration of each vehicle, and then the times are divided by the data volume of the vehicles to obtain the percentage;
the lane changing time of the quick lane changing vehicle is fixed on the whole, the minimum lane changing duration is 1.0s, the maximum lane changing duration is 16.5s, and the vehicle is shown to run in a stable state in the range;
s4: performing K-Means clustering analysis on the characteristic parameters to obtain a clustering center;
the K-Means clustering method of the step S4 specifically comprises the following steps:
s41: determining a k value, and obtaining k sets by the sample data through a clustering method;
s42: randomly selecting k data points from the sample data as a centroid;
s43: for each point in the sample data, calculating the distance between the point and each centroid, and dividing the point into a set to which the centroid belongs when the point is close to which centroid;
s44: after all sample data are grouped together, k sets are in total, the centroid of each set is recalculated, and the steps S41 to S43 are repeatedly executed;
s45: until the distance between the newly calculated centroid and the original centroid is less than the threshold, the clustering has reached the desired result, and the algorithm terminates
S5: adopting bp neural network classification to judge whether the driving behavior is in a safe range;
the bp neural network in the step S5 consists of an input layer, a hidden layer and an output layer; the specific method of step S5 is:
s51: 5 label dimensions of rapid acceleration, rapid deceleration, overspeed, rapid lane change and fatigue driving are used as input, and corresponding feature vectors are input; the hidden layer consists of 10 neurons; the output layer corresponds to the classified result, and the classified result is the category corresponding to the driver;
s52: and (3) normalizing a typical sample clustering result to be used as a training sample, distributing the sample data according to 8:2, setting the maximum training frequency to be 5000, setting the target error to be 0.001, training the bp neural network to obtain a bp neural network classifier of the driving behavior of the driver, and judging whether the driving behavior is safe or not by using the bp neural network classifier.
The above description is only for the preferred embodiment of the present invention, but the scope of the present invention is not limited thereto, and any changes or substitutions that can be easily conceived by those skilled in the art within the technical scope of the present invention are included in the scope of the present invention. Therefore, the protection scope of the present invention shall be subject to the protection scope of the claims.
Claims (9)
1. A driving behavior analysis method based on the Internet of vehicles is characterized by comprising the following steps:
s1: acquiring original information of a driving record of a driver in the Internet of vehicles to obtain original database data;
s2: cleaning the original information of the driving record by combining the Isolationforest algorithm and the SOM algorithm to obtain sample data;
s3: extracting and storing the characteristic parameters of the sample data according to the data attributes and the characteristic classification;
s4: performing K-Means clustering analysis on the characteristic parameters to obtain a clustering center;
s5: and (4) classifying by adopting a bp neural network, and judging whether the driving behavior is in a safe range.
2. The driving behavior analysis method based on the internet of vehicles as claimed in claim 1, wherein the specific method of step S2 is:
s21: randomly selecting a sample point from the training data by adopting an Isolation Forest algorithm as a subsample, and putting the subsample into a root node of a tree;
s22: randomly appointing a dimension attribute, and randomly generating a cutting point p in the current node data, wherein the cutting point p is generated between the maximum value and the minimum value of the appointed dimension in the current node data;
s23: generating a hyperplane by using a cutting point p, and dividing the data space of the current node into 2 subspaces: placing the data smaller than the cut point p in the specified dimension on the left child of the current node, and placing the data larger than or equal to the cut point p on the right child of the current node;
s24: recursion steps S22 and S23 in the child nodes, new child nodes are continuously constructed until only one piece of data which can not be cut any more in the child nodes or the child nodes reach the limited height, t pieces of iTrees are obtained, and the iForest training is finished;
s25: evaluating the test data by using the generated iForest, removing abnormal data to obtain primary screening data and storing the primary screening data in a database;
s26: performing secondary screening and abnormal data cleaning on the primary screened data by adopting an SOM algorithm, and outputting normal data and storing the normal data in a database if abnormal data do not exist; if the abnormal data still exists, the steps S21 to S26 are repeatedly executed until the abnormal data is completely eliminated.
3. The driving behavior analysis method based on the internet of vehicles according to claim 2, wherein the specific method for the iForest to evaluate the test data in the step S25 is as follows:
s251: traversing each iTree by each test data respectively;
s252: calculating the number of layers of each test data falling on each tree finally, and calculating the height average value of each test data in the forest respectively;
s253: and obtaining the average path length of each training datum, wherein the test datum with the average path length lower than the threshold value is abnormal data, and rejecting the abnormal data.
4. The driving behavior analysis method based on the internet of vehicles as claimed in claim 2, wherein the normal data output in step S26 is further processed by redundancy check, and the specific method of the redundancy check processing is as follows: the Python data cleaning uses a duplicate method to repeatedly judge normal data, returns a sequence with the same number of lines as the original data, if the data lines are not repeated, corresponds to false, otherwise corresponds to true; and returning to true to clean again by using any method and returning to true once only one true exists in the sequence.
5. The driving behavior analysis method based on the Internet of vehicles as claimed in claim 4, wherein the normal data after the redundancy check processing is further subjected to data correction, and the data correction comprises data missing supplement, data confusion correction and data deduplication, so as to obtain robust and feasible sample data.
6. The internet-of-vehicles-based driving behavior analysis method according to claim 1, wherein the characteristic parameters of step S3 include fatigue driving, overspeed, proportion of rapid acceleration and rapid deceleration, lane change;
the fatigue driving is analyzed by time series data collected in a vehicle-mounted terminal, the in-transit time of a vehicle is divided into driving time and stopping time according to data characteristics, and the time is analyzed in a segmented interval mode, if a driver continuously drives for 4 hours, the driver does not stop and rest for 20 minutes or more, and therefore the fatigue driving frequency on the day is judged to be increased by 1 time;
the overspeed is limited according to the high speed of different vehicle types, and is processed according to the overspeed if the overspeed exceeds the 10 percent of the specified rule of the intersection;
the proportion of the rapid acceleration to the rapid deceleration is firstly calculated to obtain the times of the rapid acceleration and the rapid deceleration of each vehicle, and then the times are divided by the data volume of the vehicles to obtain the percentage;
the lane changing time of the quick lane changing vehicle is fixed on the whole, the minimum lane changing duration is 1.0s, the maximum lane changing duration is 16.5s, and the vehicle is shown to run in a stable state in the range.
7. The driving behavior analysis method based on the internet of vehicles as claimed in claim 1, wherein the K-Means clustering method of step S4 is specifically:
s41: determining a k value, and obtaining k sets by the sample data through a clustering method;
s42: randomly selecting k data points from the sample data as a centroid;
s43: for each point in the sample data, calculating the distance between the point and each centroid, and dividing the point into a set to which the centroid belongs when the point is close to which centroid;
s44: after all sample data are grouped together, k sets are in total, the centroid of each set is recalculated, and the steps S41 to S43 are repeatedly executed;
s45: until the distance between the newly calculated centroid and the original centroid is less than the threshold, the clustering has reached the desired result and the algorithm terminates.
8. The Internet of vehicles based driving behavior analysis method of claim 6, wherein the bp neural network in step S5 is composed of an input layer, a hidden layer and an output layer; the specific method of step S5 is:
s51: 5 label dimensions of rapid acceleration, rapid deceleration, overspeed, rapid lane change and fatigue driving are used as input, and corresponding feature vectors are input; the hidden layer consists of 10 neurons; the output layer corresponds to the classified result, and the classified result is the category corresponding to the driver;
s52: and normalizing a typical sample clustering result to be used as a training sample, distributing the sample data, setting the maximum training times and a target error so as to train the bp neural network to obtain a bp neural network classifier of the driving behavior of the driver, and judging whether the driving behavior is safe or not by using the bp neural network classifier.
9. The driving behavior analysis method based on the internet of vehicles as claimed in claim 8, wherein the training parameters of the bp neural network in step S52 are: sample data is distributed according to the ratio of 8:2, the maximum training frequency is set to be 5000, and the target error is 0.001.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202110070391.4A CN112766373A (en) | 2021-01-19 | 2021-01-19 | Driving behavior analysis method based on Internet of vehicles |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202110070391.4A CN112766373A (en) | 2021-01-19 | 2021-01-19 | Driving behavior analysis method based on Internet of vehicles |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN112766373A true CN112766373A (en) | 2021-05-07 |
Family
ID=75703215
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202110070391.4A Pending CN112766373A (en) | 2021-01-19 | 2021-01-19 | Driving behavior analysis method based on Internet of vehicles |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN112766373A (en) |
Cited By (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN113968234A (en) * | 2021-11-29 | 2022-01-25 | 蔡定海 | Vehicle auxiliary driving control method and device and vehicle-mounted controller |
| CN116092059A (en) * | 2022-11-30 | 2023-05-09 | 南京通力峰达软件科技有限公司 | Neural network-based vehicle networking user driving behavior recognition method and system |
| CN116665342A (en) * | 2023-08-01 | 2023-08-29 | 北京简精科技有限公司 | New energy automobile driving behavior analysis method, system and equipment |
Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20150317844A1 (en) * | 2014-05-02 | 2015-11-05 | Kookmin University Industry-Academic Cooperation Foundation | Method of processing and analysing vehicle driving big data and system thereof |
| CN107153916A (en) * | 2017-04-30 | 2017-09-12 | 安徽中科美络信息技术有限公司 | A kind of driving behavior evaluation method clustered based on FCM with BP neural network |
| CN110046665A (en) * | 2019-04-17 | 2019-07-23 | 成都信息工程大学 | Based on isolated two abnormal classification point detecting method of forest, information data processing terminal |
| CN111833172A (en) * | 2020-05-25 | 2020-10-27 | 百维金科(上海)信息科技有限公司 | Consumption credit fraud detection method and system based on isolated forest |
-
2021
- 2021-01-19 CN CN202110070391.4A patent/CN112766373A/en active Pending
Patent Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20150317844A1 (en) * | 2014-05-02 | 2015-11-05 | Kookmin University Industry-Academic Cooperation Foundation | Method of processing and analysing vehicle driving big data and system thereof |
| CN107153916A (en) * | 2017-04-30 | 2017-09-12 | 安徽中科美络信息技术有限公司 | A kind of driving behavior evaluation method clustered based on FCM with BP neural network |
| CN110046665A (en) * | 2019-04-17 | 2019-07-23 | 成都信息工程大学 | Based on isolated two abnormal classification point detecting method of forest, information data processing terminal |
| CN111833172A (en) * | 2020-05-25 | 2020-10-27 | 百维金科(上海)信息科技有限公司 | Consumption credit fraud detection method and system based on isolated forest |
Non-Patent Citations (1)
| Title |
|---|
| 郑恒杰、熊昕等: "基于车联网数据挖掘的驾驶员行为分析", 《信息通信》, no. 8, pages 1 - 3 * |
Cited By (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN113968234A (en) * | 2021-11-29 | 2022-01-25 | 蔡定海 | Vehicle auxiliary driving control method and device and vehicle-mounted controller |
| CN113968234B (en) * | 2021-11-29 | 2023-05-02 | 深圳市科莱德电子有限公司 | Vehicle auxiliary driving control method and device and vehicle-mounted controller |
| CN116092059A (en) * | 2022-11-30 | 2023-05-09 | 南京通力峰达软件科技有限公司 | Neural network-based vehicle networking user driving behavior recognition method and system |
| CN116092059B (en) * | 2022-11-30 | 2023-10-20 | 南京通力峰达软件科技有限公司 | Neural network-based vehicle networking user driving behavior recognition method and system |
| CN116665342A (en) * | 2023-08-01 | 2023-08-29 | 北京简精科技有限公司 | New energy automobile driving behavior analysis method, system and equipment |
| CN116665342B (en) * | 2023-08-01 | 2023-10-03 | 北京简精科技有限公司 | New energy automobile driving behavior analysis method, system and equipment |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN112766373A (en) | Driving behavior analysis method based on Internet of vehicles | |
| Labib et al. | Road accident analysis and prediction of accident severity by using machine learning in Bangladesh | |
| CN110588658B (en) | Method for detecting risk level of driver based on comprehensive model | |
| CN109410588B (en) | Traffic accident evolution analysis method based on traffic big data | |
| Elyassami et al. | Road crashes analysis and prediction using gradient boosted and random forest trees | |
| CN111222584A (en) | Lithium battery real-time evaluation method based on big data and deep neural network | |
| CN110562261B (en) | Method for detecting risk level of driver based on Markov model | |
| CN111556016A (en) | Network flow abnormal behavior identification method based on automatic encoder | |
| CN115372828A (en) | Battery cell consistency evaluation method based on charging segment data and unsupervised algorithm | |
| US20170236071A1 (en) | Alarm management system | |
| CN113435471A (en) | Deep feature clustering high-emission mobile source pollution identification method and system | |
| Ahmed et al. | A comparative study of machine learning algorithms to predict road accident severity | |
| CN112651666A (en) | Driver risk assessment method based on driving mode transfer characteristics | |
| Rajendar et al. | Sensor Data Based Anomaly Detection in Autonomous Vehicles using Modified Convolutional Neural Network. | |
| CN113468538A (en) | Vulnerability attack database construction method based on similarity measurement | |
| CN110705628B (en) | Method for detecting risk level of driver based on hidden Markov model | |
| CN115688003A (en) | Driver identification method and device, computer equipment and readable storage medium | |
| Kwiatkowski et al. | A fuzzy-logic based multi-dimensional analysis of traffic incident data | |
| Hu et al. | Detecting socially abnormal highway driving behaviors via recurrent graph attention networks | |
| Yokoyama et al. | Do drivers' behaviors reflect their past driving histories?-large scale examination of vehicle recorder data | |
| CN114580874A (en) | Multidimensional distributed data analysis system suitable for highway risk assessment | |
| Patil et al. | Analysis of road accidents using data mining techniques: a survey | |
| Xia et al. | Real-time traffic accident severity prediction using data mining technologies | |
| Chen et al. | Driving-data-driven platform of driving behavior spectrum for vehicle networks | |
| Bereczki et al. | A Novel Machine Learning Based Traffic Congestion Recognition System |
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 |