AU2020101738A4 - Automated real-time driving behavioural modelling analysis and reporting in denser traffic using data mining - Google Patents

Automated real-time driving behavioural modelling analysis and reporting in denser traffic using data mining Download PDF

Info

Publication number
AU2020101738A4
AU2020101738A4 AU2020101738A AU2020101738A AU2020101738A4 AU 2020101738 A4 AU2020101738 A4 AU 2020101738A4 AU 2020101738 A AU2020101738 A AU 2020101738A AU 2020101738 A AU2020101738 A AU 2020101738A AU 2020101738 A4 AU2020101738 A4 AU 2020101738A4
Authority
AU
Australia
Prior art keywords
data
traffic
data mining
driver
analysis
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.)
Ceased
Application number
AU2020101738A
Inventor
Brahmananda Reddy Annapu Reddy
Durga Prasad Kavadi
Purna Chand Kollapudi
Narayana M. V.
Chiranjeevi Manike
Subhash P.
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.)
Annapu Reddy Brahmananda Reddy Dr
Kavadi Durga Prasad Dr
Kollapudi Purna Chand Dr
M V Narayana Dr
Manike Chiranjeevi Dr
P Subhash Dr
Original Assignee
Annapu Reddy Brahmananda Reddy Dr
Kavadi Durga Prasad Dr
Kollapudi Purna Chand Dr
M V Narayana Dr
Manike Chiranjeevi Dr
P Subhash Dr
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 Annapu Reddy Brahmananda Reddy Dr, Kavadi Durga Prasad Dr, Kollapudi Purna Chand Dr, M V Narayana Dr, Manike Chiranjeevi Dr, P Subhash Dr filed Critical Annapu Reddy Brahmananda Reddy Dr
Priority to AU2020101738A priority Critical patent/AU2020101738A4/en
Application granted granted Critical
Publication of AU2020101738A4 publication Critical patent/AU2020101738A4/en
Ceased legal-status Critical Current
Anticipated expiration legal-status Critical

Links

Classifications

    • GPHYSICS
    • G06COMPUTING; CALCULATING OR COUNTING
    • G06NCOMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
    • G06N20/00Machine learning
    • G06N20/10Machine learning using kernel methods, e.g. support vector machines [SVM]
    • GPHYSICS
    • G06COMPUTING; CALCULATING OR COUNTING
    • G06NCOMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
    • G06N3/00Computing arrangements based on biological models
    • G06N3/02Neural networks

Landscapes

  • Engineering & Computer Science (AREA)
  • Theoretical Computer Science (AREA)
  • Software Systems (AREA)
  • Physics & Mathematics (AREA)
  • Mathematical Physics (AREA)
  • Artificial Intelligence (AREA)
  • Data Mining & Analysis (AREA)
  • Evolutionary Computation (AREA)
  • Computing Systems (AREA)
  • General Engineering & Computer Science (AREA)
  • General Physics & Mathematics (AREA)
  • Medical Informatics (AREA)
  • Computer Vision & Pattern Recognition (AREA)
  • Health & Medical Sciences (AREA)
  • Life Sciences & Earth Sciences (AREA)
  • Biomedical Technology (AREA)
  • Biophysics (AREA)
  • Computational Linguistics (AREA)
  • General Health & Medical Sciences (AREA)
  • Molecular Biology (AREA)
  • Traffic Control Systems (AREA)

Abstract

AUTOMATED REAL-TIME DRIVING BEHAVIOURAL MODELLING ANALYSIS AND REPORTING IN DENSER TRAFFIC USING DATA MINING Abstract For emergency management, traffic safety is an essential component, and to improve the safe transit, the driving risk prediction should be sufficient. In recent days, the roads and transportation capabilities are not evolved effectively according to the expanding number of vehicles and population increase. Road traffic accidents have become the most significant health issue throughout the world. The extension of the present roads has become insufficient. Traffic congestion has become the main issue throughout the entire globe. The issues present due to traffic congestion are noise, pollution, and an increase in traveling time. Traffic prediction had paid attention and became a vital issue in smart cities. The technologies had developed so far as to know the driver behavior analysis. This research addresses the real-time driver behavioral analysis and the denser traffic using data mining technologies. The data mining emulsions are considerably used to establish and forecast the factors amongst the motor vehicle, human, and environmental considerations. The data mining algorithms are used to analyze and predict the driving risk to improve the driver's behavior by analyzing driving behavior data. This research explores the technologies to overwhelm indirect and direct traffic problems on civilization and the world. The classifiers are used to predict whether the traffic rule is violated. The classifier techniques are Decision Tree (Random Forest), SVM, and Neural network is used to know the driver behavior, prediction, and analysis and for prediction of road traffic accidents. 11 P a g e AUTOMATED REAL-TIME DRIVING BEHAVIOURAL MODELLING ANALYSIS AND REPORTING IN DENSER TRAFFIC USING DATA MINING Drawings: Data Set IFeature Engineeririg FaueS co ---------------------------------------------------------------- Pre1icionandCasifiratiori USin Daando Forestae1u Figure 1: Framework ofproposedmethodology 1aPaagae

Description

AUTOMATED REAL-TIME DRIVING BEHAVIOURAL MODELLING ANALYSIS AND REPORTING IN DENSER TRAFFIC USING DATA MINING
Drawings:
Data Set
IFeature Engineeririg FaueS co
Forestae1u USinDaando
---------------------------------------------------------------- Pre1icionandCasifiratiori
Figure 1: Framework ofproposedmethodology
1aPaagae
AUTOMATED REAL-TIME DRIVING BEHAVIOURAL MODELLING ANALYSIS AND REPORTING IN DENSER TRAFFIC USING DATA MINING
Description
Field of Invention:
This invention is related to find the real-time driver behavioral analysis and reporting about the denser traffic using data mining technology. Traffic congestion and driver distractions have become the central issue throughout the entire globe. The robust analysis can be provided using the data mining techniques in traffic management, processing the large traffic data functions, and allows the drivers and systems to make the decision better.
Background of the invention:
Traffic congestion is referred to as the excessive vehicles on the roadways at a peak time and resulting in a slower speed. It has become a significant challenge for traffic management and transportation planning. The traffic congestion can be avoided by giving the prior information to the road users regarding the road status so that, it will be helpful to minimize the opportunity of the traffic congestion and to make the better decisions by the road users while traveling. The quantifiable measures for the traffic congestion are mentioned by estimation of traffic parameters like traffic density and the travel time. These parameters are difficult to measure from the field. Due to traffic congestion, the problem faced by the traveler is missing opportunities, waste of time, and frustration.
In evaluation, for the business enterprise, the problems are alternate possibilities, loss of worker productiveness, shipping put off, and growth in the fee. Reduction of visitors congestion can help the people for safe transits, decreased injuries, decreased gasoline intake, controlling of air pollutants, discount in ready time and clean running of motors in transportation routes, and presenting the statistics for the plan and analysis. Site visitors' congestion causes could be extraordinary along with huge purple-light delay, boundaries in the street like accidents, double parking, avenue paintings, street narrowing down, random car preventing. Many answers have been proposed through the town governments, municipalities, corporations, and the researchers to find the site visitors congestion difficulty. Some of the answers for warding off problems include self-sufficient car technology, tracking the pedestrian site visitors, adaptive site visitors signals, real-time traffic feedback, vehicle sharing, and car-to-infrastructure smart corridors. The maximum of the solutions' standards is records Analytics, wireless sensor networks, and the internet of things. Other particular answers are a) constructing the brand new roads, bridges, pass roads, flyovers, and tunnels b) creating jewelry and performing the road rehabilitation. A significant quantity of records will be generated by traffic, that's accrued from the numerous varieties of devices, smart cameras, and sensors. The collection of data is not the difficulty.
In assessment, the tremendous hard problem is storage, managing, processing, studying, and managing the massive quantity of site visitors facts, to be useful for future purposes. This approach makes a specialty of analyzing a large amount of site visitors' data to extract precise visitors' statistics. However, the site visitors'pace, quantity, the arrival rate of the vehicle, and ready time prevalent.
Motive force behavior is the maximum essential to play a decisive function in retaining safety and sustainable transport. Apart from protection precautions, driver conduct additionally influences gasoline consumption, and visitors drift, air pollution, public health, personal health, and psychology. Lajunen and Ozkan described that riding style influences individual riding conduct, i.e., how the driver chooses and comes to a decision to drive. From the macroscopic factor of view in traffic control and manage, motive force conduct is maximum essential, and it is related to the visitor's congestion stages. From the microscopic factor of view, the driving assistance system is critical to improving the person riding behavior and also raises the motive force's cognizance that impacts the manner the power. Driver's will range in the manner how they choose to boost up and decelerate. Driving profiles are regarded as competitive driving (harsh accelerating, braking, cornering, lane converting, and many others.), Distracted riding (consuming, consuming, chatting or speak at the smartphone) motive force loses the eye at the riding project, danger-Taking ( using in excess velocity, violating the traffic guidelines ), Eco-riding ( using in a gas-efficient way to decrease the pollutant), and secure using (low threat using behavior). Andrei Aksjonov et al. have proposed the unconventional technique to assess the driving force distraction and consciousness of the scenario during the performance of an undertaking using system mastering and fuzzy set concept. The Evaluated effects display that the proposed technique is used to understand, discover, and calculate the extent of the driver distraction in phrases of percent based totally on secure automobile dynamic performance. The chatting on the cell smartphone is tested as the secondary task.
The driving force distraction experiments have achieved the usage of the simulations and as compared with current strategies. Nanxiang et al. has constructed the statistical models inside the form of Gaussian combination fashions for tracking the motive force's attention. The GMM version is used to quantify and examine the deviations of the driving force behavior from the ordinary driving force sample. The Secondary project is defined as the working with the cellphone, radio, and navigation device. The statistics are accrued from the actual international usage of specific sensors, including the Controller place network, Video camera, and microphone arrays. The writer has proposed the regression version to generate the parameter to identify and describe the motive force's attention to the alarm system and enhance the riding revel. Afaf Bouhoute et al. evolved a way to research and system the auto-generated information. On this method, the probabilistic graphical fashions are blended with the system, getting to know the set of rules to enhance driving force behavior. Najah and Hatem defined the evaluation of in-car and smartphone sensing talents, and the communication and the applications, offerings of driver behavior modeling like cloud-based total services. The distinct forms of entering and additives, tiers are involved in the motive force behavioral modeling. Fugiglando et al. has proposed the real-time method to categorize and examine motive force conduct in the shape of various organizations the use of a CAN community. The proposed approach utilizes the unsupervised mastering approach. The statistics are amassed from numerous sources like fuel pedal positioning, brake pedal strain, steering of wheel attitude, steerage the wheel momentum, longitudinal and lateral acceleration. Bashar et al. has added the general processes used by the drivers and passenger's popularity from the database amassed from smartphones. The probabilistic method uses capabilities primarily based on telephone inertial measurements to perceive and examine a person's behavior. Jainjun.et .al has analyzed the have a look primarily based on the using motion capture system, which's essential for site visitors protection inside the motorway. The MCS is used to capture the riding motion and motion Builder to apprehend the riding movement. Meenakshi et al. have evolved the clustering technique for the site visitors' sign choice guide to discover day durations throughout visitors' congestion automatically.
Objects of the invention:
• This research addresses the solution to the driver's behavior analysis while driving and traffic congestion because traffic management is essential to avoid the driver's risk, accidents, denser traffic. • Data Mining results are substantially used to establish and forecast the issues among human, motor vehicle, and environmentally friendly factors. • Through the analysis of driving behavior data, the data mining algorithms are used to analyze and predict the driving risk to improve the driver's behavior. • The classifier techniques are Decision Tree (Random Forest), SVM, and Neural network is used to know the driver behavior, prediction, and analysis and also for prediction of road traffic accidents.
Summary of the invention:
There are different approaches to detect and evaluate driver behavior. The following areas are i) detecting and evaluating the driver distraction, ii) capturing the behavior of driver using the speech model iii) analyzing and predicting the driver behavior. In recent days, most of the drivers are getting distracted from the visual and cognitive distractions. In addition to that, the features are embedded into vehicles such as entertainment systems, portable devices like smartphones also leads to distractions. The most common cause of traffic congestion and accident occurs because of driver distraction, mainly among young drivers. The distracted or non-distracted decisions are provided in the previous method, and it requires some additional device like a camera. In the proposed concept, the practical tool can be used to evaluate in different ways and comparing analyses of the secondary tasks on vehicle safety. For example, Lane changing, intersection decision-making, modeling of router choice, and driver profiling are the application of detecting and predicting driver behavior. Another model to know the driver's distraction can be done by knowing the driver motion behavior that includes the driver's body movements to obtain the parameters of motion that will impact on the vehicle control and safe transit in traffic while driving on the road. Based on the motion parameters, the impact of human fatigue and alcohol can be analyzed during the driving performance based on different traffic conditions. Modeling driver behavior is very important to know the environmental factors, vehicle, and driver state, predicting driver intention, and these factors will help to improve the safety transportation, reduction in traffic congestion, and improving the driver experience. The different researchers from different backgrounds like transport engineering, car industries, and psychology are trying to analyze the driving behavior of humans in a real-time situation. The relation between driver actions, the performance of vehicles and the driving environment should be understood. Road crashes should be prevented and reinforce traffic safety and enhance the driver's comfort. The traffic parameters can be predicted and categorized into four approaches. The following approaches are i) estimation and prediction of real-time traffic flow, ii) estimation and prediction of travel time at real-time, and iv) estimate and prediction of real-time traffic density. The mechanism for the real-time traffic process prediction in the urban areas allows reducing the traveling trip time using the data mining algorithms to increase the accuracy, scalability, and the adaptability of the traffic applications. The mechanism will combine all the scalable techniques of data mining like decision trees, association rules, and neural networks. The traffic parameters and the existing data will be used as the input. The existing traffic data is used to predict the traffic flow in the short term using the artificial neural network. The input parameters used here are speed, density, time, volume. Two approaches are present to predict the traffic parameters i) Model driven, ii) Data-driven. The Model-driven approaches use the simulation for road network infrastructure. In model-driven approaches, the number of parameters and the structure is fixed, and it will not reflect the changes of the road network to maintain the accurate results. The Data driven will aim to examine and organize the road traffic data to analyze and interpret the traffic situation. Most of the traffic systems use ad-hoc sensors like loops and cameras, but these types of sensors are not allowed to monitor the road locally because their installation and maintenance are expensive. Social networks like Twitter and Facebook are considered as the social sensor, and it is widely used as the source for getting information related to traffic events like traffic congestion, traffic accidents. The primary measures of the traffic congestion on the roadways are that traffic density, throughput, fuel consumption, safety, travel time, etc. The Real-time traffic density is done based on the i) aerial photography using the loop detector, ii) the data-driven approach based on the regression, Artificial Neural Network, K-Nearest Neighbor, etc. and iii) the Image processing technique uses the K-Nearest Neighbors and Artificial Neural Network (ANN) as the machine learning technique to estimate time and traffic density. The phases to develop the transportation and control systems are collection, pre-processing, analysis, storage, Communication, Maintenance, and archiving. Traffic data are collected using various methods. To observe the road traffic visually, surveillance cameras are used in the area for recording or streaming the images or videos, and this method is known as image or video-based method. The captured images or videos will be transferred to the control rooms, and this will be used widely for managing the road traffic. This method is efficient and easy to maintain. But this method requires a lot of space to store the image and video content, network bandwidth, and computational complexity. Sensor-based methods like RFIDs, Ultrasonic sensors, photoelectric sensors, radar, lasers, and vehicle probe data are used. The Wi-Fi, GPRS, Bluetooth, and WiMAX are used by the vehicle to vehicle and vehicle to infrastructure communication. Two or more methods are used to combine, which is known as Hybrid based methods. The Raw data, which is collected from any of the above methods, will be subjected to the noise present in data, missing values, and inconsistent data because of the failure of the sensor, measurement errors, and data link errors. Due to this reason, the data manipulation is needed for filtering, cleansing, fusion, and reduction. Data cleaning is the method to remove the noise, malfunction detection, and recovering the missing data. Reduction in the dimensionality of data by using the manifold learning, non-negative matrix factorization, or kernel dimensional reduction to improve the performance of the driven tasks. The Redundant features are reduced from the original features using heterogeneous learning, which is known as sparsity analysis. The processing of various sources of data, which is required, is known as Data Fusion. Data analysis is used to provide information about the estimation of the total number of vehicles for a specified area. The error data element is identified, and the measurement of the different data-driven process is to be done to ensure the quality present in analyzed data. Advanced data processing techniques and cloud computing tools are used to analyze the broad set of traffic data to create effective traffic decisions. Apart from that, some of the learning tools are used to control the lane signals, traffic light, a vehicle message system, and traffic information. This approach will be based on machine learning, data mining, and the artificial intelligence algorithm. The fast growth of the traffic leads to the demands of storage technologies. Cloud storage is used to store the secure the broad set of traffic data for creating traffic decisions. Data communication uses and shares traffic data. Traffic data is used to monitor traffic systems. Traffic data communication helps the policymakers, planners, transportation department, researchers and identifies the way for the system to be active and efficient. Sharing of traffic data will help the researchers to improve the decisions that will be clear with the high quality. Maintenance of the data is the process of continuous improvement, and the system checks the correction and verification. Maintaining in higher-level ensures the proper functioning of all the required systems. Data archiving can be done to move and store the data, which is less commonly used out of the active systems. The database which is specialized and archived to optimize the performance and to achieve the cost-effective and it will be useful for future retrieval. For example, the transfer of smartphone sensors data streams about driving behavior is not a straightforward method. The data which is gathered from the smartphones will usually consist of some issues like noise because they will be affected by the roadway conditions and the smartphone position of the car. During the monitoring stage, different types of sensors will be used, such as accelerometer, GPS, gyroscope, etc. This is the primary step for the collection of data. After data collection from the sensors, the data cleaning process should be done for the removal of noise and other pre-processing techniques. The removal of noise data from smartphone data is the main challenge. Noise may arise from the data with other transportation, walking, passenger trips, infrastructure related to noise, etc. In data pre-processing, the data filtering with a low-pass filter, vehicle coordinates from a smartphone, improving the quality of collected data. The features should extract from the data collection. The pre-processing of data is done by following the feature extraction. Based on this, trips can be recognized from the different modes, detecting crashes, identifying whether the user is the passenger or driver can be done. The car trips can be detected based on the speed information which can be collected from the GPS and accelerometer sensor. Data mining techniques can be used to understand some essential driving events which are monitored through smartphone sensors. In contrast, the machine learning approach is used for aggressive behavior detection like accelerating, braking, cornering, and identifying the driver's distraction by recognizing mobile usage while driving. The driver's behavior is the key to understand for improving road safety and level of service. Various methods are present to monitor the driving behavior from statistical methods to machine learning techniques.
Detailed Description of the Invention:
For users, the driving risk prediction is a strategy, and it is needed to make some in-depth analysis of the driver's behavior. The driving behavior is considered as the database. The data which is gathered (smartphone sensor data, Different types of sensor data, camera, image, video, speech, motion parameters, environmental factors) will usually consist of some issues like noise because they will be affected by the roadway conditions and the position of the car. This is the primary step for the collection of data. After the data collection, the data cleaning process should be done for the removal of noise and other pre-processing techniques. The removal of noise data from the database is the main challenge. Noise may arise from the data with other transportation, walking, passenger trips, infrastructure related to noise, etc. The pre-processing of data is done by following the feature extraction. Based on this, trips can be recognized from the different modes, detecting crashes, identifying whether the user is the passenger or driver. For example, car trips can be detected based on the speed information which can be collected from the GPS and accelerometer sensor. Data Mining is the method of analyzing, predicting, and discovering the new data pattern and hidden pattern form the broad set of data that is stored in repositories like database and data warehouse. This mining method includes the statistical models, mathematical algorithms, and the machine learning concept. The robust analysis can be provided using the data mining technology in traffic management, processing the large traffic data functions, and allows the drivers and systems to make the decision better. Data mining is developed as the primary research for several decades, like a large volume of data is needed for mining to grow massive data storage. It is the knowledge discovery in a database that helps to analyze and categorize the data from different dimensions and also extracts the knowledge from the data. Data mining techniques can be used to understand some essential driving events which are monitored through the imaging or video method or sensing method or hybrid method or vehicle to vehicle and vehicle to infrastructure method. The driver's behavior is the key to understand for improving road safety and level of service. Various methods are present to monitor the driving behavior from statistical methods to machine learning techniques. The features will be applied to the classifier for vehicle records to check whether it violates traffic rules.
From Figure 1, the framework is detailed in the proposed work. The data is collected from multiple sources. A large amount of data will be collected from the driver who is in the process of driving the vehicle. To evaluate driver behavior, the data has to be integrated from the various sources. The raw data consists of two approaches i) instantaneous data ii) trajectory data. The time, instantaneous speed, distance traveled in the vehicle moving process are the records present in the instantaneous data records. The GPS information in the vehicle moving process, time is recorded in the trajectory data records. GPS data will let us know the vehicle's trajectory location in real time, and it is essential to calculate the turning data, which cannot be calculated by the availability of instantaneous data. Data pre-processing is a crucial stage to handle the data before using the data mining algorithms. This step includes cleaning, feature selection, transformation, and normalization. Data cleaning has to be done for trajectory data and instantaneous data so that the correct data can be maintained. In the data cleaning process, the consistency should be checked. Data cleaning and filling the wanted data will provide the stability of data. After the pre-processing of data, a statistical analysis can be done on the routed data. In this, the maximum, minimum, and average speed of the driver can be calculated. The attribute selection is the concept of feature selection or variable selection. It is the process of selecting the features for further use. The feature extraction is to be done, such as speed and night travel combination. Iftwo drivers drive in different miles for day and night, the driver risk will be different in terms. Separate statistical analysis is better to know the accurate driver's behavior. After feature extraction, the classifiers are to be built for prediction models. Data Mining algorithms have to be performed after the pre-processing steps. By using data mining algorithms that are to be performed on the dataset is to find the prediction of the traffic accident. The various task will be performed by the data mining like classification, predictions, clustering, and association rule mining. The classification techniques are used to classify the data into the predefined class label. The classification consists of a two-step process. The learning phase comes under the first step to build the model that will describe the predefined set of classes and to analyze the training data. The classification phase is considered as the second phase, where the accuracy is estimated using the test data. In case, if the accuracy is acceptable, the rules or models will use the new unlabeled data, and it will be used in the decision-making. For data classification, the different techniques are present like neural networks, nave Bayes classifiers, SVM, and decision trees. The classifiers like support vector machine (SVM), Random Forest (RF), and Neural Networks (NN) are used. The kernel function is used by SVM to map the attributes to the higher dimensional space. It uses the hyperplane to maximize the distinction between the drivers to know whether the traffic rules violate. RF method integrates the multiple decision trees to classify the driving behaviors. The neural network can classify the driving behaviors by mapping the features in a non-linear way.
A. Decision Tree: The most popular classifier is the decision tree, and it is used as the classification technique. The tree is constructed based on the given data set and also uses the attributes selection like ratio measure. The decision tree is also one of the data mining technique, which is used to build the classification model, and it is a practical method. This method is speedy and does not requires any of the parameter settings, deals with the multidimensional data, and generates simple classification rules which will be understandable by humans. Decision trees have good accuracy. B. Neural Networks: Artificial neural network is the data modeling tool for prediction and classification. There are different kinds of ANN techniques. Among them, back propagation is the most widely used network. The advantage of ANN is that flexibility while dealing with the noisy and missing data. Neural networks are suitable for continuous valued and also speeds up the computation procedure. It can extract rules from the trained neural networks, which makes it useful for classification and prediction purposes in data mining.
C. Support Vector Machine: A Support Vector Machine is also a classification and prediction algorithm for both linear and non-linear data. During the training time, SVM will be slow and highly accurate. The performance will be perfect on the data sets, which will have the number of features, and it is easy to train.

Claims (7)

AUTOMATED REAL-TIME DRIVING BEHAVIOURAL MODELLING ANALYSIS AND REPORTING IN DENSER TRAFFIC USING DATA MINING Claims We claim,
1. With the help of data mining techniques, the driver's behavior and traffic congestion can be predicted and analyzed.
2. Data Mining results are substantially used to define and foresee the factors amongst human, motor vehicle, and conservation issues.
3. Through the analysis of driving behavior data, the data mining algorithms are used to analyze and predict the driving risk to improve the driver's behavior.
4. The classifier techniques are Decision Tree (Random Forest), SVM, and Neural network is used to know the driver behavior, prediction, and analysis and also for prediction of road traffic accidents.
5. The kernel function is used by SVM to map the attributes to the better dimensional space. It uses the hyperplane to increase the difference among the drivers to know whether the traffic rules violate. RF method integrates the multiple decision trees to classify the driving behaviors. The neural network can classify the driving behaviors by mapping the features in a non-linear way.
6. Data pre-processing is an important stage to handle the data before using the data mining algorithms. This step includes cleaning, feature selection, transformation, and normalization.
7. Data cleaning will remove the noise from the data and makes it to be consistent.
1 Pag e
AUTOMATED REAL-TIME DRIVING BEHAVIOURAL MODELLING ANALYSIS Aug 2020
AND REPORTING IN DENSER TRAFFIC USING DATA MINING
Drawings: 2020101738
Figure 1: Framework of proposed methodology
1|Page
AU2020101738A 2020-08-10 2020-08-10 Automated real-time driving behavioural modelling analysis and reporting in denser traffic using data mining Ceased AU2020101738A4 (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
AU2020101738A AU2020101738A4 (en) 2020-08-10 2020-08-10 Automated real-time driving behavioural modelling analysis and reporting in denser traffic using data mining

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
AU2020101738A AU2020101738A4 (en) 2020-08-10 2020-08-10 Automated real-time driving behavioural modelling analysis and reporting in denser traffic using data mining

Publications (1)

Publication Number Publication Date
AU2020101738A4 true AU2020101738A4 (en) 2020-09-17

Family

ID=72432503

Family Applications (1)

Application Number Title Priority Date Filing Date
AU2020101738A Ceased AU2020101738A4 (en) 2020-08-10 2020-08-10 Automated real-time driving behavioural modelling analysis and reporting in denser traffic using data mining

Country Status (1)

Country Link
AU (1) AU2020101738A4 (en)

Cited By (8)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN110491121A (en) * 2019-07-26 2019-11-22 同济大学 A kind of heterogeneity traffic accident causation analysis method and apparatus
CN113076661A (en) * 2021-04-28 2021-07-06 中国人民解放军国防科技大学 Uncertain data driven radar early warning detection modeling method
CN113190997A (en) * 2021-04-29 2021-07-30 贵州数据宝网络科技有限公司 Big data terminal data restoration method and system
CN113674122A (en) * 2021-07-22 2021-11-19 华设设计集团股份有限公司 Urban resident travel rule rapid extraction method suitable for high-concurrency travel data
CN114613130A (en) * 2022-02-18 2022-06-10 北京理工大学 Driving credibility analysis method in traffic and delivery system
CN114637931A (en) * 2022-03-29 2022-06-17 北京工业大学 Travel mode detection method based on manifold upper sequence subspace clustering
CN116227714A (en) * 2023-03-14 2023-06-06 西华大学 Travel mode selection prediction and analysis method and system
CN117173899A (en) * 2023-11-03 2023-12-05 武汉网信安全技术股份有限公司 Smart city data processing method

Cited By (13)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN110491121B (en) * 2019-07-26 2022-04-05 同济大学 Heterogeneous traffic accident cause analysis method and equipment
CN110491121A (en) * 2019-07-26 2019-11-22 同济大学 A kind of heterogeneity traffic accident causation analysis method and apparatus
CN113076661B (en) * 2021-04-28 2022-08-12 中国人民解放军国防科技大学 Uncertain data driven radar early warning detection modeling method
CN113076661A (en) * 2021-04-28 2021-07-06 中国人民解放军国防科技大学 Uncertain data driven radar early warning detection modeling method
CN113190997A (en) * 2021-04-29 2021-07-30 贵州数据宝网络科技有限公司 Big data terminal data restoration method and system
CN113674122A (en) * 2021-07-22 2021-11-19 华设设计集团股份有限公司 Urban resident travel rule rapid extraction method suitable for high-concurrency travel data
CN114613130A (en) * 2022-02-18 2022-06-10 北京理工大学 Driving credibility analysis method in traffic and delivery system
CN114637931A (en) * 2022-03-29 2022-06-17 北京工业大学 Travel mode detection method based on manifold upper sequence subspace clustering
CN114637931B (en) * 2022-03-29 2024-04-02 北京工业大学 Travel mode detection method based on manifold sequence subspace clustering
CN116227714A (en) * 2023-03-14 2023-06-06 西华大学 Travel mode selection prediction and analysis method and system
CN116227714B (en) * 2023-03-14 2023-10-27 西华大学 Travel mode selection prediction and analysis method and system
CN117173899A (en) * 2023-11-03 2023-12-05 武汉网信安全技术股份有限公司 Smart city data processing method
CN117173899B (en) * 2023-11-03 2024-01-30 武汉网信安全技术股份有限公司 Smart city traffic travel quality evaluation method

Similar Documents

Publication Publication Date Title
AU2020101738A4 (en) Automated real-time driving behavioural modelling analysis and reporting in denser traffic using data mining
Alsrehin et al. Intelligent transportation and control systems using data mining and machine learning techniques: A comprehensive study
CN111583639B (en) Road traffic jam early warning method and system
Ma et al. Driving style recognition and comparisons among driving tasks based on driver behavior in the online car-hailing industry
Matarazzo et al. Crowdsourcing bridge dynamic monitoring with smartphone vehicle trips
US11467579B2 (en) Probabilistic neural network for predicting hidden context of traffic entities for autonomous vehicles
Bolovinou et al. Driving style recognition for co-operative driving: A survey
Li et al. Modeling vehicle merging position selection behaviors based on a finite mixture of linear regression models
EP4021027A1 (en) Distributed learning to learn context-specific driving patterns
Singh et al. Road pothole detection from smartphone sensor data using improved LSTM
Prezioso et al. Machine Learning Insights for Behavioral Data Analysis Supporting the Autonomous Vehicles Scenario
Jadhav et al. Road accident analysis and prediction of accident severity using machine learning
Paredes et al. Intelligent collision risk detection in medium-sized cities of developing countries, using naturalistic driving: A review
Scora et al. Real-time roadway emissions estimation using visual traffic measurements
Krishna et al. A Computational Data Science Based Detection of Road Traffic Anomalies
Ranacher et al. A model to estimate and interpret the energy-efficiency of movement patterns in urban road traffic
Huang et al. A data-driven operational integrated driving behavioral model on highways
Raslan et al. IoT for measuring road network quality index
CN114492544B (en) Model training method and device and traffic incident occurrence probability evaluation method and device
Soultana et al. Context-awareness in the smart car: Study and analysis
He et al. Feasibility of economic forecasting model based on intelligent algorithm of smart city
Ali et al. Employment of instrumented vehicles to identify real-time snowy weather conditions on freeways using supervised machine learning techniques–A naturalistic driving study
Manikandan et al. Traffic Flow Prediction For Intelligent Transportation System Using Machine Learning
Talebloo et al. Dynamic and Systematic Survey of Deep Learning Approaches for Driving Behavior Analysis
Guo et al. Freeway Merging Decision Analysis on the Acceleration Lane

Legal Events

Date Code Title Description
FGI Letters patent sealed or granted (innovation patent)
MK22 Patent ceased section 143a(d), or expired - non payment of renewal fee or expiry