Bayesian Network-Based Road Traffic Accident Causality Analysis

Size: px
Start display at page:

Download "Bayesian Network-Based Road Traffic Accident Causality Analysis"

Transcription

1 2010 WASE International Conference on Information Engineering Bayesian Networ-Based Road Traffic Accident Causality Analysis Xu Hongguo Zong Fang (Contact Author) Abstract Traffic accident causality analysis is an important aspect in the traffic safety research field. Based on data survey and statistical analysis, a Bayesian networ for traffic accident causality analysis was developed. The structure and parameter of the Bayesian networ was learnt with K2 algorithm and Bayesian parameter estimation respectively. With the Junction Tree algorithm, the effect of road cross-section on the accident casualties was inferred. The results show that the Bayesian networ can express the complicated relationship between the traffic accident and the causes, as well the correlations among the factors of causes. The results of analysis provide the valuable information on how to reveal the traffic accident causality mechanisms and how to tae effective measures to improve the traffic safety situations. Keywords-traffic accident; bayesian networ; K2 algorithm; accident causality Ⅰ. INTRODUCTION With the development of economy and society in China, the car ownership and the road traffic accidents have increased greatly. In order to reduce the traffic accidents, it is necessary to analyze traffic accident causality in the theory. The activity can reveal the correlation between the traffic accident and the elements of traffic system, and provide the valuable information for road construction, traffic safety control, vehicle safety design, and so on. The causes of traffic accident include the subjective factors and objective one. The subjective one is human factors, composed of motor driver, non-motor driver, Zhang Huiyong jluzhy@yahoo.com.cn zongfang@jlu.edu.cn passenger and pedestrian. The objective one consists of vehicle, road environment and environment: The vehicle factors, for example vehicle performance; The road elements include types, alignment, intersection, pavement conditions, etc.; The environment factors consist of natural conditions, traffic volume, etc. Road traffic system is a complex dynamic coupling one, composed of human, vehicle, road and environment. Road accident is caused by disturbance of these elements. As a result, it has the characteristic of randomness and uncertainty. This paper will give the theory of traffic accident causality with Bayesian networ. It is organized as follows: In section 2, the previous researches are reviewed. In section 3, the Bayesian model is introduced. This is followed by developing a Bayesian networ for traffic accident causality analysis in section 4. In section 5, the correlation between the traffic accident and the causes is analyzed. Summary and conclusions are made in section 6. Ⅱ. REVIEW A. Traffic Accident Causality Theory The traditional accident causality theories include Accident Proneness theory, Accident Causality Sequence theory, Energy Transfer theory, Surry's Accident model, Orbit Intersecting theory, Haddon model, etc. Based on these, many researchers have given the traffic accident causality theory in recent years. The main methods include Principle Component Analysis, Layer-Interrelating Analysis method, Grey Forecasting /10 $ IEEE DOI /ICIE

2 Model, BP Neural Networ, Fuzzy Clustering Analysis and so on. For instance, Chong Miao and Abraham Ajith (2004, 2005) studied the methods of traffic accident causality and traffic accident data mining with machine learning paradigm [1,2]. Zhang Haiyi et al. (2001) designed a traffic accident analysis system based on association rules [3]. Han Wenbin (2007) studied on the relationship between road alignment and traffic accidents with Principle Component analysis and Multivariate Linear Regression method [4]. Sha aimin (2006) analyzed the accident causes by the accident data,and developed a Grey forecasting model predicting traffic accident [5]. Yuan Chunmiao et al. (2005) analyzed the causes of accidents based on BP Neural Networ [6]. Wei Qingyao et al. (2005) analyzed road factors in traffic accident using Layer-Interrelating Analysis method [7]. Yu hongqi (2005) analyzed traffic accident causality factors using the Clustering Method in data mining and the technique of the Open Database Connectivity (ODBC) [8]. B. Bayesian Method In terms of traffic safety, the Bayesian method is mainly applied to traffic accident duration prediction. Qin Xiaohu (2005) developed a traffic accident prediction model based on Bayesian networ [9]. Wang Fazhi(2006) created a Bayesian networ assessing traffic accidental event situation and analyzed the effect of several factors, such as whether, traffic volume, vehicle type, etc., on the traffic accident duration [10]. Ji Yang Beibei (2008) studied traffic accident duration prediction method with Bayesian method-based decision tree classification algorithm [11]. Above reviews show that most of the researches are unitary analysis. This can only reveal the inherent laws of traffic accident on a certain aspect. However, the traffic accident causality is multidimensional and there are correlation and logic relationships among the causality factors. The existing wors are practice-oriented, but without the character of generality. Although some studied the traffic accident causality based on the theory of complex systems, the whole road traffic system are not taen as complex systems. In addition, most of the existing wors are static analysis that ignore the time-variation of the factors. The application of Bayesian networ in the accident duration forecasting indicates that: (1) Comparing with the other methods, Bayesian networ can describe the logic relationships among the variables in the networ. (2) The networ can embody the dynamic coupling mechanism among the variables; (3) Bayesian method provides a way to introduce more information into the model, such as expert suggestions, other studies and empirical distributions. The above merits show that Bayesian networ can solve the problems in the existing researches. Therefore, a Bayesian networ for traffic accident causality analysis will be developed in this study, in order to explore the application of Bayesian method in the field of traffic accident reasons. Ⅲ. METHODOLOTY A. Bayesian Networ A Bayesian networ, belief networ or directed acyclic graphical model is a probabilistic graphical model that represents a set of random variables and their conditional dependencies via a directed acyclic graph (DAG). A Bayesian networ consists of the following elements: (1)A set of variables and a set of directed lins among variables; (2)The variables coupled with the directed lins construct a DAG; (3) Each variable with parents has a conditional probability table. The basis of Bayesian networ is Bayes' formula for conditional probability: H) D H ) H D) = (1) D) H is a hypothesis, and D is the data. H) is the prior probability of H: the probability that H is correct before the data D was seen. D H) is the conditional probability of seeing the data D given that the hypothesis H is true. D H) is called the lielihood. D) is the marginal probability of D. H D) is the posterior probability: the probability that the hypothesis is true, given the data and the previous state of belief about the hypothesis. B. Modeling of Bayesian Networ The modeling of Bayesian networ consists of two steps: structure learning and parameter learning. 414

3 The purpose of structure learning is to define the directed lins between the nodes in order to form the structure of the Bayesian networ. The main methods of structure learning are Exhaustive search, K2 algorithm, Hill-climbing, etc. The most popular method is K2 algorithm (Cooper and Hersovits, 1992), which is a ind of greedy search algorithm that wors as follows: Initially each node has no parents. It then adds incrementally the parent whose addition most increases the score of the resulting structure. When the addition of no single parent can increase the score, it stops adding parents to the node [12]. The aim of parameter learning is to estimate the posterior distribution of the nodes when the structure of the Bayesian networ and the prior distribution of some nodes are already nown. In many previous researches, the conditional probability tables were defined by the experts. However, there is a big deviation between the estimated results and the actual one. The popular method is to learn the distribution of the variables from the data, which maes the parameter learning more universal. When learning parameter, the prior distribution of the nodes can be defined by experts or by analyzing and estimating. The main methods of parameter learning are maximum lielihood parameter estimation and Bayesian parameter estimation. Ⅳ. CONSTRUCTION STRUCTION OF BAYESIAN NETWORK A. Data Based on traffic accidents data of some main expressways (including Chang-Ping, Chang-Ji, Yan-Tu, Jing-Ha, etc.) in Jilin province during 2003~2006, the training data, consisting of 3019 sample data, are set up. The main data items (shown in TableⅠ) indicate that the data about road condition and the consequence of the accident are in detail. Data type Environment factors Road factors Traffic accident TABLEⅠ. DATA ITEMS OF THE TRAINING DATA Data items Terrain, Weather, Traffic control Road type, Pavement type, Road alignment, Type of intersection and road section, Road cross-section, Road condition Cause of accident, Accident form, Accident type, deaths, serious, light, Property damage No data about driver, passenger and pedestrian, vehicle types, speed, traffic volume, etc. are given, which maes the modeling difficult. B. Structure Learning Using the K2 algorithm and the Full-BNT toolbox of Matlab, the structure of the Bayesian networ is learnt with the training data, as shown in Figure 1. Road cross-section Cause of accident Figure 1. The structure of the Bayesian networ The Bayesian networ is composed of ten nodes and concerned lins. The ten nodes come from the ten variables in Table Ⅱ, and the lins represent the correlations between the nodes. For instance, as shown in the networ, road type has influence on accident form and accident type; accident type affects the number of deaths. TABLE Ⅱ. VARIABLES IN THE BAYESIAN NETWORK Terrain 1.Plains, 2.Hills, 3.Mountains 1.Expressway, 2.Class Ⅰ highway, 3. Class Ⅱ highway, Road type 4.Class Ⅲ highway, 5. Class Ⅳ highway, 6. Sub-standard road, 7.Urban road Road 1.Road divided by lanes and directions, 2.Road divided by cross-section lanes, 3.Road divided by directions, 4.Lane-direction mixed 1.Vehicle breadown, 2.Violation of motor vehicle, Cause of 3.Violation of non-motor vehicle, passenger or pedestrian, accident 4.Unexpected reasons, 5.Others 1. Front collision, 2. Side collision, 3. Rear collision, 4. Scraping in the opposite direction, 5. Scraping in the same Accident form direction, 6. Rolling over, 7. Rolling, 8. Hitting a stationary object, 9. Hitting a stationary vehicle, 10. Falling, 11. Catching fire, 12. Others Accident type 1. Injury, 2. Property damage, 3. Fatal than 4 deaths serious than than 3 light Property damage (Yuan) Number of deaths Terrain Road type Accident form Accident type serious light Environment factors Road factors Traffic accident Accident casualties Property damage No less than

4 C. Parameter Learning The variables in the Bayesian networ are estimated by the Bayesian parameter estimation method and the Full-BNT toolbox in Matlab. During the process of parameter learning, the prior distributions of all the variables are assumed to be Dirichlet distribution (see reference [12] for the detailed content about the Dirichlet distribution). The variable of accident type will be taen as an example to illustrate the parameter learning process. TABLE Ⅲ. ESTIMATIONS OF THE ACCIDENT TYPE Road type Conditional probability Conditional probability of Training data of injury accident property damage accident Conditional probability Training data Training data Sample size of fatal accident Expressway Class Ⅰ highway Class Ⅱ highway Class Ⅲ highway Class Ⅳ highway Sub-standard road Urban road D. Validity Test The validity of the estimations will be tested by two methods: (1) Comparing the forecasts with the training data; (2) Calculating the Hit Ratio of the model. The comparing of the forecasts and the training data are shown in Table Ⅲ. Results show that the maximum absolute error of the forecasts compared with the training data is , and the mean absolute error is Because of the sample size of the property damage accidents on sub-standard road is 0, the error in this case is very large, and the relative error is 1. If the data is not taen into account, the maximum relative error of the model is , and the mean relative error is The parameter of Hit Ratio is calculated as: Defining P i as the forecasting probability of the accident type i for the piece of data. Defining d = i. If Pi is the max for all the values of i, defining if and only if P i traverses all i. Then S The Hit ratio is: 1, = d = 0, d = i, (2) R h n Si = = 1 (3) n The reference [15] shows that if the Hit Ratio is greater than 80%, the estimation can be considered to be credible. Based on formula (3), the Hit Ratio of the model in accident type forecast is calculated as 100%. The estimations of other variables in the model are tested with the same two methods. The results show high accuracy of the model. Ⅴ. APPLICATION IN THE ACCIDENT CASUALTIES ANALYSIS Based on the Bayesian networ, the conditional probability of any node in the networ can be inferred. The main methods of inference include Junction Tree algorithm, Variable Elimination algorithm, Global Inference methods and so on. The Junction Tree algorithm, which is the basis of all the inference methods, is most widely used. Therefore, the Junction Tree algorithm will be selected. The inferred results of the accident casualties under the influence of the road cross-section are shown in Table Ⅳ, which is taen as an example to illustrate the inferring process. 416

5 Road crosssection TABLE Ⅳ. THE CONDITIONAL PROBABILITY OF ACCIDENT CASUALTIES UNDER THE INFLUENCE OF ROAD CROSS-SECTION deaths than 4 serious than 3 light than 3 lanes and directions lanes directions Lane-directio n mixed Assuming the number of deaths, serious and light are all 0, the effect of the road cross-section on the accident casualties is shown in Figure 2. The results show that the better the condition of road cross-section, the greater the lielihood of no accident casualty lanes and directions no death no serious injury no light injury lanes directions Lane-direction mixed Figure 2. The effect of the road cross-section on the accident casualties With the same method, the analysis of the other sets of data in Table Ⅳ indicates that the road cross-section is an important factor for accident casualties. When the road cross-section is divided by lanes and directions, the casualties are reduced to a great extent. The casualties increase gradually with the worsening of the road cross-section. It means that the better the condition of road facilities, the lower the casualties. The influence of other factors on the traffic accident can be inferred with the same method. Ⅵ. CONCLUSIONS Based on data survey and statistical analysis, a Bayesian networ for traffic accident causality analysis was given. The structure and parameter of the Bayesian networ is learnt with K2 algorithm and Bayesian parameter estimation method respectively. With the Junction Tree algorithm, the effect of the factor of road cross-section on the accident casualties is inferred. It is an attempt to explore new methods for traffic accident causality analysis. The results of analysis provide the valuable information on how to reveal the traffic accident causality mechanism and how to tae effective measures to improve the traffic safety conditions. The limitations include: (1) Due to the lac of some data, the factors of human, vehicle, speed and traffic volume, etc. are not introduced in the Bayesian networ. This problem results in the incompletion of the Bayesian networ. The required data should be complemented to improve the model in the future. (2) It can be improved in prior distribution definition and Bayesian networ application. ACKNOWLEDGEMENTS The research is funded by National High Technology Research and Development Program (2009AA11Z201). REFERENCES [1] C. Miao, A. Ajith, Traffic accident analysis using machine learning paradigms, Computational Intelligence in Data Mining, 2005, vol. 29, No. 5, pp [2] C. Miao, A. Ajith, Traffic Accident Data Mining Using Machine Learning Paradigms, Fourth International Conference on Intelligent Systems Design and Applications (ISDA'04), Hungary, 2004, pp [3] H. Zhang, B. Bac, W. Zhou, The Design and Implementation of a Traffic Accident Analysis System, Proceedings of the 14th International conference on Industrial and engineering applications of artificial intelligence and expert systems: engineering of intelligent systems, 2001, pp [4] H. Wenbin, Research on the Relation between Road Alignment and Traffic Accident, Master Degree Dissertation of, [5] S. Aimin, Research on the Freeway Traffic Accidents and Countermeasures, Master Degree Dissertation of Dongnan University, [6] Y. Chunmiao, C. Baozhi, L. Chang, Analyzing Methods of Contributing Causes for Accidents Based on BP Neural Networ, Industrial Safety and Dust Control, 2005, vol. 31, No. 10, pp [7] W. Qing-yao, C. Bin, J. Weidong, F. Rui, Y. Wei, Analyzing Road Factor in Traffic Accident Based on Layer-interrelating Analysis Method, Journal of Changsha Communications University, 2005, vol. 21, No. 1, pp [8] Yu hongqi, Analysis of the Road Traffic Accident Reason in Clustering, Master Degree Dissertation of, [9] Qin Xiaohu, The Model and Application of Urban Traffic Emergency Management and Safety System, Doctor Degree Dissertation of Chongqing University, [10] W. Fazhi, Methods Based on the Bayesian Networs for Traffic Accidental Event Situation Assessment, Master Degree Dissertation of Dalian University of Technology, [11] J. Beibei, Research on Prediction Method of Traffic Incident Duration, Doctor Degree Dissertation of Tongji University, [12] G.Cooper, E. Hersovits, A Bayesian Method for the Induction of Probabilistic Networs from Data, Machine Learning, 1992, No. 9, pp

Global Journal of Engineering Science and Research Management

Global Journal of Engineering Science and Research Management PREDICTIVE COMPLEX EVENT PROCESSING USING EVOLVING BAYESIAN NETWORK HUI GAO, YONGHENG WANG* * College of Computer Science and Electronic Engineering, Hunan University, Changsha 410082, China DOI: 10.5281/zenodo.1204185

More information

Estimating Traffic Accidents in Turkey Using Differential Evolution Algorithm

Estimating Traffic Accidents in Turkey Using Differential Evolution Algorithm SSP - JOURNAL OF CIVIL ENGINEERING Vol. 12, Issue 1, 2017 DOI: 10.1515/sspjce-2017-0008 Estimating Traffic Accidents in Turkey Using Differential Evolution Algorithm Ali Payıdar Akgüngör, Ersin Korkmaz

More information

Research Article A Bayesian Network Approach to Causation Analysis of Road Accidents Using Netica

Research Article A Bayesian Network Approach to Causation Analysis of Road Accidents Using Netica Hindawi Journal of Advanced Transportation Volume 217, Article ID 2525481, 18 pages https://doi.org/1.1155/217/2525481 Research Article A Bayesian Network Approach to Causation Analysis of Road Accidents

More information

Matching method for emergency plans of highway traffic based on fuzzy sets and rough sets

Matching method for emergency plans of highway traffic based on fuzzy sets and rough sets Journal of Intelligent & Fuzzy Systems 29 (2015) 2421 2427 DOI:10.3233/IFS-151942 IOS Press 2421 Matching method for emergency plans of highway traffic based on fuzzy sets and rough sets Gan Chai a,, Min-min

More information

Two-stage Pedestrian Detection Based on Multiple Features and Machine Learning

Two-stage Pedestrian Detection Based on Multiple Features and Machine Learning 38 3 Vol. 38, No. 3 2012 3 ACTA AUTOMATICA SINICA March, 2012 1 1 1, (Adaboost) (Support vector machine, SVM). (Four direction features, FDF) GAB (Gentle Adaboost) (Entropy-histograms of oriented gradients,

More information

Risk Assessment of Pedestrian Accident Area Using Spatial Analysis and Deep Learning

Risk Assessment of Pedestrian Accident Area Using Spatial Analysis and Deep Learning Risk Assessment of Pedestrian Accident Area Using Spatial Analysis and Deep Learning Ilyoung Hong*, Hanseung Choi, Songpyo Hong Department of GIS Engineering, Namseoul University, Republic of Korea. *

More information

AN ARTIFICIAL NEURAL NETWORK MODEL FOR ROAD ACCIDENT PREDICTION: A CASE STUDY OF KHULNA METROPOLITAN CITY

AN ARTIFICIAL NEURAL NETWORK MODEL FOR ROAD ACCIDENT PREDICTION: A CASE STUDY OF KHULNA METROPOLITAN CITY Proceedings of the 4 th International Conference on Civil Engineering for Sustainable Development (ICCESD 2018), 9~11 February 2018, KUET, Khulna, Bangladesh (ISBN-978-984-34-3502-6) AN ARTIFICIAL NEURAL

More information

Probabilistic Reasoning

Probabilistic Reasoning Course 16 :198 :520 : Introduction To Artificial Intelligence Lecture 7 Probabilistic Reasoning Abdeslam Boularias Monday, September 28, 2015 1 / 17 Outline We show how to reason and act under uncertainty.

More information

Influence Regularity of Fog on Expressway in China

Influence Regularity of Fog on Expressway in China Influence Regularity of Fog on Expressway in China Tang Jun-jun 1, Bai Song-ping 2, He Yong 1, Gao Hai-long 1 1 Research Institute of Highway, MOC, Key Laboratory of Road Safety Technology, MOC, PRC. 8

More information

Stochastic prediction of train delays with dynamic Bayesian networks. Author(s): Kecman, Pavle; Corman, Francesco; Peterson, Anders; Joborn, Martin

Stochastic prediction of train delays with dynamic Bayesian networks. Author(s): Kecman, Pavle; Corman, Francesco; Peterson, Anders; Joborn, Martin Research Collection Other Conference Item Stochastic prediction of train delays with dynamic Bayesian networks Author(s): Kecman, Pavle; Corman, Francesco; Peterson, Anders; Joborn, Martin Publication

More information

An Empirical-Bayes Score for Discrete Bayesian Networks

An Empirical-Bayes Score for Discrete Bayesian Networks An Empirical-Bayes Score for Discrete Bayesian Networks scutari@stats.ox.ac.uk Department of Statistics September 8, 2016 Bayesian Network Structure Learning Learning a BN B = (G, Θ) from a data set D

More information

Learning a probabalistic model of rainfall using graphical models

Learning a probabalistic model of rainfall using graphical models Learning a probabalistic model of rainfall using graphical models Byoungkoo Lee Computational Biology Carnegie Mellon University Pittsburgh, PA 15213 byounko@andrew.cmu.edu Jacob Joseph Computational Biology

More information

Neural Networks & Fuzzy Logic

Neural Networks & Fuzzy Logic Journal of Computer Applications ISSN: 0974 1925, Volume-5, Issue EICA2012-4, February 10, 2012 Neural Networks & Fuzzy Logic Elakkiya Prabha T Pre-Final B.Tech-IT, M.Kumarasamy College of Engineering,

More information

Multi-wind Field Output Power Prediction Method based on Energy Internet and DBPSO-LSSVM

Multi-wind Field Output Power Prediction Method based on Energy Internet and DBPSO-LSSVM , pp.128-133 http://dx.doi.org/1.14257/astl.16.138.27 Multi-wind Field Output Power Prediction Method based on Energy Internet and DBPSO-LSSVM *Jianlou Lou 1, Hui Cao 1, Bin Song 2, Jizhe Xiao 1 1 School

More information

Introduction to Artificial Intelligence. Unit # 11

Introduction to Artificial Intelligence. Unit # 11 Introduction to Artificial Intelligence Unit # 11 1 Course Outline Overview of Artificial Intelligence State Space Representation Search Techniques Machine Learning Logic Probabilistic Reasoning/Bayesian

More information

EFFECT OF HIGHWAY GEOMETRICS ON ACCIDENT MODELING

EFFECT OF HIGHWAY GEOMETRICS ON ACCIDENT MODELING Sustainable Solutions in Structural Engineering and Construction Edited by Saha, S., Lloyd, N., Yazdani, S., and Singh, A. Copyright 2015 ISEC Press ISBN: 978-0-9960437-1-7 EFFECT OF HIGHWAY GEOMETRICS

More information

Using GIS to Identify Pedestrian- Vehicle Crash Hot Spots and Unsafe Bus Stops

Using GIS to Identify Pedestrian- Vehicle Crash Hot Spots and Unsafe Bus Stops Using GIS to Identify Pedestrian-Vehicle Crash Hot Spots and Unsafe Bus Stops Using GIS to Identify Pedestrian- Vehicle Crash Hot Spots and Unsafe Bus Stops Long Tien Truong and Sekhar V. C. Somenahalli

More information

Calculation of Surrounding Rock Pressure Based on Pressure Arch Theory. Yuxiang SONG1,2, a

Calculation of Surrounding Rock Pressure Based on Pressure Arch Theory. Yuxiang SONG1,2, a 5th International Conference on Advanced Materials and Computer Science (ICAMCS 2016) Calculation of Surrounding Rock Pressure Based on Pressure Arch Theory Yuxiang SONG1,2, a 1 School of Civil Engineering,

More information

The Research of Urban Rail Transit Sectional Passenger Flow Prediction Method

The Research of Urban Rail Transit Sectional Passenger Flow Prediction Method Journal of Intelligent Learning Systems and Applications, 2013, 5, 227-231 Published Online November 2013 (http://www.scirp.org/journal/jilsa) http://dx.doi.org/10.4236/jilsa.2013.54026 227 The Research

More information

HIGHWAY TRAFFIC ACCIDENGT CAUSES ANALYSIS BASED ON EXPERT INVESTIGATION AND STATISTICS ANALYSIS METHOD

HIGHWAY TRAFFIC ACCIDENGT CAUSES ANALYSIS BASED ON EXPERT INVESTIGATION AND STATISTICS ANALYSIS METHOD HIGHWAY TRAFFIC ACCIDENGT CAUSES ANALYSIS BASED ON EXPERT INVESTIGATION AND STATISTICS ANALYSIS METHOD Xu Nuo, Chen Lei Research Institute of Highway M.O.T, Road Safety Research Center M.O.T, 00088, Beijing

More information

CS6220: DATA MINING TECHNIQUES

CS6220: DATA MINING TECHNIQUES CS6220: DATA MINING TECHNIQUES Matrix Data: Classification: Part 2 Instructor: Yizhou Sun yzsun@ccs.neu.edu September 21, 2014 Methods to Learn Matrix Data Set Data Sequence Data Time Series Graph & Network

More information

Uncertainty. Introduction to Artificial Intelligence CS 151 Lecture 2 April 1, CS151, Spring 2004

Uncertainty. Introduction to Artificial Intelligence CS 151 Lecture 2 April 1, CS151, Spring 2004 Uncertainty Introduction to Artificial Intelligence CS 151 Lecture 2 April 1, 2004 Administration PA 1 will be handed out today. There will be a MATLAB tutorial tomorrow, Friday, April 2 in AP&M 4882 at

More information

Method for Optimizing the Number and Precision of Interval-Valued Parameters in a Multi-Object System

Method for Optimizing the Number and Precision of Interval-Valued Parameters in a Multi-Object System Method for Optimizing the Number and Precision of Interval-Valued Parameters in a Multi-Object System AMAURY CABALLERO*, KANG YEN*, *, JOSE L. ABREU**, ALDO PARDO*** *Department of Electrical & Computer

More information

Soil geochemical characteristics and prospecting direction of Pengboshan area Inner Mongolia

Soil geochemical characteristics and prospecting direction of Pengboshan area Inner Mongolia 34 4 2015 12 GLOBAL GEOLOGY Vol. 34 No. 4 Dec. 2015 1004-5589 2015 04-0993 - 09 1 2 3 4 1 1 3 1. 150036 2. 157000 3. 130061 4. 024005 66 4 P632. 1 A doi 10. 3969 /j. issn. 1004-5589. 2015. 04. 011 Soil

More information

Recursive estimation of average vehicle time headway using single inductive loop detector data

Recursive estimation of average vehicle time headway using single inductive loop detector data Loughborough University Institutional Repository Recursive estimation of average vehicle time headway using single inductive loop detector data This item was submitted to Loughborough University's Institutional

More information

Bayesian Networks Inference with Probabilistic Graphical Models

Bayesian Networks Inference with Probabilistic Graphical Models 4190.408 2016-Spring Bayesian Networks Inference with Probabilistic Graphical Models Byoung-Tak Zhang intelligence Lab Seoul National University 4190.408 Artificial (2016-Spring) 1 Machine Learning? Learning

More information

EE562 ARTIFICIAL INTELLIGENCE FOR ENGINEERS

EE562 ARTIFICIAL INTELLIGENCE FOR ENGINEERS EE562 ARTIFICIAL INTELLIGENCE FOR ENGINEERS Lecture 16, 6/1/2005 University of Washington, Department of Electrical Engineering Spring 2005 Instructor: Professor Jeff A. Bilmes Uncertainty & Bayesian Networks

More information

Combine Monte Carlo with Exhaustive Search: Effective Variational Inference and Policy Gradient Reinforcement Learning

Combine Monte Carlo with Exhaustive Search: Effective Variational Inference and Policy Gradient Reinforcement Learning Combine Monte Carlo with Exhaustive Search: Effective Variational Inference and Policy Gradient Reinforcement Learning Michalis K. Titsias Department of Informatics Athens University of Economics and Business

More information

A Wavelet Neural Network Forecasting Model Based On ARIMA

A Wavelet Neural Network Forecasting Model Based On ARIMA A Wavelet Neural Network Forecasting Model Based On ARIMA Wang Bin*, Hao Wen-ning, Chen Gang, He Deng-chao, Feng Bo PLA University of Science &Technology Nanjing 210007, China e-mail:lgdwangbin@163.com

More information

Lecture 10: Introduction to reasoning under uncertainty. Uncertainty

Lecture 10: Introduction to reasoning under uncertainty. Uncertainty Lecture 10: Introduction to reasoning under uncertainty Introduction to reasoning under uncertainty Review of probability Axioms and inference Conditional probability Probability distributions COMP-424,

More information

Bayesian Networks BY: MOHAMAD ALSABBAGH

Bayesian Networks BY: MOHAMAD ALSABBAGH Bayesian Networks BY: MOHAMAD ALSABBAGH Outlines Introduction Bayes Rule Bayesian Networks (BN) Representation Size of a Bayesian Network Inference via BN BN Learning Dynamic BN Introduction Conditional

More information

CS6220: DATA MINING TECHNIQUES

CS6220: DATA MINING TECHNIQUES CS6220: DATA MINING TECHNIQUES Chapter 8&9: Classification: Part 3 Instructor: Yizhou Sun yzsun@ccs.neu.edu March 12, 2013 Midterm Report Grade Distribution 90-100 10 80-89 16 70-79 8 60-69 4

More information

The Reliability Evaluation of Electromagnetic Valve of EMUs Based on Two-Parameter Exponential Distribution

The Reliability Evaluation of Electromagnetic Valve of EMUs Based on Two-Parameter Exponential Distribution Send Orders for Reprints to reprints@benthamscience.ae 63 The Open Mechanical Engineering Journal 25 9 63-636 Open Access The Reliability Evaluation of Electromagnetic Valve of EMUs Based on Two-Parameter

More information

Application Research of ARIMA Model in Rainfall Prediction in Central Henan Province

Application Research of ARIMA Model in Rainfall Prediction in Central Henan Province Application Research of ARIMA Model in Rainfall Prediction in Central Henan Province Lulu Xu 1, Dexian Zhang 1, Xin Zhang 1 1School of Information Science and Engineering, Henan University of Technology,

More information

Belief Update in CLG Bayesian Networks With Lazy Propagation

Belief Update in CLG Bayesian Networks With Lazy Propagation Belief Update in CLG Bayesian Networks With Lazy Propagation Anders L Madsen HUGIN Expert A/S Gasværksvej 5 9000 Aalborg, Denmark Anders.L.Madsen@hugin.com Abstract In recent years Bayesian networks (BNs)

More information

Weighted Fuzzy Time Series Model for Load Forecasting

Weighted Fuzzy Time Series Model for Load Forecasting NCITPA 25 Weighted Fuzzy Time Series Model for Load Forecasting Yao-Lin Huang * Department of Computer and Communication Engineering, De Lin Institute of Technology yaolinhuang@gmail.com * Abstract Electric

More information

ENTROPIES OF FUZZY INDISCERNIBILITY RELATION AND ITS OPERATIONS

ENTROPIES OF FUZZY INDISCERNIBILITY RELATION AND ITS OPERATIONS International Journal of Uncertainty Fuzziness and Knowledge-Based Systems World Scientific ublishing Company ENTOIES OF FUZZY INDISCENIBILITY ELATION AND ITS OEATIONS QINGUA U and DAEN YU arbin Institute

More information

Texas A&M University

Texas A&M University Texas A&M University CVEN 658 Civil Engineering Applications of GIS Hotspot Analysis of Highway Accident Spatial Pattern Based on Network Spatial Weights Instructor: Dr. Francisco Olivera Author: Zachry

More information

Humanoid Based Intelligence Control Strategy of Plastic Cement Die Press Work-Piece Forming Process for Polymer Plastics

Humanoid Based Intelligence Control Strategy of Plastic Cement Die Press Work-Piece Forming Process for Polymer Plastics Journal of Materials Science and Chemical Engineering, 206, 4, 9-6 Published Online June 206 in SciRes. http://www.scirp.org/journal/msce http://dx.doi.org/0.4236/msce.206.46002 Humanoid Based Intelligence

More information

Research on Freeway Passenger Flow Prediction Based on Neural. Network

Research on Freeway Passenger Flow Prediction Based on Neural. Network International Journal of Research in Engineering and Science (IJRES) ISSN (Online): 2320-9364, ISSN (Print): 2320-9356 Volume 4 Issue 7 ǁ July. 2016 ǁ PP. 59-66 Research on Freeway Passenger Flow Prediction

More information

CS 188: Artificial Intelligence Fall 2009

CS 188: Artificial Intelligence Fall 2009 CS 188: Artificial Intelligence Fall 2009 Lecture 14: Bayes Nets 10/13/2009 Dan Klein UC Berkeley Announcements Assignments P3 due yesterday W2 due Thursday W1 returned in front (after lecture) Midterm

More information

4: Parameter Estimation in Fully Observed BNs

4: Parameter Estimation in Fully Observed BNs 10-708: Probabilistic Graphical Models 10-708, Spring 2015 4: Parameter Estimation in Fully Observed BNs Lecturer: Eric P. Xing Scribes: Purvasha Charavarti, Natalie Klein, Dipan Pal 1 Learning Graphical

More information

A Multi-Factor HMM-based Forecasting Model for Fuzzy Time Series

A Multi-Factor HMM-based Forecasting Model for Fuzzy Time Series A Multi-Factor HMM-based Forecasting Model for Fuzzy Time Series Hui-Chi Chuang, Wen-Shin Chang, Sheng-Tun Li Department of Industrial and Information Management Institute of Information Management National

More information

Unobserved Heterogeneity and the Statistical Analysis of Highway Accident Data. Fred Mannering University of South Florida

Unobserved Heterogeneity and the Statistical Analysis of Highway Accident Data. Fred Mannering University of South Florida Unobserved Heterogeneity and the Statistical Analysis of Highway Accident Data Fred Mannering University of South Florida Highway Accidents Cost the lives of 1.25 million people per year Leading cause

More information

Seismic hazard expression in risk assessment

Seismic hazard expression in risk assessment Earthquae Resistant Engineering Structures VI 299 Seismic hazard expression in ris assessment X.-X. Tao 1, 2, Z.-R. Tao 2 & P. Li 1 1 Harbin Institute of Technology, People s Republic of China 2 Institute

More information

Predicting Causes of Traffic Road Accidents Using Multi-class Support Vector Machines

Predicting Causes of Traffic Road Accidents Using Multi-class Support Vector Machines Journal of Communication and Computer 11(2014) 441-447 doi: 10.17265/1548-7709/2014.05 004 D DAVID PUBLISHING Predicting Causes of Traffic Road Accidents Using Multi-class Support Vector Machines Elfadil

More information

Development of Decision Support Tools to Assess Pedestrian and Bicycle Safety: Focus on Population, Demographic and Socioeconomic FINAL REPORT

Development of Decision Support Tools to Assess Pedestrian and Bicycle Safety: Focus on Population, Demographic and Socioeconomic FINAL REPORT TRCLC 14-07 June 30, 2016 Development of Decision Support Tools to Assess Pedestrian and Bicycle Safety: Focus on Population, Demographic and Socioeconomic Spectra FINAL REPORT Deo Chimba, PhD, PE., PTOE

More information

Bayesian Networks: Construction, Inference, Learning and Causal Interpretation. Volker Tresp Summer 2014

Bayesian Networks: Construction, Inference, Learning and Causal Interpretation. Volker Tresp Summer 2014 Bayesian Networks: Construction, Inference, Learning and Causal Interpretation Volker Tresp Summer 2014 1 Introduction So far we were mostly concerned with supervised learning: we predicted one or several

More information

Selection of the Appropriate Lag Structure of Foreign Exchange Rates Forecasting Based on Autocorrelation Coefficient

Selection of the Appropriate Lag Structure of Foreign Exchange Rates Forecasting Based on Autocorrelation Coefficient Selection of the Appropriate Lag Structure of Foreign Exchange Rates Forecasting Based on Autocorrelation Coefficient Wei Huang 1,2, Shouyang Wang 2, Hui Zhang 3,4, and Renbin Xiao 1 1 School of Management,

More information

GREEN SHEET. California Department of Forestry and Fire Protection (CAL FIRE)

GREEN SHEET. California Department of Forestry and Fire Protection (CAL FIRE) GREEN SHEET California Department of Forestry and Fire Protection (CAL FIRE) Informational Summary Report of Serious CAL FIRE Injuries, Illnesses, Accidents and Near Serious Accidents Fire Engine Rollover

More information

Beyond Uniform Priors in Bayesian Network Structure Learning

Beyond Uniform Priors in Bayesian Network Structure Learning Beyond Uniform Priors in Bayesian Network Structure Learning (for Discrete Bayesian Networks) scutari@stats.ox.ac.uk Department of Statistics April 5, 2017 Bayesian Network Structure Learning Learning

More information

ECE 6504: Advanced Topics in Machine Learning Probabilistic Graphical Models and Large-Scale Learning

ECE 6504: Advanced Topics in Machine Learning Probabilistic Graphical Models and Large-Scale Learning ECE 6504: Advanced Topics in Machine Learning Probabilistic Graphical Models and Large-Scale Learning Topics Summary of Class Advanced Topics Dhruv Batra Virginia Tech HW1 Grades Mean: 28.5/38 ~= 74.9%

More information

The Failure-tree Analysis Based on Imprecise Probability and its Application on Tunnel Project

The Failure-tree Analysis Based on Imprecise Probability and its Application on Tunnel Project 463 A publication of CHEMICAL ENGINEERING TRANSACTIONS VOL. 59, 2017 Guest Editors: Zhuo Yang, Junjie Ba, Jing Pan Copyright 2017, AIDIC Servizi S.r.l. ISBN 978-88-95608-49-5; ISSN 2283-9216 The Italian

More information

VISUAL EXPLORATION OF SPATIAL-TEMPORAL TRAFFIC CONGESTION PATTERNS USING FLOATING CAR DATA. Candra Kartika 2015

VISUAL EXPLORATION OF SPATIAL-TEMPORAL TRAFFIC CONGESTION PATTERNS USING FLOATING CAR DATA. Candra Kartika 2015 VISUAL EXPLORATION OF SPATIAL-TEMPORAL TRAFFIC CONGESTION PATTERNS USING FLOATING CAR DATA Candra Kartika 2015 OVERVIEW Motivation Background and State of The Art Test data Visualization methods Result

More information

Discussion Papers In Economics And Business

Discussion Papers In Economics And Business Discussion Papers In Economics And Business Smeed s Law and the Role of Hospitals in Modeling Fatalities and Traffic Accidents Yueh-Tzu Lu and Mototsugu Fukushige Discussion Paper 17-22 Graduate School

More information

Related Concepts: Lecture 9 SEM, Statistical Modeling, AI, and Data Mining. I. Terminology of SEM

Related Concepts: Lecture 9 SEM, Statistical Modeling, AI, and Data Mining. I. Terminology of SEM Lecture 9 SEM, Statistical Modeling, AI, and Data Mining I. Terminology of SEM Related Concepts: Causal Modeling Path Analysis Structural Equation Modeling Latent variables (Factors measurable, but thru

More information

CHAPTER-17. Decision Tree Induction

CHAPTER-17. Decision Tree Induction CHAPTER-17 Decision Tree Induction 17.1 Introduction 17.2 Attribute selection measure 17.3 Tree Pruning 17.4 Extracting Classification Rules from Decision Trees 17.5 Bayesian Classification 17.6 Bayes

More information

Bayesian Networks to design optimal experiments. Davide De March

Bayesian Networks to design optimal experiments. Davide De March Bayesian Networks to design optimal experiments Davide De March davidedemarch@gmail.com 1 Outline evolutionary experimental design in high-dimensional space and costly experimentation the microwell mixture

More information

Indirect Clinical Evidence of Driver Inattention as a Cause of Crashes

Indirect Clinical Evidence of Driver Inattention as a Cause of Crashes University of Iowa Iowa Research Online Driving Assessment Conference 2007 Driving Assessment Conference Jul 10th, 12:00 AM Indirect Clinical Evidence of Driver Inattention as a Cause of Crashes Gary A.

More information

Study on Shandong Expressway Network Planning Based on Highway Transportation System

Study on Shandong Expressway Network Planning Based on Highway Transportation System Study on Shandong Expressway Network Planning Based on Highway Transportation System Fei Peng a, Yimeng Wang b and Chengjun Shi c School of Automobile, Changan University, Xian 71000, China; apengfei0799@163.com,

More information

Freeway Travel Time Forecast Using Artifical Neural Networks With Cluster Method

Freeway Travel Time Forecast Using Artifical Neural Networks With Cluster Method 12th International Conference on Information Fusion Seattle, WA, USA, July 6-9, 2009 Freeway Travel Time Forecast Using Artifical Neural Networks With Cluster Method Ying LEE Department of Hospitality

More information

Bayesian Inference. Definitions from Probability: Naive Bayes Classifiers: Advantages and Disadvantages of Naive Bayes Classifiers:

Bayesian Inference. Definitions from Probability: Naive Bayes Classifiers: Advantages and Disadvantages of Naive Bayes Classifiers: Bayesian Inference The purpose of this document is to review belief networks and naive Bayes classifiers. Definitions from Probability: Belief networks: Naive Bayes Classifiers: Advantages and Disadvantages

More information

Uncertainty and Bayesian Networks

Uncertainty and Bayesian Networks Uncertainty and Bayesian Networks Tutorial 3 Tutorial 3 1 Outline Uncertainty Probability Syntax and Semantics for Uncertainty Inference Independence and Bayes Rule Syntax and Semantics for Bayesian Networks

More information

Outline. CSE 573: Artificial Intelligence Autumn Agent. Partial Observability. Markov Decision Process (MDP) 10/31/2012

Outline. CSE 573: Artificial Intelligence Autumn Agent. Partial Observability. Markov Decision Process (MDP) 10/31/2012 CSE 573: Artificial Intelligence Autumn 2012 Reasoning about Uncertainty & Hidden Markov Models Daniel Weld Many slides adapted from Dan Klein, Stuart Russell, Andrew Moore & Luke Zettlemoyer 1 Outline

More information

Effect of Environmental Factors on Free-Flow Speed

Effect of Environmental Factors on Free-Flow Speed Effect of Environmental Factors on Free-Flow Speed MICHAEL KYTE ZAHER KHATIB University of Idaho, USA PATRICK SHANNON Boise State University, USA FRED KITCHENER Meyer Mohaddes Associates, USA ABSTRACT

More information

Directed Graphical Models

Directed Graphical Models CS 2750: Machine Learning Directed Graphical Models Prof. Adriana Kovashka University of Pittsburgh March 28, 2017 Graphical Models If no assumption of independence is made, must estimate an exponential

More information

A FUZZY NEURAL NETWORK MODEL FOR FORECASTING STOCK PRICE

A FUZZY NEURAL NETWORK MODEL FOR FORECASTING STOCK PRICE A FUZZY NEURAL NETWORK MODEL FOR FORECASTING STOCK PRICE Li Sheng Institute of intelligent information engineering Zheiang University Hangzhou, 3007, P. R. China ABSTRACT In this paper, a neural network-driven

More information

arxiv: v1 [cs.cv] 28 Nov 2017

arxiv: v1 [cs.cv] 28 Nov 2017 A fatal point concept and a low-sensitivity quantitative measure for traffic safety analytics arxiv:1711.10131v1 [cs.cv] 28 Nov 2017 Shan Suthaharan Department of Computer Science University of North Carolina

More information

Traffic Signal Control with Swarm Intelligence

Traffic Signal Control with Swarm Intelligence 009 Fifth International Conference on Natural Computation Traffic Signal Control with Swarm Intelligence David Renfrew, Xiao-Hua Yu Department of Electrical Engineering, California Polytechnic State University

More information

Bayesian Networks: Construction, Inference, Learning and Causal Interpretation. Volker Tresp Summer 2016

Bayesian Networks: Construction, Inference, Learning and Causal Interpretation. Volker Tresp Summer 2016 Bayesian Networks: Construction, Inference, Learning and Causal Interpretation Volker Tresp Summer 2016 1 Introduction So far we were mostly concerned with supervised learning: we predicted one or several

More information

Probabilistic Graphical Networks: Definitions and Basic Results

Probabilistic Graphical Networks: Definitions and Basic Results This document gives a cursory overview of Probabilistic Graphical Networks. The material has been gleaned from different sources. I make no claim to original authorship of this material. Bayesian Graphical

More information

Explaining Results of Neural Networks by Contextual Importance and Utility

Explaining Results of Neural Networks by Contextual Importance and Utility Explaining Results of Neural Networks by Contextual Importance and Utility Kary FRÄMLING Dep. SIMADE, Ecole des Mines, 158 cours Fauriel, 42023 Saint-Etienne Cedex 2, FRANCE framling@emse.fr, tel.: +33-77.42.66.09

More information

Bayesian Approach 2. CSC412 Probabilistic Learning & Reasoning

Bayesian Approach 2. CSC412 Probabilistic Learning & Reasoning CSC412 Probabilistic Learning & Reasoning Lecture 12: Bayesian Parameter Estimation February 27, 2006 Sam Roweis Bayesian Approach 2 The Bayesian programme (after Rev. Thomas Bayes) treats all unnown quantities

More information

The Analysis of Traffic Accidents in Erzurum Province and Its Districts Through Use of Geographical Information Systems

The Analysis of Traffic Accidents in Erzurum Province and Its Districts Through Use of Geographical Information Systems Journal of Traffic and Logistics Engineering Vol. 3, No. 2, December 2015 The Analysis of Traffic Accidents in Erzurum Province and Its Districts Through Use of Geographical Information Systems Ahmet Tortum1,

More information

Research on Heat Conduction Inverse Problem of Continuous Long Downhill Truck Brake

Research on Heat Conduction Inverse Problem of Continuous Long Downhill Truck Brake International Conference on Civil, Transportation and Environment (ICCTE 2016) Research on Heat Conduction Inverse Problem of Continuous Long Downhill Truck Brake Shun Zeng1, a,heng Zhang2,b,Yunwei Meng1,c

More information

DEVELOPMENT OF TRAFFIC ACCIDENT ANALYSIS SYSTEM USING GIS

DEVELOPMENT OF TRAFFIC ACCIDENT ANALYSIS SYSTEM USING GIS DEVELOPMENT OF TRAFFIC ACCIDENT ANALYSIS SYSTEM USING GIS Masayuki HIRASAWA Researcher Traffic Engineering Division Civil Engineering Research Institute of Hokkaido 1-3 Hiragishi, Toyohira-ku, Sapporo,

More information

Learning in Bayesian Networks

Learning in Bayesian Networks Learning in Bayesian Networks Florian Markowetz Max-Planck-Institute for Molecular Genetics Computational Molecular Biology Berlin Berlin: 20.06.2002 1 Overview 1. Bayesian Networks Stochastic Networks

More information

Lecture 6: Graphical Models: Learning

Lecture 6: Graphical Models: Learning Lecture 6: Graphical Models: Learning 4F13: Machine Learning Zoubin Ghahramani and Carl Edward Rasmussen Department of Engineering, University of Cambridge February 3rd, 2010 Ghahramani & Rasmussen (CUED)

More information

STA 4273H: Statistical Machine Learning

STA 4273H: Statistical Machine Learning STA 4273H: Statistical Machine Learning Russ Salakhutdinov Department of Statistics! rsalakhu@utstat.toronto.edu! http://www.utstat.utoronto.ca/~rsalakhu/ Sidney Smith Hall, Room 6002 Lecture 3 Linear

More information

Data Mining Part 5. Prediction

Data Mining Part 5. Prediction Data Mining Part 5. Prediction 5.5. Spring 2010 Instructor: Dr. Masoud Yaghini Outline How the Brain Works Artificial Neural Networks Simple Computing Elements Feed-Forward Networks Perceptrons (Single-layer,

More information

K. Nishijima. Definition and use of Bayesian probabilistic networks 1/32

K. Nishijima. Definition and use of Bayesian probabilistic networks 1/32 The Probabilistic Analysis of Systems in Engineering 1/32 Bayesian probabilistic bili networks Definition and use of Bayesian probabilistic networks K. Nishijima nishijima@ibk.baug.ethz.ch 2/32 Today s

More information

Investigation of Road Traffic Fatal Accidents Using Data Mining Techniques

Investigation of Road Traffic Fatal Accidents Using Data Mining Techniques Investigation of Road Traffic Fatal Accidents Using Data Mining Techniques M. Nagaraju *1 B. Bhavani *2 K.Rohith *3 *1,2,3 Assistant Professor Department of Computer Science & Engineering *1,2,3 Nalla

More information

AN ANALYSIS ON THE TRAFFIC ACCIDENTS TOURIST AT CASE STUDY: NANTOU COUNTY

AN ANALYSIS ON THE TRAFFIC ACCIDENTS TOURIST AT CASE STUDY: NANTOU COUNTY AN ANALYSIS ON THE TRAFFIC ACCIDENTS TOURIST AT CASE STUDY: NANTOU COUNTY Jau-Ming Su 1, Yu-Ming Wang 2 1 Chung Hua University, Ph.D.program of Technology Management, No. 707, Sec. 2, WuFu Rd., Hsin Chu,

More information

CS 354R: Computer Game Technology

CS 354R: Computer Game Technology CS 354R: Computer Game Technology AI Fuzzy Logic and Neural Nets Fall 2017 Fuzzy Logic Philosophical approach Decisions based on degree of truth Is not a method for reasoning under uncertainty that s probability

More information

Bayesian Concept Learning

Bayesian Concept Learning Learning from positive and negative examples Bayesian Concept Learning Chen Yu Indiana University With both positive and negative examples, it is easy to define a boundary to separate these two. Just with

More information

Traffic Accident Analysis of Sun Glare and Twilight Shortly Before and after Sunset in Chiba Prefecture, Japan

Traffic Accident Analysis of Sun Glare and Twilight Shortly Before and after Sunset in Chiba Prefecture, Japan Traffic Accident Analysis of Sun Glare and Twilight Shortly Before and after in Chiba Prefecture, Japan Kenji Hagita a, Kenji Mori b a,b Traffic Science Division, National Research Institute of Police

More information

Accident Analysis and Prevention

Accident Analysis and Prevention Accident Analysis and Prevention 94 (2016) 59 64 Contents lists available at ScienceDirect Accident Analysis and Prevention journal homepage: www.elsevier.com/locate/aap Utilizing the eigenvectors of freeway

More information

Machine learning: lecture 20. Tommi S. Jaakkola MIT CSAIL

Machine learning: lecture 20. Tommi S. Jaakkola MIT CSAIL Machine learning: lecture 20 ommi. Jaakkola MI CAI tommi@csail.mit.edu opics Representation and graphical models examples Bayesian networks examples, specification graphs and independence associated distribution

More information

System Reliability Allocation Based on Bayesian Network

System Reliability Allocation Based on Bayesian Network Appl. Math. Inf. Sci. 6, No. 3, 681-687 (2012) 681 Applied Mathematics & Information Sciences An International Journal System Reliability Allocation Based on Bayesian Network Wenxue Qian 1,, Xiaowei Yin

More information

Information. A Fitting Approach. 1. Introduction

Information. A Fitting Approach. 1. Introduction www.ijcsi.org 150 A Fitting Approach to Mend Defective Urban Traffic Flow Information Based on SARBF Neural Networks Ning Chen, Weibing Weng and Xing Xu School of Mechanical and Automotive Engineering,

More information

CS 188: Artificial Intelligence Fall 2008

CS 188: Artificial Intelligence Fall 2008 CS 188: Artificial Intelligence Fall 2008 Lecture 14: Bayes Nets 10/14/2008 Dan Klein UC Berkeley 1 1 Announcements Midterm 10/21! One page note sheet Review sessions Friday and Sunday (similar) OHs on

More information

A new Approach to Drawing Conclusions from Data A Rough Set Perspective

A new Approach to Drawing Conclusions from Data A Rough Set Perspective Motto: Let the data speak for themselves R.A. Fisher A new Approach to Drawing Conclusions from Data A Rough et Perspective Zdzisław Pawlak Institute for Theoretical and Applied Informatics Polish Academy

More information

Outlier Detection and Correction for the Deviations of Tooth Profiles of Gears

Outlier Detection and Correction for the Deviations of Tooth Profiles of Gears 0.2478/msr-203-003 MEASUREMENT SCIENCE REVIEW, Volume 3, No. 2, 203 Outlier Detection and Correction for the Deviations of Tooth Profiles of Gears Han Lianfu,2, Fu Changfeng 2, Wang Jun, Tang Wenyan Institute

More information

Bayesian Networks Basic and simple graphs

Bayesian Networks Basic and simple graphs Bayesian Networks Basic and simple graphs Ullrika Sahlin, Centre of Environmental and Climate Research Lund University, Sweden Ullrika.Sahlin@cec.lu.se http://www.cec.lu.se/ullrika-sahlin Bayesian [Belief]

More information

Machine Learning Overview

Machine Learning Overview Machine Learning Overview Sargur N. Srihari University at Buffalo, State University of New York USA 1 Outline 1. What is Machine Learning (ML)? 2. Types of Information Processing Problems Solved 1. Regression

More information

COMP538: Introduction to Bayesian Networks

COMP538: Introduction to Bayesian Networks COMP538: Introduction to Bayesian Networks Lecture 9: Optimal Structure Learning Nevin L. Zhang lzhang@cse.ust.hk Department of Computer Science and Engineering Hong Kong University of Science and Technology

More information

Machine Learning Summer School

Machine Learning Summer School Machine Learning Summer School Lecture 3: Learning parameters and structure Zoubin Ghahramani zoubin@eng.cam.ac.uk http://learning.eng.cam.ac.uk/zoubin/ Department of Engineering University of Cambridge,

More information

The Effect of Sun Glare on Traffic Accidents in Chiba Prefecture, Japan

The Effect of Sun Glare on Traffic Accidents in Chiba Prefecture, Japan Asian Transport Studies, Volume 3, Issue 2 (2014), 205 219. 2014 ATS All rights reserved The Effect of Sun Glare on Traffic Accidents in Chiba Prefecture, Japan Kenji HAGITA a*, Kenji MORI b a Traffic

More information

A Generalized Decision Logic in Interval-set-valued Information Tables

A Generalized Decision Logic in Interval-set-valued Information Tables A Generalized Decision Logic in Interval-set-valued Information Tables Y.Y. Yao 1 and Qing Liu 2 1 Department of Computer Science, University of Regina Regina, Saskatchewan, Canada S4S 0A2 E-mail: yyao@cs.uregina.ca

More information

Brief Introduction of Machine Learning Techniques for Content Analysis

Brief Introduction of Machine Learning Techniques for Content Analysis 1 Brief Introduction of Machine Learning Techniques for Content Analysis Wei-Ta Chu 2008/11/20 Outline 2 Overview Gaussian Mixture Model (GMM) Hidden Markov Model (HMM) Support Vector Machine (SVM) Overview

More information