EVALUVATINGROAD TRAFFIC ACCIDENTS USING DATAMININGTECHNOLOGY ABSTRACTRoadtraffic safety is an important perturbation for government transportauthorities as well as common people.
Road accidents are ambivalent and notable to be predict the accidents and their survey requires the factorsaffecting them. Road accidents cause difficulties which are higher at analarming rate. Controlling the traffic accidents on roads is a crucial task. Togive safe driving suggestions, clear and careful study of roadway traffic datais critical. Increasing the number of vehicles from past few years has put lotof pressure on the existing roads and ultimately resulting in increasing theroad accidents. A road traffic accident is any harm due to collisionoriginating from, terminating with or involving a vehicle partially or fully ona public road.1.
INTRODUCTION In modern life, accidents havebecome daily happening. Every day we hear the news of theaccident on thetelevision, or through internet .During accident many people die at the spot, someothers may injured very severely. By witnessing an accident one can understandthe horror of it. There are several reasons for road accidents, some of themare increasing the number of vehicles, careless driving, violating trafficrules etc. Whenever a road accident occur there are various types of damagetakes place ,which could be in the form of human beings, infrastructure whichis damage to the government and many other administration damages .
Poorroadway maintenance also contributes accidents. But still many people continueto neglect and ignore the danger involved in the accidents. In this paper weare analyzing some methods and algorithms to find out the problems occur inroad accidents.Section2 elucidate literature survey, Section 3 elucidate conclusion.
2. LITERATURE SURVEY Thepaper 1 describes the association rule mining, its classifications and theatmospheric components like roadway surface, climate, and light condition donot strongly influence the fatal accident rate. But the human factors likebeing alcoholic or not, and the impact have strongly affect on the fatal accidentrate. A common mechanism to recognize the relations between the data stored inhuge database and plays a very significant role in repeated object set miningis association rule mining algorithm. A classical association rule miningmethod is the Apriori algorithm whose main aim is to identify repeated object setsto analyze the roadway traffic data. Classification in data mining methodology focusatbuilding a classifier model from a training data set that is used to classifyrecords of unrevealed class labels. The Naïve Bayes technique is one of the probabilitybased methods for classification and is based on the Bayes’ hypothesis with theprobability of self rule between every set of variables.
The author applies statisticsanalysis and Fatal Accident Reporting System (FARS) tosolve this problem. Fromthe clustering result some regions have larger fatal rate but some others havesmaller. When driving within those risky or dangerous states,people take moreattention. When the task performed, data seems never to be sufficient to make astrong choice. If non-fatal accident data, weather condition data, mileagedata, and so on are available, more test could be executed thus more advicecould be made from the data.In paper 2, K-modes clustering techniqueis a frameworkthat is used as an initialwork for divisionof different road accidents on road network.Then association rule mining are used to recognize the various situations thatare related with the occurrence of an accident for the entire data set (EDS)and the clusters recognized by K-modes clustering algorithm. Six clusters (C1toC6)are used based on properties accident type, road type, lightning on road androad feature identified by K modes clustering method.
On each cluster associationrule mining is applied as well as on EDS to create rules. Powerful methods withhigher raise values are taken for the inspection. Rules for various clustersdisclose the situations related with the accidents within that cluster. Theserules are compared with the rules created for the EDS and resemblance showsthat association rules for EDS does not disclose correct data that can berelated with an accident. If more feature are presented large information canbe identified that is associated with an accident.
To buildup our methodology,we also performed analysis of all clusters and EDS on monthly or hourly basis.The results of analysis assist methodology that performing clustering prior toanalysis helps to identify better and useful results that cannot obtain withoutusing cluster analysis.The paper 3 performsstatistical and empirical analysison State Highways and Ordinary District Roads accidental datasets.
The need ofthe study is to analyze the traffic accident data of SH’s and ODR’s to assignthe black spots and accidental elements, part to control the harm caused by theaccidents. The basic necessity of the analysis is to check the trafficassociated dataset through Exploratory Visualization Techniques, K-means andKNN Algorithms using Rstudio.. The term accident black spot in management ofroad traffic safety defines a place where accidents are been focus historicallyand to analyze the accidental data using exploratory visualization techniquesand machine learning algorithms. These techniques and algorithms are used onthe traffic accidental dataset to get the desired output in order to reduce theaccident frequency. ExploratoryVisualization Technique is a technique to anatomize and examine the sets ofdata in order to abridge and encapsulate the important characteristics withvisual and pictorial method.
Exploratory Visualization analysis can beperformed using scatter plot, correlation analysis, barplot, clustered barplot,histogram, pie chart etc. Machine learning concentrates on algorithm designingand makes predictions on sets of data. It includes Supervised (KNN Algorithm)and Unsupervised learning (K-means Algorithm).This paper present result byresembling the above three miningtechniques and assigns the cause of accident, accident prone area, analyze thetime of accident, examine the cause of accident and scrutinize the litigators vehicle.In paper 4, describes about a frame work that usesK-mode clustering technique as aprimary task for dividing 11574 accidents onroad network of Dehradun (India) from 2009 to 2014. Then an association miningrule are used to find out the various context associated with instance of anaccident for both the whole data set and clusters find out by K-modesclustering algorithm. Then compare the findings from cluster based analysis andentire data set.
The results shows that the amalgamation of k mode clusteringand association mining rule is very encouraging, as it produces important factsthat would remain hidden if no segmentation has been performed prior to generateassociation rules. Also a trend analysis has been performed on each clustersand entire data set. By trendanalysis it shows that before analysis, priorsegmentation of data is very important. This paper put forward a frame workbased on cluster analysis using k-mode algorithm and association mining rule.
By using cluster analysis as a primary task can group the data into differenthomogeneous parts. It is the first time that both association and clustering ruleare used together to analyze the data’s for road accidents. The output of thestudy proves that by using cluster analysis as a primary task, it can help inremoving heterogeneity to some extent in the road accident data.) Based onattributes accident type, road type, lightning on road and road feature, K -modesclustering find six cluster (C1–C6). Association mining rule have been appliedon each cluster as well as on entire data set to generate rules. For thisanalysis strong rules with high lift values are used. Thepaper 5 describes purpose of data mining methods in the field of roadaccident investigation. Association rules are used to identify the patterns andrules that are subjected the cause the occurrence of road accidents.
Anefficient method for updating the index year after year could be designed.Additionally, further analysis of traffic safety data using data miningtechniques are allowed.Clusteranalysis evaluates data objects without consulting a common class label.The objects are clustered or arranged on the basis of maximizing the intraclass similarity and minimizing the interclass similarity.
Outlier analysis: A database having dataobjects that do not satisfies the general behavior or model of the data. Thesedata objects are also called outliers.Evolution analysis which defines and models consistencies or trends forobjects whose behavior changes over time.We are currently build up byconsidering several issues, changes in clash occurrence may have some after effectfor traffic safety measures in certain countries.
The determination of specificprecautionary measures to overcome clashes requires study of other factors suchas the identification of specific road sections that need work, etc..Itanalyzed the traffic accident using data mining technique that could possiblyreduce the fatality rate. Using a road safety database enables to reduce thefatality by implementing road safety programs at local and national levels. Thepaper 6 describes datamining techniques to analyze high-frequency accident locations and further identifydifferent factors that affect road accidents at specifying locations. We firstpartitioned the accident locations into k groupsbased on their accident frequency poll using k means clustering algorithm.
Association rule mining algorithm is used to reveal the correlation betweendifferent elements in the accident data and understand the characteristics ofthese locations. Hence, the major significance will be the evaluation of theoutcomes. Datamining has been proven as a reliable technique to analyzing road accident data.Several data mining techniques such as clustering, classification andassociation rule mining are widely used in the literature to identify reasonsthat affect the severity of road accidents. It is the first time that k-means algorithmis used to identify high- and low-frequency accident locations based onaccident count as it provides some technical measures to divide the accidentlocations based on threshold values.The road accident dataset and its analysisusing k-means clustering and association rulemining algorithm shows that this approach can be reused on other accident datawith more attributes to identify various other factors associated with roadaccidents.In paper 7 describes the results from analysis of traffic accidents onthe Finnish roads by applying large scale data mining methods.
The set of datacollected from road traffic accidents are vast, multidimensional and diverse.TheFinnish Road Administration between 2004 and 2008 data was collected for thisstudy. This set of data contains more than 83000 accidents and 1203 of whichare fatal. The main aim of this is to examine the usability of robust clustering,association and frequent itemsets, and visualization methods to the roadtrafficaccident analysis. The output shows that the pick out data miningmethods areable to produce intelligible patterns from the data, detectingmore information thatcould be increased with more detailed and comprehensive data sets.Most of the fatalaccidents occur due to the condition of single roadway mainroads outside built-upareas where the permitted speed varies typically between 80-100km/h. Aged andyoung drivers have large contribution to the high risk accidents inhighways.Most of the surveys reported that one of the major reasons for accidents amongyoung people are consumption of alcohol.
From the analysis it is understandthat failure of roads and end user groups are responsible for accidents atcertain limit. This paper 8 is to represent aTraffic Accident Report and Analysis System (TARAS) through data mining usingClustering technique. Detect the causes of accidents is the main aim of thispaper. The transport department of government of India produced the dataset forthe study contains traffic accident records of the yearand look into theperformance of J48. The classification accuracy on the test result disclosesthe three cases such as accident, vehicle and casualty. Genetic Algorithms isused for the future selection to lower the measurements of the dataset..Moredetailed area specific information from accident locations and circumstancesare needed.
With the help of this paper, the analysis can be done and thereforepreventive measures can be taken. It can help the government to keep track ofrecords of the accidents, causes of accident, vehicle number, vehicle owner’sname and address.. With the current data it is possible to identify the riskyroad segments and the road user groups responsible for accidents in certain environments.
The viewer or user can also make their own account for viewing the site .youcan view the data about causality .Our system will provide the graphical viewof the accidents with respect to the data entered into the system according tothe period .This system will provide the solutions as accidents causes. So thatwith the help of thissystem government can take the necessary actionsaccordingaccidents cases.1) Accurate Location ofaccident2) GPS integration3) Government ID Authenticationfor user Data4) Advanced Filtertechnique Accident Solutionprediction.
The paper9 describes application of data mining techniques on road accidents by usingmachine learning algorithms that determines accident rate in the future todecrease clash deaths and wounds.The accident dataset contains traffic accidentreport of various cities examined by using machine learningalgorithms topredict the accident rate.It implemented hybrid approach that performed withhigheraccuracy rate as compared to other methods to be described. The machinelearning techniques is used for to reduce accidents and saves life.We have toexpand the classification accuracy of road traffic accidents types, dataquality has to be added.In paper 10describes about a method called Innovators Marketplace on Data Jackets.Innovators Marketplace on Data Jackets used to externalize the value of datathrough ally. For analyzing the rate of traffic accidents on urban area methods such as factor analysis, structureequation modeling and data mining are used here.
To construct traffic accidentrisk evaluation model different indexes such as total number of accidentsreported, fatality rate injury rate arecombined. To identify the connection between different factors population structureinformation, vehicle information, road characters are used. In here we focusedon urban data, applied structural equation modeling to find out theimportantfactors associated with traffic accident. Important factors are population structure, vehicle information,structure of road etc. This paper describes six factors by constructing anaccident risk causal framework based on urban data and thecomponent factor setsof each feature and influence on traffic accident.3. CONCLUSIONIn this paper,we have collected different researchers works together in one document asanalysis and examined about the contribution towards the effects of road andtraffic accident on human life and society.
This survey focused the number ofapproaches used to avoid the accident happened in various cities and countries.The study on road traffic accident is to identify the key element quickly andefficiently to provide instructional methods to prevent or to reduce the roadtraffic accidents. Meanwhile, it would be helpful for improving the efficiencyand security service level of the road transportation system.
The paper alsodiscussing about various data mining techniques which is proved supporting toresolve traffic accident severity problem and conclude which one could beoptimal technique in road traffic accident scenario. From our study, we conclude that Associationrule is an important method to analyze road traffic accidents. The brief surveywill also help us to find better mining technique in this kind of problem. REFERENCE1 “Analysis of Road Traffic Fatal AccidentsUsing Data Mining Techniques” LilingLi, Sharad Shrestha, Gongzhu Hu2 “Analyzing road accident data using associationrule mining” Sachin Kumar; DurgaToshniwa3 “Black Spot and Accidental AttributesIdentification on State Highways and Ordinary District Roads Using Data Mining Techniques”. Gagandeep Kaur 4″A data mining framework to analyze road accident data” Sachin Kumar, Durga Toshniwal5″An overview of data mining in road traffic and accident analysis”. K.
Jayasudha, Dr. C. Chandrasekar6″ A data mining approach to characterize road accident locations”. Sachin KumarEmail author,Durga Toshniwal7″Miningroad traffic accidents”. Sami Ayramo,Pasi Pirtala,Janne Kauttonen,KashifNaveed,Tommi Karkk ainen8 “Traffic Accident Report Analysis using DataMining Techniques”. Mrs.
Kanchan Gawande1 Ambikesh Pandey9″ A Radical Approach to Forecast the RoadAccident Using Data Mining Technique”. AnupamaMakkar, Harpreet Singh Gill10 Evaluating model oftraffic accident rate on urban dataJianshi Wang,Yukio Ohsawa