SlideShare a Scribd company logo
An Approach For Predicting
Road Accident Severity
B. Sikander
MTech Scholar 188150002
Under the Supervision of:
Dr. Anant Ram
Dept. of Computer Engineering & Applications
CONTENT
Area of Research
Introduction
Motivation
Applications
Issues and Challenges
Objective
Proposed Approach
Result
References
Area of Research
Accident Analysis and Prevention
3
Introduction
• Road traffic injuries and deaths have been a major public health issue globally. According to
World Health Organization (WHO), approximately 1.35 million people die from roadway traffic
accidents each year, while 20∼50 million people suffer nonfatal injuries with many resulting in
disabilities [1].
• Road accidents in India claimed over 1.5 lakh lives in the country in 2018, with over-speeding
of vehicles being the biggest reason for casualties[2].
• The ministry of road transport and highways issued a report on Road accidents in India in 2018,
which showed that road accidents last year increased by 0.46% as compared to 2017[2].
4
Cont’d…
5
Figure 1: Road Accidents Statistics from 2007 to 2017 [3]
Motivation
6
• Increased highway accidents and rise in death toll every day.
• A total of 4,67,044 road accidents have been reported by States and Union Territories (UTs) in
the calendar year 2018, claiming 1,51,417 lives and causing injuries to 4,69,418 persons,” the
report said. Over-speeding accounted for 64.4% of the persons killed[2].
Figure 2. Car Accident due to over speeding[2]
Cont’d…
7
Figure 3: Number of Road Accidents, Deaths and Injuries [4]
Applications
8
Help in designing road geometry
Help in finding the those spot on road where chance of accident is high.
Help in installing the road surveillance
Road network simulation
Help in study of Road Intersection
Issues and Challenges
9
Proper selection of features which can cause a road accident.
Removal of unwanted features.
Collected data should be valid and related to real life accident situations.
Removal of outlier instances
Objective
10
To evaluate factors contributing to road accidents.
To review related systems and models for predicting the likelihood of causing an accident.
To develop an algorithm for predicting the likelihood of causing an accident.
To test and validate the developed algorithm.
Proposed Approach
11
• In this research we build a model which able to predict the road accident based on road conditions.
• First we consider 50 possible features/variables which can cause road accidents.
• Vehicle type : car, truck, motorcycle.
• Vehicle length : near by vehicle length.
• Road Type : One way street, Single carriageway, dual carriageway.
• Light condition : daylight, darkness – lights lit, darkness – no lighting.
• Weather condition : fine no high winds, raining no high winds, snowing, etc.
• Road surface condition : dry, wet, snow, mud, etc.
• Number of vehicles
• Speed limit
Cont’d…
12
• Number of passenger
• Pedestrian location
• Age of driver
• Engine capacity
• Age of vehicle
(manufacture)
• Vehicle maneuver
• Pedestrian movement
• Breaking behavior
• Passenger in adjacent
seat
• Cellphone use
• Driver seatbelt
• Speeding
• Low speed
• Drowsiness
• Alcohol Drug
Impairment
• Travel lanes
• Traffic Density
• Traffic flow
• Traffic control
• Vehicle to vehicle
distance
• Fatigue
• Tire pressure
• Acceleration
• Deceleration
• Pedestrian crossing
• Pedestrian type
• Sex of driver
• Longitude and
latitude
Cont’d…
13
We categorized the stated features into four categories as shown in figure 5.
Figure 5: Road Accidents Causes
Cont’d…
14
Assume D = { 𝑑1, 𝑑2, 𝑑3,… 𝑑 𝑛} where D is the dataset with number of instances around 1 Lakh
X = {𝑥1, 𝑥2, 𝑥3,… 𝑥 𝑛} where X is the set of Features with the number of features are 50.
T = {𝑡1, 𝑡2, 𝑡3} where T is the set of Target Values i.e. {Slight, Fatal, Serious}
Propose Work Process
Step 1: Manual Selection of features which are related to the on-road condition
X = {𝑥1, 𝑥2, 𝑥3,… 𝑥 𝑛} where X is the set of Features with the number of features are 34.
Step 2: In this step, we also filter out those features which have some direct impact on road accidents.
X = {𝑥1, 𝑥2, 𝑥3,… 𝑥 𝑛} where X is the set of Features with the number of features are 18.
Cont’d…
15
Step 3: In this step, we label the instance with the target value by identifying the impact of the accident.
D: X → T {Slight, Fatal, Serious}
Step 4: Now we use some machine learning features selection techniques to find the important features and
reduce the curse of dimensionality.
we use Univariate Selection, Feature Importance, Recursive Feature Elimination.
Cont’d…
16
Step 4 (a): Univariate Features Selection
• Statistical tests can be used to select those features that have the strongest relationship with
the output variable.
• The scikit-learn library provides the SelectKBest class that can be used with a suite of
different statistical tests to select a specific number of features.
• We uses the chi-squared (chi²) statistical test for non-negative features to select 11 of the best
features from the Dataset.
• The Formula for Chi-squared test is
𝑋𝑐
𝑖
=
(𝑂 𝑖 − 𝐸 𝑖 )2
𝐸 𝑖
Where c = Degrees of freedom, O = Observed value(s) and E = Expected value(s)
Cont’d…
17
Step 4 (b): Feature Importance
• We can get the feature importance of each feature of the dataset by using the feature
importance property of the model.
• Feature importance gives you a score for each feature of the data, the higher the score more
important or relevant is the feature towards your output variable.
• Feature importance is an inbuilt class that comes with Tree Based Classifiers, we will be
using Extra Tree Classifier for extracting the top 11 features for the dataset.
• Scikit-learn calculates a nodes importance using Gini Importance, assuming only two child
nodes
𝑛𝑖𝑗 = 𝑤𝑗 𝐶𝑗 − 𝑤𝑙𝑒𝑓𝑡 𝑗 𝐶𝑙𝑒𝑓𝑡 𝑗 − 𝑤 𝑟𝑖𝑔ℎ𝑡 𝑗 𝐶 𝑟𝑖𝑔ℎ𝑡 𝑗
Cont’d…
18
Where
𝑛𝑖𝑗 = the importance of node j
𝑤𝑗 = 𝑤𝑒𝑖𝑔ℎ𝑡𝑒𝑑 𝑛𝑢𝑚𝑏𝑒𝑟 𝑜𝑓 𝑠𝑎𝑚𝑝𝑙𝑒𝑠 𝑟𝑒𝑎𝑐ℎ𝑖𝑛𝑔 𝑛𝑜𝑑𝑒 𝑗
𝐶𝑗 = the impurity value of node j
𝑙𝑒𝑓𝑡 𝑗 = child node from left split on node j
𝑟𝑖𝑔ℎ𝑡 𝑗 = child node from left split on node j
The importance for each feature on a decision tree is then calculated as:
𝑓𝑖 𝑗 =
𝑗:𝑛𝑜𝑑𝑒 𝑗 𝑠𝑝𝑙𝑖𝑡𝑠 𝑜𝑛 𝑓𝑒𝑎𝑡𝑢𝑟𝑒 𝑖 𝑛𝑖𝑗
𝑘∈𝑎𝑙𝑙 𝑛𝑜𝑑𝑒𝑠 𝑛𝑖 𝑘
𝑓𝑖 𝑗 = 𝑡ℎ𝑒 𝑖𝑚𝑝𝑜𝑟𝑡𝑎𝑛𝑐𝑒 𝑜𝑓 𝑓𝑒𝑎𝑡𝑢𝑟𝑒 𝑖
𝑛𝑖𝑗 = 𝑡ℎ𝑒 𝑖𝑚𝑝𝑜𝑟𝑡𝑎𝑛𝑐𝑒 𝑜𝑓 𝑛𝑜𝑑𝑒 𝑗
Cont’d…
19
These can then be normalized to a value between 0 and 1 by dividing by the sum of all feature importance values:
𝑛𝑜𝑟𝑚𝑓𝑖 𝑗 =
𝑓𝑖 𝑗
𝑗∈𝑎𝑙𝑙 𝑓𝑒𝑎𝑡𝑢𝑟𝑒𝑠 𝑓𝑖 𝑗
The final feature importance, at the Random Forest level, is it’s average over all the trees. The sum of the feature’s
importance value on each trees is calculated and divided by the total number of trees:
𝑅𝐹𝑓𝑖𝑖 =
𝑗∈𝑎𝑙𝑙 𝑡𝑟𝑒𝑒𝑠 𝑛𝑜𝑟𝑚𝑓𝑖𝑖𝑗
𝑇
𝑅𝐹𝑓𝑖𝑖 = 𝑡ℎ𝑒 𝑖𝑚𝑝𝑜𝑟𝑡𝑎𝑛𝑐𝑒 𝑜𝑓 𝑓𝑒𝑎𝑡𝑢𝑟𝑒 𝑖 𝑐𝑎𝑙𝑐𝑢𝑙𝑎𝑡𝑒𝑑 𝑓𝑟𝑜𝑚 𝑎𝑙𝑙 𝑡𝑟𝑒𝑒𝑠 𝑖𝑛 𝑡ℎ𝑒 𝑅𝑎𝑛𝑑𝑜𝑚 𝐹𝑜𝑟𝑒𝑠𝑡 𝑚𝑜𝑑𝑒𝑙
𝑛𝑜𝑟𝑚𝑓𝑖𝑖𝑗 = 𝑡ℎ𝑒 𝑛𝑜𝑟𝑚𝑎𝑙𝑖𝑧𝑒𝑑 𝑓𝑒𝑎𝑡𝑢𝑟𝑒 𝑖𝑚𝑝𝑜𝑟𝑡𝑎𝑛𝑐𝑒 𝑓𝑜𝑟 𝑖 𝑖𝑛 𝑡𝑟𝑒𝑒 𝑗
T = total number of trees
Cont’d…
20
Step 4 (c): Recursive Feature Elimination
Given an external estimator that assigns weights to features (e.g., the coefficients of a linear model), the goal of
ursive feature elimination (RFE) is to select features by recursively considering smaller and smaller sets of
features. First, the estimator is trained on the initial set of features and the importance of each feature is obtained
either through a coef_ attribute or through a feature_importances_ attribute. Then, the least important features are
pruned from current set of features. That procedure is recursively repeated on the pruned set until the desired
number of features to select is eventually reached.
Result
21
Step 5: Following model has been used and test:
A. Naïve Bayes
Accuracy = 79.8% Classification Error = 20.2%
Confusion
Matrix
True Slight True Serious True Fatal Class Prediction
Pred. Slight 27965 6605 468 79.81%
Pred. Serious 0 0 0 0.00%
Pred. Fatal 0 0 0 0.00%
Class recall 100.00% 0.00% 0.00%
Cont’d…
22
B. Logistic Regression
Accuracy = 79.5% Classification Error = 90.5%
Confusion
Matrix
True Slight True Serious True Fatal Class Prediction
Pred. Slight 2862 439 7 86.52%
Pred. Serious 0 0 0 0.00%
Pred. Fatal 25039 6220 472 1.49%
Class recall 10.26% 0.00% 98.54%
Cont’d…
23
C. Decision Tree
Accuracy = 79.8% Classification Error = 20.2%
Confusion
Matrix
True Slight True Serious True Fatal Class Prediction
Pred. Slight 27965 6605 468 79.81%
Pred. Serious 0 0 0 0.00%
Pred. Fatal 0 0 0 0.00%
Class recall 100.00% 0.00% 0.00%
Cont’d…
24
D. Random Forest
Accuracy = 80.0% Classification Error = 20.2%
Confusion
Matrix
True Slight True Serious True Fatal Class Prediction
Pred. Slight 14240 3373 235 79.78%
Pred. Serious 8 2 0 20.00%
Pred. Fatal 0 0 0 0.00%
Class recall 99.94% 0.06% 0.00%
Cont’d…
25
E. Support Vector Machine
Accuracy = 79.9% Classification Error = 20.6%
Confusion
Matrix
True Slight True Serious True Fatal Class Prediction
Pred. Slight 3404 822 59 79.44%
Pred. Serious 0 0 0 0.00%
Pred. Fatal 0 0 0 0.00%
Class recall 100.00% 0.00% 0.00%
Propose work flowchart 26
Input Raw Data
Top 11 Features
Remove null and noisy data
Input all road condition
related features
Remove the dependent
features
Perform features selection
techniques (Univariate
Features selection, Feature
Importance & recursive
feature elimination)
Perform the Classification
techniques (Naive Bayes,
Logistic Regression, Decision
Tree, Random Forest, Support
Vector Machine )
Select the Best Prediction
Method
Resultant Final Model
References
1. Global status report on road safety 2018. “https://www.who.int/violence_injury_prevention/road_safety_status/2018/en/”
(27/05/2020 04:37 PM).
2. Road accidents claimed over 1.5 lakh lives in 2018, over-speeding major killer - The Economic Times.
“https://economictimes.indiatimes.com/news/politics-and-nation/road-accidents-claimed-over-1-5-lakh-lives-in-2018-over-
speeding-major-killer/articleshow/72127418.cms?from=mdr” (27/05/2020 04:37 PM).
3. India way behind 2020 target, road accidents still kill over a lakh a year | India News - Times of India.
“https://timesofindia.indiatimes.com/india/india-way-off-road-safety-targets-for-2020-road-accidents-still-kill-over-a-lakh-a-
year/articleshow/65765549.cms” (27/05/2020 04:55 PM).
4. Government of India Ministry of Road Transport & Highways Transport Research Wing 2018.
5. Xiaoxia Xiong , Long Chen , and Jun Liang : Discrete Dynamics in Nature and Society Volume 2018, "Vehicle Driving Risk
Prediction Based on Markov Chain Model.“
6. Chunjiao Dong ,1,2,3 Chunfu Shao,1,2 Juan Li,1 and Zhihua Xiong : Journal of Advanced Transportation Volume 2018, "An
Improved Deep Learning Model for Traffic Crash Prediction.“
7. Nasim Arbabzadeh and Mohsen Jafari : IEEE TRANSACTIONS ON INTELLIGENT TRANSPORTATION SYSTEMS, “A
Data-Driven Approach for Driving Safety Risk Prediction Using Driver Behavior and Roadway Information Data”
8. Yutao Ba a,⇑, Wei Zhang b, Qinhua Wang a, Ronggang Zhou c, Changrui Ren a : Transportation Research Part C, “Crash
prediction with behavioral and physiological features for advanced vehicle collision avoidance system”
9. Yiping Chen1, Jingkang Wang2, Jonathan Li#1,3, Cewu Lu#2, Zhipeng Luo1, Han Xue2, and Cheng Wang1 : “LiDAR-
Video Driving Dataset: Learning Driving Policies Effectively”
10. Rishu Chhabra, Dr. Seema Verma and Dr. C. Rama Krishna : A survey on driver behavior detection techniques for intelligent
transportation systems.
11. Loukas Dimitrioua,⁎, Katerina Stylianoua, Mohamed A. Abdel-Atyb : Assessing rear-end crash potential in urban locations
based on vehicle-byvehicle interactions, geometric characteristics and operational conditions.
27
Published and Communicated
Papers
28
• B. Sikander and Anant Ram, “SURVEY ON SEVERITY RATE OF ROAD ACCIDENT
ASSESSMENT AND ESTIMATION USING DATA MINING TECHNIQUES”, TEST ENGINEERING
AND MANAGEMENT (ACCEPTED FOR PUBLICATION) –SCOPUS
• B. Sikander and Anant Ram, “Urban Road Accident Evaluation and Road
Accident Severity Prediction”, International Journal of Mathematical, Engineering
and Management Sciences (Scopus), - Communicated
29

More Related Content

What's hot

Disease Prediction by Machine Learning Over Big Data From Healthcare Communities
Disease Prediction by Machine Learning Over Big Data From Healthcare CommunitiesDisease Prediction by Machine Learning Over Big Data From Healthcare Communities
Disease Prediction by Machine Learning Over Big Data From Healthcare Communities
Khulna University of Engineering & Tecnology
 
A thesis presentation on pothole detection
A thesis presentation on pothole detectionA thesis presentation on pothole detection
A thesis presentation on pothole detection
PrimeAsia University
 
House Price Prediction Using Machine Learning Via Data Analysis
House Price Prediction Using Machine Learning Via Data AnalysisHouse Price Prediction Using Machine Learning Via Data Analysis
House Price Prediction Using Machine Learning Via Data Analysis
IRJET Journal
 
Unsupervised learning
Unsupervised learningUnsupervised learning
Unsupervised learning
amalalhait
 
Diabetes prediction with r(using knn)
Diabetes prediction with r(using knn)Diabetes prediction with r(using knn)
Diabetes prediction with r(using knn)
tanujoshi98
 
Machine Learning for Disease Prediction
Machine Learning for Disease PredictionMachine Learning for Disease Prediction
Machine Learning for Disease Prediction
Mustafa Oğuz
 
An Introduction to Supervised Machine Learning and Pattern Classification: Th...
An Introduction to Supervised Machine Learning and Pattern Classification: Th...An Introduction to Supervised Machine Learning and Pattern Classification: Th...
An Introduction to Supervised Machine Learning and Pattern Classification: Th...
Sebastian Raschka
 
Machine Learning With Logistic Regression
Machine Learning  With Logistic RegressionMachine Learning  With Logistic Regression
Machine Learning With Logistic Regression
Knoldus Inc.
 
Stock Market Prediction using Machine Learning
Stock Market Prediction using Machine LearningStock Market Prediction using Machine Learning
Stock Market Prediction using Machine Learning
Aravind Balaji
 
Emotion recognition
Emotion recognitionEmotion recognition
Emotion recognition
Madhusudhan G
 
Data analysis of weather forecasting
Data analysis of weather forecastingData analysis of weather forecasting
Data analysis of weather forecasting
Trupti Shingala, WAS, CPACC, CPWA, JAWS, CSM
 
Classification Based Machine Learning Algorithms
Classification Based Machine Learning AlgorithmsClassification Based Machine Learning Algorithms
Classification Based Machine Learning Algorithms
Md. Main Uddin Rony
 
Satellite Image Classification with Deep Learning Survey
Satellite Image Classification with Deep Learning SurveySatellite Image Classification with Deep Learning Survey
Satellite Image Classification with Deep Learning Survey
ijtsrd
 
Automatic number-plate-recognition
Automatic number-plate-recognitionAutomatic number-plate-recognition
Automatic number-plate-recognition
Devang Tailor
 
Artificial intelligence in transportation system
Artificial intelligence in transportation systemArtificial intelligence in transportation system
Artificial intelligence in transportation system
PoojaBele1
 
Drowsiness Detection using machine learning (1).pptx
Drowsiness Detection using machine learning (1).pptxDrowsiness Detection using machine learning (1).pptx
Drowsiness Detection using machine learning (1).pptx
sathiyasowmi
 
DATA MINING FRAMEWORK TO ANALYZE ROAD ACCIDENT DATA
DATA MINING FRAMEWORK TO ANALYZE ROAD ACCIDENT DATADATA MINING FRAMEWORK TO ANALYZE ROAD ACCIDENT DATA
DATA MINING FRAMEWORK TO ANALYZE ROAD ACCIDENT DATA
Aishwarya Saseendran
 
Dendral
DendralDendral
Dendral
gupta8741
 
Machine Learning In Insurance
Machine Learning In InsuranceMachine Learning In Insurance
Machine Learning In Insurance
Accenture Insurance
 

What's hot (20)

Disease Prediction by Machine Learning Over Big Data From Healthcare Communities
Disease Prediction by Machine Learning Over Big Data From Healthcare CommunitiesDisease Prediction by Machine Learning Over Big Data From Healthcare Communities
Disease Prediction by Machine Learning Over Big Data From Healthcare Communities
 
A thesis presentation on pothole detection
A thesis presentation on pothole detectionA thesis presentation on pothole detection
A thesis presentation on pothole detection
 
Random forest
Random forestRandom forest
Random forest
 
House Price Prediction Using Machine Learning Via Data Analysis
House Price Prediction Using Machine Learning Via Data AnalysisHouse Price Prediction Using Machine Learning Via Data Analysis
House Price Prediction Using Machine Learning Via Data Analysis
 
Unsupervised learning
Unsupervised learningUnsupervised learning
Unsupervised learning
 
Diabetes prediction with r(using knn)
Diabetes prediction with r(using knn)Diabetes prediction with r(using knn)
Diabetes prediction with r(using knn)
 
Machine Learning for Disease Prediction
Machine Learning for Disease PredictionMachine Learning for Disease Prediction
Machine Learning for Disease Prediction
 
An Introduction to Supervised Machine Learning and Pattern Classification: Th...
An Introduction to Supervised Machine Learning and Pattern Classification: Th...An Introduction to Supervised Machine Learning and Pattern Classification: Th...
An Introduction to Supervised Machine Learning and Pattern Classification: Th...
 
Machine Learning With Logistic Regression
Machine Learning  With Logistic RegressionMachine Learning  With Logistic Regression
Machine Learning With Logistic Regression
 
Stock Market Prediction using Machine Learning
Stock Market Prediction using Machine LearningStock Market Prediction using Machine Learning
Stock Market Prediction using Machine Learning
 
Emotion recognition
Emotion recognitionEmotion recognition
Emotion recognition
 
Data analysis of weather forecasting
Data analysis of weather forecastingData analysis of weather forecasting
Data analysis of weather forecasting
 
Classification Based Machine Learning Algorithms
Classification Based Machine Learning AlgorithmsClassification Based Machine Learning Algorithms
Classification Based Machine Learning Algorithms
 
Satellite Image Classification with Deep Learning Survey
Satellite Image Classification with Deep Learning SurveySatellite Image Classification with Deep Learning Survey
Satellite Image Classification with Deep Learning Survey
 
Automatic number-plate-recognition
Automatic number-plate-recognitionAutomatic number-plate-recognition
Automatic number-plate-recognition
 
Artificial intelligence in transportation system
Artificial intelligence in transportation systemArtificial intelligence in transportation system
Artificial intelligence in transportation system
 
Drowsiness Detection using machine learning (1).pptx
Drowsiness Detection using machine learning (1).pptxDrowsiness Detection using machine learning (1).pptx
Drowsiness Detection using machine learning (1).pptx
 
DATA MINING FRAMEWORK TO ANALYZE ROAD ACCIDENT DATA
DATA MINING FRAMEWORK TO ANALYZE ROAD ACCIDENT DATADATA MINING FRAMEWORK TO ANALYZE ROAD ACCIDENT DATA
DATA MINING FRAMEWORK TO ANALYZE ROAD ACCIDENT DATA
 
Dendral
DendralDendral
Dendral
 
Machine Learning In Insurance
Machine Learning In InsuranceMachine Learning In Insurance
Machine Learning In Insurance
 

Similar to An Approach For Predicting Road Accident Severity

thesisppt-200603142908.pdf
thesisppt-200603142908.pdfthesisppt-200603142908.pdf
thesisppt-200603142908.pdf
nebulaa2
 
AIRLINE FARE PRICE PREDICTION
AIRLINE FARE PRICE PREDICTIONAIRLINE FARE PRICE PREDICTION
AIRLINE FARE PRICE PREDICTION
IRJET Journal
 
Deep Learning: Introduction & Chapter 5 Machine Learning Basics
Deep Learning: Introduction & Chapter 5 Machine Learning BasicsDeep Learning: Introduction & Chapter 5 Machine Learning Basics
Deep Learning: Introduction & Chapter 5 Machine Learning Basics
Jason Tsai
 
Application of Machine Learning in Agriculture
Application of Machine  Learning in AgricultureApplication of Machine  Learning in Agriculture
Application of Machine Learning in Agriculture
Aman Vasisht
 
Data mining with weka
Data mining with wekaData mining with weka
Data mining with weka
Hein Min Htike
 
CSL0777-L07.pptx
CSL0777-L07.pptxCSL0777-L07.pptx
CSL0777-L07.pptx
KonkoboUlrichArthur
 
Modeling Road Traffic Accidents
Modeling Road Traffic AccidentsModeling Road Traffic Accidents
Modeling Road Traffic Accidents
Julius Nyerere
 
Machine Learning Notes for beginners ,Step by step
Machine Learning Notes for beginners ,Step by stepMachine Learning Notes for beginners ,Step by step
Machine Learning Notes for beginners ,Step by step
SanjanaSaxena17
 
Dimensionality Reduction
Dimensionality ReductionDimensionality Reduction
Dimensionality Reduction
Saad Elbeleidy
 
机器学习Adaboost
机器学习Adaboost机器学习Adaboost
机器学习Adaboost
Shocky1
 
Working with the data for Machine Learning
Working with the data for Machine LearningWorking with the data for Machine Learning
Working with the data for Machine Learning
Mehwish690898
 
CONSTRUCTING A FUZZY NETWORK INTRUSION CLASSIFIER BASED ON DIFFERENTIAL EVOLU...
CONSTRUCTING A FUZZY NETWORK INTRUSION CLASSIFIER BASED ON DIFFERENTIAL EVOLU...CONSTRUCTING A FUZZY NETWORK INTRUSION CLASSIFIER BASED ON DIFFERENTIAL EVOLU...
CONSTRUCTING A FUZZY NETWORK INTRUSION CLASSIFIER BASED ON DIFFERENTIAL EVOLU...
IJCNCJournal
 
2014-mo444-final-project
2014-mo444-final-project2014-mo444-final-project
2014-mo444-final-projectPaulo Faria
 
Scalable machine learning
Scalable machine learningScalable machine learning
Scalable machine learning
Tien-Yang (Aiden) Wu
 
Class 2 variables, classes methods...
Class 2   variables, classes methods...Class 2   variables, classes methods...
Class 2 variables, classes methods...
Fernando Loizides
 
BS LAB Manual (1).pdf
BS LAB Manual  (1).pdfBS LAB Manual  (1).pdf
BS LAB Manual (1).pdf
ssuser476810
 
Application Of The Three-In-One Control Platform Based On OPC In The Lifting-...
Application Of The Three-In-One Control Platform Based On OPC In The Lifting-...Application Of The Three-In-One Control Platform Based On OPC In The Lifting-...
Application Of The Three-In-One Control Platform Based On OPC In The Lifting-...
IJRES Journal
 
Real Estate Investment Advising Using Machine Learning
Real Estate Investment Advising Using Machine LearningReal Estate Investment Advising Using Machine Learning
Real Estate Investment Advising Using Machine Learning
IRJET Journal
 
2014-mo444-practical-assignment-04-paulo_faria
2014-mo444-practical-assignment-04-paulo_faria2014-mo444-practical-assignment-04-paulo_faria
2014-mo444-practical-assignment-04-paulo_fariaPaulo Faria
 
AINL 2016: Strijov
AINL 2016: StrijovAINL 2016: Strijov
AINL 2016: Strijov
Lidia Pivovarova
 

Similar to An Approach For Predicting Road Accident Severity (20)

thesisppt-200603142908.pdf
thesisppt-200603142908.pdfthesisppt-200603142908.pdf
thesisppt-200603142908.pdf
 
AIRLINE FARE PRICE PREDICTION
AIRLINE FARE PRICE PREDICTIONAIRLINE FARE PRICE PREDICTION
AIRLINE FARE PRICE PREDICTION
 
Deep Learning: Introduction & Chapter 5 Machine Learning Basics
Deep Learning: Introduction & Chapter 5 Machine Learning BasicsDeep Learning: Introduction & Chapter 5 Machine Learning Basics
Deep Learning: Introduction & Chapter 5 Machine Learning Basics
 
Application of Machine Learning in Agriculture
Application of Machine  Learning in AgricultureApplication of Machine  Learning in Agriculture
Application of Machine Learning in Agriculture
 
Data mining with weka
Data mining with wekaData mining with weka
Data mining with weka
 
CSL0777-L07.pptx
CSL0777-L07.pptxCSL0777-L07.pptx
CSL0777-L07.pptx
 
Modeling Road Traffic Accidents
Modeling Road Traffic AccidentsModeling Road Traffic Accidents
Modeling Road Traffic Accidents
 
Machine Learning Notes for beginners ,Step by step
Machine Learning Notes for beginners ,Step by stepMachine Learning Notes for beginners ,Step by step
Machine Learning Notes for beginners ,Step by step
 
Dimensionality Reduction
Dimensionality ReductionDimensionality Reduction
Dimensionality Reduction
 
机器学习Adaboost
机器学习Adaboost机器学习Adaboost
机器学习Adaboost
 
Working with the data for Machine Learning
Working with the data for Machine LearningWorking with the data for Machine Learning
Working with the data for Machine Learning
 
CONSTRUCTING A FUZZY NETWORK INTRUSION CLASSIFIER BASED ON DIFFERENTIAL EVOLU...
CONSTRUCTING A FUZZY NETWORK INTRUSION CLASSIFIER BASED ON DIFFERENTIAL EVOLU...CONSTRUCTING A FUZZY NETWORK INTRUSION CLASSIFIER BASED ON DIFFERENTIAL EVOLU...
CONSTRUCTING A FUZZY NETWORK INTRUSION CLASSIFIER BASED ON DIFFERENTIAL EVOLU...
 
2014-mo444-final-project
2014-mo444-final-project2014-mo444-final-project
2014-mo444-final-project
 
Scalable machine learning
Scalable machine learningScalable machine learning
Scalable machine learning
 
Class 2 variables, classes methods...
Class 2   variables, classes methods...Class 2   variables, classes methods...
Class 2 variables, classes methods...
 
BS LAB Manual (1).pdf
BS LAB Manual  (1).pdfBS LAB Manual  (1).pdf
BS LAB Manual (1).pdf
 
Application Of The Three-In-One Control Platform Based On OPC In The Lifting-...
Application Of The Three-In-One Control Platform Based On OPC In The Lifting-...Application Of The Three-In-One Control Platform Based On OPC In The Lifting-...
Application Of The Three-In-One Control Platform Based On OPC In The Lifting-...
 
Real Estate Investment Advising Using Machine Learning
Real Estate Investment Advising Using Machine LearningReal Estate Investment Advising Using Machine Learning
Real Estate Investment Advising Using Machine Learning
 
2014-mo444-practical-assignment-04-paulo_faria
2014-mo444-practical-assignment-04-paulo_faria2014-mo444-practical-assignment-04-paulo_faria
2014-mo444-practical-assignment-04-paulo_faria
 
AINL 2016: Strijov
AINL 2016: StrijovAINL 2016: Strijov
AINL 2016: Strijov
 

Recently uploaded

Gen AI Study Jams _ For the GDSC Leads in India.pdf
Gen AI Study Jams _ For the GDSC Leads in India.pdfGen AI Study Jams _ For the GDSC Leads in India.pdf
Gen AI Study Jams _ For the GDSC Leads in India.pdf
gdsczhcet
 
Heap Sort (SS).ppt FOR ENGINEERING GRADUATES, BCA, MCA, MTECH, BSC STUDENTS
Heap Sort (SS).ppt FOR ENGINEERING GRADUATES, BCA, MCA, MTECH, BSC STUDENTSHeap Sort (SS).ppt FOR ENGINEERING GRADUATES, BCA, MCA, MTECH, BSC STUDENTS
Heap Sort (SS).ppt FOR ENGINEERING GRADUATES, BCA, MCA, MTECH, BSC STUDENTS
Soumen Santra
 
CME397 Surface Engineering- Professional Elective
CME397 Surface Engineering- Professional ElectiveCME397 Surface Engineering- Professional Elective
CME397 Surface Engineering- Professional Elective
karthi keyan
 
Industrial Training at Shahjalal Fertilizer Company Limited (SFCL)
Industrial Training at Shahjalal Fertilizer Company Limited (SFCL)Industrial Training at Shahjalal Fertilizer Company Limited (SFCL)
Industrial Training at Shahjalal Fertilizer Company Limited (SFCL)
MdTanvirMahtab2
 
Student information management system project report ii.pdf
Student information management system project report ii.pdfStudent information management system project report ii.pdf
Student information management system project report ii.pdf
Kamal Acharya
 
Forklift Classes Overview by Intella Parts
Forklift Classes Overview by Intella PartsForklift Classes Overview by Intella Parts
Forklift Classes Overview by Intella Parts
Intella Parts
 
block diagram and signal flow graph representation
block diagram and signal flow graph representationblock diagram and signal flow graph representation
block diagram and signal flow graph representation
Divya Somashekar
 
Pile Foundation by Venkatesh Taduvai (Sub Geotechnical Engineering II)-conver...
Pile Foundation by Venkatesh Taduvai (Sub Geotechnical Engineering II)-conver...Pile Foundation by Venkatesh Taduvai (Sub Geotechnical Engineering II)-conver...
Pile Foundation by Venkatesh Taduvai (Sub Geotechnical Engineering II)-conver...
AJAYKUMARPUND1
 
AP LAB PPT.pdf ap lab ppt no title specific
AP LAB PPT.pdf ap lab ppt no title specificAP LAB PPT.pdf ap lab ppt no title specific
AP LAB PPT.pdf ap lab ppt no title specific
BrazilAccount1
 
Sachpazis:Terzaghi Bearing Capacity Estimation in simple terms with Calculati...
Sachpazis:Terzaghi Bearing Capacity Estimation in simple terms with Calculati...Sachpazis:Terzaghi Bearing Capacity Estimation in simple terms with Calculati...
Sachpazis:Terzaghi Bearing Capacity Estimation in simple terms with Calculati...
Dr.Costas Sachpazis
 
weather web application report.pdf
weather web application report.pdfweather web application report.pdf
weather web application report.pdf
Pratik Pawar
 
一比一原版(SFU毕业证)西蒙菲莎大学毕业证成绩单如何办理
一比一原版(SFU毕业证)西蒙菲莎大学毕业证成绩单如何办理一比一原版(SFU毕业证)西蒙菲莎大学毕业证成绩单如何办理
一比一原版(SFU毕业证)西蒙菲莎大学毕业证成绩单如何办理
bakpo1
 
DfMAy 2024 - key insights and contributions
DfMAy 2024 - key insights and contributionsDfMAy 2024 - key insights and contributions
DfMAy 2024 - key insights and contributions
gestioneergodomus
 
Basic Industrial Engineering terms for apparel
Basic Industrial Engineering terms for apparelBasic Industrial Engineering terms for apparel
Basic Industrial Engineering terms for apparel
top1002
 
Cosmetic shop management system project report.pdf
Cosmetic shop management system project report.pdfCosmetic shop management system project report.pdf
Cosmetic shop management system project report.pdf
Kamal Acharya
 
一比一原版(UofT毕业证)多伦多大学毕业证成绩单如何办理
一比一原版(UofT毕业证)多伦多大学毕业证成绩单如何办理一比一原版(UofT毕业证)多伦多大学毕业证成绩单如何办理
一比一原版(UofT毕业证)多伦多大学毕业证成绩单如何办理
ydteq
 
NUMERICAL SIMULATIONS OF HEAT AND MASS TRANSFER IN CONDENSING HEAT EXCHANGERS...
NUMERICAL SIMULATIONS OF HEAT AND MASS TRANSFER IN CONDENSING HEAT EXCHANGERS...NUMERICAL SIMULATIONS OF HEAT AND MASS TRANSFER IN CONDENSING HEAT EXCHANGERS...
NUMERICAL SIMULATIONS OF HEAT AND MASS TRANSFER IN CONDENSING HEAT EXCHANGERS...
ssuser7dcef0
 
road safety engineering r s e unit 3.pdf
road safety engineering  r s e unit 3.pdfroad safety engineering  r s e unit 3.pdf
road safety engineering r s e unit 3.pdf
VENKATESHvenky89705
 
在线办理(ANU毕业证书)澳洲国立大学毕业证录取通知书一模一样
在线办理(ANU毕业证书)澳洲国立大学毕业证录取通知书一模一样在线办理(ANU毕业证书)澳洲国立大学毕业证录取通知书一模一样
在线办理(ANU毕业证书)澳洲国立大学毕业证录取通知书一模一样
obonagu
 
Planning Of Procurement o different goods and services
Planning Of Procurement o different goods and servicesPlanning Of Procurement o different goods and services
Planning Of Procurement o different goods and services
JoytuBarua2
 

Recently uploaded (20)

Gen AI Study Jams _ For the GDSC Leads in India.pdf
Gen AI Study Jams _ For the GDSC Leads in India.pdfGen AI Study Jams _ For the GDSC Leads in India.pdf
Gen AI Study Jams _ For the GDSC Leads in India.pdf
 
Heap Sort (SS).ppt FOR ENGINEERING GRADUATES, BCA, MCA, MTECH, BSC STUDENTS
Heap Sort (SS).ppt FOR ENGINEERING GRADUATES, BCA, MCA, MTECH, BSC STUDENTSHeap Sort (SS).ppt FOR ENGINEERING GRADUATES, BCA, MCA, MTECH, BSC STUDENTS
Heap Sort (SS).ppt FOR ENGINEERING GRADUATES, BCA, MCA, MTECH, BSC STUDENTS
 
CME397 Surface Engineering- Professional Elective
CME397 Surface Engineering- Professional ElectiveCME397 Surface Engineering- Professional Elective
CME397 Surface Engineering- Professional Elective
 
Industrial Training at Shahjalal Fertilizer Company Limited (SFCL)
Industrial Training at Shahjalal Fertilizer Company Limited (SFCL)Industrial Training at Shahjalal Fertilizer Company Limited (SFCL)
Industrial Training at Shahjalal Fertilizer Company Limited (SFCL)
 
Student information management system project report ii.pdf
Student information management system project report ii.pdfStudent information management system project report ii.pdf
Student information management system project report ii.pdf
 
Forklift Classes Overview by Intella Parts
Forklift Classes Overview by Intella PartsForklift Classes Overview by Intella Parts
Forklift Classes Overview by Intella Parts
 
block diagram and signal flow graph representation
block diagram and signal flow graph representationblock diagram and signal flow graph representation
block diagram and signal flow graph representation
 
Pile Foundation by Venkatesh Taduvai (Sub Geotechnical Engineering II)-conver...
Pile Foundation by Venkatesh Taduvai (Sub Geotechnical Engineering II)-conver...Pile Foundation by Venkatesh Taduvai (Sub Geotechnical Engineering II)-conver...
Pile Foundation by Venkatesh Taduvai (Sub Geotechnical Engineering II)-conver...
 
AP LAB PPT.pdf ap lab ppt no title specific
AP LAB PPT.pdf ap lab ppt no title specificAP LAB PPT.pdf ap lab ppt no title specific
AP LAB PPT.pdf ap lab ppt no title specific
 
Sachpazis:Terzaghi Bearing Capacity Estimation in simple terms with Calculati...
Sachpazis:Terzaghi Bearing Capacity Estimation in simple terms with Calculati...Sachpazis:Terzaghi Bearing Capacity Estimation in simple terms with Calculati...
Sachpazis:Terzaghi Bearing Capacity Estimation in simple terms with Calculati...
 
weather web application report.pdf
weather web application report.pdfweather web application report.pdf
weather web application report.pdf
 
一比一原版(SFU毕业证)西蒙菲莎大学毕业证成绩单如何办理
一比一原版(SFU毕业证)西蒙菲莎大学毕业证成绩单如何办理一比一原版(SFU毕业证)西蒙菲莎大学毕业证成绩单如何办理
一比一原版(SFU毕业证)西蒙菲莎大学毕业证成绩单如何办理
 
DfMAy 2024 - key insights and contributions
DfMAy 2024 - key insights and contributionsDfMAy 2024 - key insights and contributions
DfMAy 2024 - key insights and contributions
 
Basic Industrial Engineering terms for apparel
Basic Industrial Engineering terms for apparelBasic Industrial Engineering terms for apparel
Basic Industrial Engineering terms for apparel
 
Cosmetic shop management system project report.pdf
Cosmetic shop management system project report.pdfCosmetic shop management system project report.pdf
Cosmetic shop management system project report.pdf
 
一比一原版(UofT毕业证)多伦多大学毕业证成绩单如何办理
一比一原版(UofT毕业证)多伦多大学毕业证成绩单如何办理一比一原版(UofT毕业证)多伦多大学毕业证成绩单如何办理
一比一原版(UofT毕业证)多伦多大学毕业证成绩单如何办理
 
NUMERICAL SIMULATIONS OF HEAT AND MASS TRANSFER IN CONDENSING HEAT EXCHANGERS...
NUMERICAL SIMULATIONS OF HEAT AND MASS TRANSFER IN CONDENSING HEAT EXCHANGERS...NUMERICAL SIMULATIONS OF HEAT AND MASS TRANSFER IN CONDENSING HEAT EXCHANGERS...
NUMERICAL SIMULATIONS OF HEAT AND MASS TRANSFER IN CONDENSING HEAT EXCHANGERS...
 
road safety engineering r s e unit 3.pdf
road safety engineering  r s e unit 3.pdfroad safety engineering  r s e unit 3.pdf
road safety engineering r s e unit 3.pdf
 
在线办理(ANU毕业证书)澳洲国立大学毕业证录取通知书一模一样
在线办理(ANU毕业证书)澳洲国立大学毕业证录取通知书一模一样在线办理(ANU毕业证书)澳洲国立大学毕业证录取通知书一模一样
在线办理(ANU毕业证书)澳洲国立大学毕业证录取通知书一模一样
 
Planning Of Procurement o different goods and services
Planning Of Procurement o different goods and servicesPlanning Of Procurement o different goods and services
Planning Of Procurement o different goods and services
 

An Approach For Predicting Road Accident Severity

  • 1. An Approach For Predicting Road Accident Severity B. Sikander MTech Scholar 188150002 Under the Supervision of: Dr. Anant Ram Dept. of Computer Engineering & Applications
  • 2. CONTENT Area of Research Introduction Motivation Applications Issues and Challenges Objective Proposed Approach Result References
  • 3. Area of Research Accident Analysis and Prevention 3
  • 4. Introduction • Road traffic injuries and deaths have been a major public health issue globally. According to World Health Organization (WHO), approximately 1.35 million people die from roadway traffic accidents each year, while 20∼50 million people suffer nonfatal injuries with many resulting in disabilities [1]. • Road accidents in India claimed over 1.5 lakh lives in the country in 2018, with over-speeding of vehicles being the biggest reason for casualties[2]. • The ministry of road transport and highways issued a report on Road accidents in India in 2018, which showed that road accidents last year increased by 0.46% as compared to 2017[2]. 4
  • 5. Cont’d… 5 Figure 1: Road Accidents Statistics from 2007 to 2017 [3]
  • 6. Motivation 6 • Increased highway accidents and rise in death toll every day. • A total of 4,67,044 road accidents have been reported by States and Union Territories (UTs) in the calendar year 2018, claiming 1,51,417 lives and causing injuries to 4,69,418 persons,” the report said. Over-speeding accounted for 64.4% of the persons killed[2]. Figure 2. Car Accident due to over speeding[2]
  • 7. Cont’d… 7 Figure 3: Number of Road Accidents, Deaths and Injuries [4]
  • 8. Applications 8 Help in designing road geometry Help in finding the those spot on road where chance of accident is high. Help in installing the road surveillance Road network simulation Help in study of Road Intersection
  • 9. Issues and Challenges 9 Proper selection of features which can cause a road accident. Removal of unwanted features. Collected data should be valid and related to real life accident situations. Removal of outlier instances
  • 10. Objective 10 To evaluate factors contributing to road accidents. To review related systems and models for predicting the likelihood of causing an accident. To develop an algorithm for predicting the likelihood of causing an accident. To test and validate the developed algorithm.
  • 11. Proposed Approach 11 • In this research we build a model which able to predict the road accident based on road conditions. • First we consider 50 possible features/variables which can cause road accidents. • Vehicle type : car, truck, motorcycle. • Vehicle length : near by vehicle length. • Road Type : One way street, Single carriageway, dual carriageway. • Light condition : daylight, darkness – lights lit, darkness – no lighting. • Weather condition : fine no high winds, raining no high winds, snowing, etc. • Road surface condition : dry, wet, snow, mud, etc. • Number of vehicles • Speed limit
  • 12. Cont’d… 12 • Number of passenger • Pedestrian location • Age of driver • Engine capacity • Age of vehicle (manufacture) • Vehicle maneuver • Pedestrian movement • Breaking behavior • Passenger in adjacent seat • Cellphone use • Driver seatbelt • Speeding • Low speed • Drowsiness • Alcohol Drug Impairment • Travel lanes • Traffic Density • Traffic flow • Traffic control • Vehicle to vehicle distance • Fatigue • Tire pressure • Acceleration • Deceleration • Pedestrian crossing • Pedestrian type • Sex of driver • Longitude and latitude
  • 13. Cont’d… 13 We categorized the stated features into four categories as shown in figure 5. Figure 5: Road Accidents Causes
  • 14. Cont’d… 14 Assume D = { 𝑑1, 𝑑2, 𝑑3,… 𝑑 𝑛} where D is the dataset with number of instances around 1 Lakh X = {𝑥1, 𝑥2, 𝑥3,… 𝑥 𝑛} where X is the set of Features with the number of features are 50. T = {𝑡1, 𝑡2, 𝑡3} where T is the set of Target Values i.e. {Slight, Fatal, Serious} Propose Work Process Step 1: Manual Selection of features which are related to the on-road condition X = {𝑥1, 𝑥2, 𝑥3,… 𝑥 𝑛} where X is the set of Features with the number of features are 34. Step 2: In this step, we also filter out those features which have some direct impact on road accidents. X = {𝑥1, 𝑥2, 𝑥3,… 𝑥 𝑛} where X is the set of Features with the number of features are 18.
  • 15. Cont’d… 15 Step 3: In this step, we label the instance with the target value by identifying the impact of the accident. D: X → T {Slight, Fatal, Serious} Step 4: Now we use some machine learning features selection techniques to find the important features and reduce the curse of dimensionality. we use Univariate Selection, Feature Importance, Recursive Feature Elimination.
  • 16. Cont’d… 16 Step 4 (a): Univariate Features Selection • Statistical tests can be used to select those features that have the strongest relationship with the output variable. • The scikit-learn library provides the SelectKBest class that can be used with a suite of different statistical tests to select a specific number of features. • We uses the chi-squared (chi²) statistical test for non-negative features to select 11 of the best features from the Dataset. • The Formula for Chi-squared test is 𝑋𝑐 𝑖 = (𝑂 𝑖 − 𝐸 𝑖 )2 𝐸 𝑖 Where c = Degrees of freedom, O = Observed value(s) and E = Expected value(s)
  • 17. Cont’d… 17 Step 4 (b): Feature Importance • We can get the feature importance of each feature of the dataset by using the feature importance property of the model. • Feature importance gives you a score for each feature of the data, the higher the score more important or relevant is the feature towards your output variable. • Feature importance is an inbuilt class that comes with Tree Based Classifiers, we will be using Extra Tree Classifier for extracting the top 11 features for the dataset. • Scikit-learn calculates a nodes importance using Gini Importance, assuming only two child nodes 𝑛𝑖𝑗 = 𝑤𝑗 𝐶𝑗 − 𝑤𝑙𝑒𝑓𝑡 𝑗 𝐶𝑙𝑒𝑓𝑡 𝑗 − 𝑤 𝑟𝑖𝑔ℎ𝑡 𝑗 𝐶 𝑟𝑖𝑔ℎ𝑡 𝑗
  • 18. Cont’d… 18 Where 𝑛𝑖𝑗 = the importance of node j 𝑤𝑗 = 𝑤𝑒𝑖𝑔ℎ𝑡𝑒𝑑 𝑛𝑢𝑚𝑏𝑒𝑟 𝑜𝑓 𝑠𝑎𝑚𝑝𝑙𝑒𝑠 𝑟𝑒𝑎𝑐ℎ𝑖𝑛𝑔 𝑛𝑜𝑑𝑒 𝑗 𝐶𝑗 = the impurity value of node j 𝑙𝑒𝑓𝑡 𝑗 = child node from left split on node j 𝑟𝑖𝑔ℎ𝑡 𝑗 = child node from left split on node j The importance for each feature on a decision tree is then calculated as: 𝑓𝑖 𝑗 = 𝑗:𝑛𝑜𝑑𝑒 𝑗 𝑠𝑝𝑙𝑖𝑡𝑠 𝑜𝑛 𝑓𝑒𝑎𝑡𝑢𝑟𝑒 𝑖 𝑛𝑖𝑗 𝑘∈𝑎𝑙𝑙 𝑛𝑜𝑑𝑒𝑠 𝑛𝑖 𝑘 𝑓𝑖 𝑗 = 𝑡ℎ𝑒 𝑖𝑚𝑝𝑜𝑟𝑡𝑎𝑛𝑐𝑒 𝑜𝑓 𝑓𝑒𝑎𝑡𝑢𝑟𝑒 𝑖 𝑛𝑖𝑗 = 𝑡ℎ𝑒 𝑖𝑚𝑝𝑜𝑟𝑡𝑎𝑛𝑐𝑒 𝑜𝑓 𝑛𝑜𝑑𝑒 𝑗
  • 19. Cont’d… 19 These can then be normalized to a value between 0 and 1 by dividing by the sum of all feature importance values: 𝑛𝑜𝑟𝑚𝑓𝑖 𝑗 = 𝑓𝑖 𝑗 𝑗∈𝑎𝑙𝑙 𝑓𝑒𝑎𝑡𝑢𝑟𝑒𝑠 𝑓𝑖 𝑗 The final feature importance, at the Random Forest level, is it’s average over all the trees. The sum of the feature’s importance value on each trees is calculated and divided by the total number of trees: 𝑅𝐹𝑓𝑖𝑖 = 𝑗∈𝑎𝑙𝑙 𝑡𝑟𝑒𝑒𝑠 𝑛𝑜𝑟𝑚𝑓𝑖𝑖𝑗 𝑇 𝑅𝐹𝑓𝑖𝑖 = 𝑡ℎ𝑒 𝑖𝑚𝑝𝑜𝑟𝑡𝑎𝑛𝑐𝑒 𝑜𝑓 𝑓𝑒𝑎𝑡𝑢𝑟𝑒 𝑖 𝑐𝑎𝑙𝑐𝑢𝑙𝑎𝑡𝑒𝑑 𝑓𝑟𝑜𝑚 𝑎𝑙𝑙 𝑡𝑟𝑒𝑒𝑠 𝑖𝑛 𝑡ℎ𝑒 𝑅𝑎𝑛𝑑𝑜𝑚 𝐹𝑜𝑟𝑒𝑠𝑡 𝑚𝑜𝑑𝑒𝑙 𝑛𝑜𝑟𝑚𝑓𝑖𝑖𝑗 = 𝑡ℎ𝑒 𝑛𝑜𝑟𝑚𝑎𝑙𝑖𝑧𝑒𝑑 𝑓𝑒𝑎𝑡𝑢𝑟𝑒 𝑖𝑚𝑝𝑜𝑟𝑡𝑎𝑛𝑐𝑒 𝑓𝑜𝑟 𝑖 𝑖𝑛 𝑡𝑟𝑒𝑒 𝑗 T = total number of trees
  • 20. Cont’d… 20 Step 4 (c): Recursive Feature Elimination Given an external estimator that assigns weights to features (e.g., the coefficients of a linear model), the goal of ursive feature elimination (RFE) is to select features by recursively considering smaller and smaller sets of features. First, the estimator is trained on the initial set of features and the importance of each feature is obtained either through a coef_ attribute or through a feature_importances_ attribute. Then, the least important features are pruned from current set of features. That procedure is recursively repeated on the pruned set until the desired number of features to select is eventually reached.
  • 21. Result 21 Step 5: Following model has been used and test: A. Naïve Bayes Accuracy = 79.8% Classification Error = 20.2% Confusion Matrix True Slight True Serious True Fatal Class Prediction Pred. Slight 27965 6605 468 79.81% Pred. Serious 0 0 0 0.00% Pred. Fatal 0 0 0 0.00% Class recall 100.00% 0.00% 0.00%
  • 22. Cont’d… 22 B. Logistic Regression Accuracy = 79.5% Classification Error = 90.5% Confusion Matrix True Slight True Serious True Fatal Class Prediction Pred. Slight 2862 439 7 86.52% Pred. Serious 0 0 0 0.00% Pred. Fatal 25039 6220 472 1.49% Class recall 10.26% 0.00% 98.54%
  • 23. Cont’d… 23 C. Decision Tree Accuracy = 79.8% Classification Error = 20.2% Confusion Matrix True Slight True Serious True Fatal Class Prediction Pred. Slight 27965 6605 468 79.81% Pred. Serious 0 0 0 0.00% Pred. Fatal 0 0 0 0.00% Class recall 100.00% 0.00% 0.00%
  • 24. Cont’d… 24 D. Random Forest Accuracy = 80.0% Classification Error = 20.2% Confusion Matrix True Slight True Serious True Fatal Class Prediction Pred. Slight 14240 3373 235 79.78% Pred. Serious 8 2 0 20.00% Pred. Fatal 0 0 0 0.00% Class recall 99.94% 0.06% 0.00%
  • 25. Cont’d… 25 E. Support Vector Machine Accuracy = 79.9% Classification Error = 20.6% Confusion Matrix True Slight True Serious True Fatal Class Prediction Pred. Slight 3404 822 59 79.44% Pred. Serious 0 0 0 0.00% Pred. Fatal 0 0 0 0.00% Class recall 100.00% 0.00% 0.00%
  • 26. Propose work flowchart 26 Input Raw Data Top 11 Features Remove null and noisy data Input all road condition related features Remove the dependent features Perform features selection techniques (Univariate Features selection, Feature Importance & recursive feature elimination) Perform the Classification techniques (Naive Bayes, Logistic Regression, Decision Tree, Random Forest, Support Vector Machine ) Select the Best Prediction Method Resultant Final Model
  • 27. References 1. Global status report on road safety 2018. “https://www.who.int/violence_injury_prevention/road_safety_status/2018/en/” (27/05/2020 04:37 PM). 2. Road accidents claimed over 1.5 lakh lives in 2018, over-speeding major killer - The Economic Times. “https://economictimes.indiatimes.com/news/politics-and-nation/road-accidents-claimed-over-1-5-lakh-lives-in-2018-over- speeding-major-killer/articleshow/72127418.cms?from=mdr” (27/05/2020 04:37 PM). 3. India way behind 2020 target, road accidents still kill over a lakh a year | India News - Times of India. “https://timesofindia.indiatimes.com/india/india-way-off-road-safety-targets-for-2020-road-accidents-still-kill-over-a-lakh-a- year/articleshow/65765549.cms” (27/05/2020 04:55 PM). 4. Government of India Ministry of Road Transport & Highways Transport Research Wing 2018. 5. Xiaoxia Xiong , Long Chen , and Jun Liang : Discrete Dynamics in Nature and Society Volume 2018, "Vehicle Driving Risk Prediction Based on Markov Chain Model.“ 6. Chunjiao Dong ,1,2,3 Chunfu Shao,1,2 Juan Li,1 and Zhihua Xiong : Journal of Advanced Transportation Volume 2018, "An Improved Deep Learning Model for Traffic Crash Prediction.“ 7. Nasim Arbabzadeh and Mohsen Jafari : IEEE TRANSACTIONS ON INTELLIGENT TRANSPORTATION SYSTEMS, “A Data-Driven Approach for Driving Safety Risk Prediction Using Driver Behavior and Roadway Information Data” 8. Yutao Ba a,⇑, Wei Zhang b, Qinhua Wang a, Ronggang Zhou c, Changrui Ren a : Transportation Research Part C, “Crash prediction with behavioral and physiological features for advanced vehicle collision avoidance system” 9. Yiping Chen1, Jingkang Wang2, Jonathan Li#1,3, Cewu Lu#2, Zhipeng Luo1, Han Xue2, and Cheng Wang1 : “LiDAR- Video Driving Dataset: Learning Driving Policies Effectively” 10. Rishu Chhabra, Dr. Seema Verma and Dr. C. Rama Krishna : A survey on driver behavior detection techniques for intelligent transportation systems. 11. Loukas Dimitrioua,⁎, Katerina Stylianoua, Mohamed A. Abdel-Atyb : Assessing rear-end crash potential in urban locations based on vehicle-byvehicle interactions, geometric characteristics and operational conditions. 27
  • 28. Published and Communicated Papers 28 • B. Sikander and Anant Ram, “SURVEY ON SEVERITY RATE OF ROAD ACCIDENT ASSESSMENT AND ESTIMATION USING DATA MINING TECHNIQUES”, TEST ENGINEERING AND MANAGEMENT (ACCEPTED FOR PUBLICATION) –SCOPUS • B. Sikander and Anant Ram, “Urban Road Accident Evaluation and Road Accident Severity Prediction”, International Journal of Mathematical, Engineering and Management Sciences (Scopus), - Communicated
  • 29. 29