In For this research we will use Apriori Algorithm

In 11 author study the
relation between number of injuries and accident distance in this researcher
author conclude from the findings that when the distance rate of the
designation is increasing death and accident rate of such type of travelling
are more. Author concludes this result by using regression data mining
techniques on the data set.The
main purpose of this research is to find the causes of the road accidents and
for this reason we will use data mining techniques, data mining is mainly used
to find the meaning full information from the vast information area similarly
in this research we will apply data mining techniques on the information and
data set to figure out the facts by analyzing the data to predict the results. Data
mining has so many techniques for our research we will use association and
classification rule in order to discover patterns. By using these techniques,
we will find factors which are involved and cause of the accident like weather,
speed limit, road conditions.Association Rule Association
rule is mainly analysis the previous data and predicting future in this
research this rule will implement to get the pattern for the future prediction.
Association rule mainly has two measures which are ·        
Support
and ·        
Confidence
·        
For
this research we will use Apriori Algorithm to analyses previous data.Steps:
o   The first step of this rule is to
scan the data and then set the support of each item o   Generate Candidate ko   Scan candidate k and generate item
set o   Add item set in the frequent item
set till It will become null o   Each frequent item generates non
empty subset

o   Non empty subset -> confidenceThis
research is mainly for the Sindh and this report is helpful for finding facts
and providing the factors like no of injuries in Sindh by road accident in each
year, no of killed peoples, no of vehicles involved, type of accidents (fatal,
non-fatal)

We Will Write a Custom Essay Specifically
For You For Only $13.90/page!


order now

 

TOOLS

Weka Toolkit

WEKA
stands for Waikato Environment for Knowledge Analysis. Widely used for machine
learning and data mining techniques, developed by the university of Waikato in
New Zealand. This tool is written in java, this tool is mainly used for
clustering, classification, association this is an open source software

mlpy machine learning Python

This
algorithm is mainly for the regression and classification. Various algorithms
are also obtainable which include error evaluation, peak finding algorithm