System for analysis of Road Accidents

Road accident is one amongst the crucial areas in India. The information to be analyzed is collected from various sources and has various aspects. It's a challenge to collect all such relevant data, detect and analyze it together to get decision trees that give information about previous accidents. One amongst the most objectives in accident data analysis is to spot the most factors of road accidents. For this matter, there's a necessity of machine learning to research the speed of accident. The results of this project are utilized to place some accident prevention efforts within the areas identified for a various variety of accidents to beat the number of accidents.

Looking to build projects on Analytics?:

Analytics Kit will be shipped to you and you can learn and build using tutorials. You can start for free today!

1. Data Analytics using R

Project Description

This project is predicated on data analysis within which we try to analyze an outsized set of knowledge not able of being analyzed by typical database or data analysis software like Excel. Patterns involved in a severe accident can be detected if we develop suitable prediction models capable of automatic classification of a variety of injury severity of assorted traffic accidents. To beat this, we try to implement distributed processing using the Hadoop and acquire the required result with Apache Zeppelin to research and visualize the information set and generate a choice tree. This project is helpful for several purposes such as:

• Identify the essential nature of accidents happening within the selected highway & roads.
• Identify the most explanation for accidents supported the collected data.
• Identify the features of the road to blame for the accident.
• Identify road intersection types and also the frequency of accidents.
• Run instructional recommendation system.

Modules used in this project

LOGIN

• Login cardinals
• Starting Hadoop distributed file system

PRE-PROCESSING DATA

• Converting Unstructured data to Structured data.
• Data cleaning
• Removing Missing Values & noise.
• Removing duplicate records.
• Integration of knowledge sets.
• Fragmentation and replication for Hadoop.

CLUSTERING MODULE

• Fitting Master and Slave nodes.
• Processing MapReduce jobs in a very parallel environment.
• Fetching MapReduce Gain output to Zeppelin for tree induction.

ATTRIBUTE SELECTION AND TREE INDUCTION

• An information Mining functionality for generating Decision Tree.
• Preparing Training Data set.
• Preparing Validation Data set.

VISUALIZATION WITH APACHE zeppelin

• Data Visualization.

Project Implementation

This framework has many steps to urge the finalized analysis of knowledge.

they're as follows:

Hadoop login: The Apache Hadoop may be a collection of open-source software that facilitate a framework for distributed storage and processing of enormous data sets using the MapReduce programming model.

• During this framework, firstly you've got to log in during this software together with your own login cardinals.
• After this, the information set you'll enter get Converted into

Structured data for Pre-Processing.

• After the information gets structured and free from all the errors, it'll transfer to the cluster module for the task of grouping a collection of objects.
• Next step is that the Decision Tree Visualization, it performs the foundations for the prediction of the target variable. With the assistance of this algorithm, the critical distribution of the information is definitely understandable.
• Lastly, the information gets visualize for the output in Apache Zeppelin for Severity Vs User Defined Attribute.

HARDWARE REQUIREMENTS

• Processor – i3
• Hard Disk – 1 TB
• Memory – 4 GB RAM

SOFTWARE REQUIREMENTS

• Windows 8, Windows 10 (ultimate, enterprise)
• Front End: Apache Zeppelin
• Back End: Scala-Spark

Advantages:

• This given framework overcomes the above-mentioned issues in an efficient way.

• Protection is developed within the locations which are more at risk of accidents found from the analysis of the information.

Disadvantages:

• If more features were established then more information is identified that's related to an accident.
• It has memory restrictions and variable data, it only gives a precise level of knowledge.

Did you know

Skyfi Labs helps students learn practical skills by building real-world projects.

You can enrol with friends and receive kits at your doorstep

You can learn from experts, build working projects, showcase skills to the world and grab the best jobs.
Get started today!

• No kit required

• Data mining

System for analysis of Road Accidents