Data Analysis of Driving Events on Processing Sensor Data through Hadoop Ecosystem

Author Name(s): Pooja M. Gupta, Shrikant D. Zade
Author Email: 910poojagupta@gmail.com

Abstract

Moreover, Insurance Company decided the premium according to the risk of accidental ratio, average speed, driving patterns of owner of a vehicle. Minimum speed of the vehicle should be 30-50, onwards it considered as more chances of a accidental stage. In past papers the hardware were used to detect the speed and driving patterns of a vehicle by GSM/GPS modem. SD card consist of each data record within a certain limitation. But there is no certain analysis on this input files generated from the hardware used. Hence in this paper the main task is to process input data file in hadoop ecosystem for analysis. This will speed up the process of tracking the vehicle, mapping and calculating the mileage and various operations of vehicle. This application consume large amount of data and this data should not be loss and it should be manageable with historical and current data for this hadoop is the best feature for better analysis and fault tolerant. Other than this we can process number of applications for processing through hadoop ecosystem. But here we have taken vehicular data which is used to analyze the each drivers of vehicle in the form of speed, date and time. It is also used to derive driving patterns of the vehicle and each driver and also help to manage the speed of a vehicle and to analyze the accidental ratio.

Keywords

Hadoop Ecosystem, SD Card, GPS/GSM Data File.

Introduction

Big Data

Big Data is used for collections of extremely large dataset which increases day by day which may be analysed computationally to find the patterns, different association-relationship, trends mostly related to human interactions and behaviour. It is difficult to process Big Data using traditional data management systems[1].Therefore, Hadoop is a framework introduced by Apache Software Foundation to solve Big Data management and processing challenges.

  1. Year over year, the large number of client and variety of devices increasing to serve the facilities.
  2. The amount of data and having different variety collected continuously due to expand use of social, mobile, and embedded technologies.
  3. In a competitive marketplace the need of large amount of data for deep understanding of business to maintain stability[2].

Conclusion

To get familiar with end to end implantation of a big data project using various tools in Hadoop ecosystem. The company wants to use this data to better understand risk. It can used in a several business objective to minimize the time. It analyses the data in large amount with specific results and to implement machine learning and implementation of several frameworks and projects for enhancing the better results in shorter period of time. The data can be analyzed in the form of pie charts, graph, bar graphs.

823 total views, no views today

Download PDF File

About the author: admin