International Journal of Computer Applications (0975 – 8887) Volume 131 – No.12, December2015 25 Diwasi: Intelligent GPS based Accident Prevention System V.Y. Attigala Sri Lanka Institute of Information Technology, Sri Lanka D.H. Hettiarachchi Sri Lanka Institute of Information Technology,Sri Lanka R.S.D.S. Jayasekara Sri Lanka Institute of Information Technology,Sri Lanka V.N. Vithana Sri Lanka Institute of Information Technology,Sri Lanka ABSTRACT In a demanding environment, accidents are unavoidable disasters. This has caused accidents to fall under the top 10 of the world’s death rate. When considering Sri Lanka the accidents rate are in vast amount compared to world. The majority of accidents are caused by unawareness of critical zones rather than human errors. The research is focused on finding patterns within each accident with the aid of data mining techniques. The main objective was to minimize accidents caused by unawareness of the drivers. When considering the accident management systems available in the market, most of them are automobile integrated systems which are quite expensive and can hardly be affordable to ordinary people. In order to prevent this, the research team has created an accident management mobile app which is not integrated to the automobile and inexpensive. The application is built on a Global Position System (GPS) and is based on mobile phone platform considering the cost and the modern trend of using smart phones all over the world. The Data mining model which was built by the Research team, are effective mostly on B Conditions mid-range (20km-50km) roads in Sri Lankan Road conditions. This research on the accident management system mainly focus researchers to study further concerning vehicle independent systems and involve in implementing such inventive systems. Usage of the system will have an impact on considerable reduction of road accidents occurrences, due to the predictions of probable accidents that can happen in critical areas, which previously the driver was unaware of. General Terms Data Mining, GPS, Accident, Patterns, Intelligence, Microsoft Clustering Algorithm, Microsoft Decision Tree Algorithm, CART Method Keywords Diwasi, Mobile Based Applications, Vehicle Independent, Real time Data mining, 297B Accident Report Application. 1. INTRODUCTION Data mining and warehousing is a concept use to discover patterns and trends in a large data base such as a data warehouse [1]. The research filed area covers the above mention technology. The main approach is based on young and fast-growing field of data mining. The system analyses the accident data recorded in Sri Lanka police central database and identify accident pattern, and the system is intelligent for predicting situation, when vehicle reaches the same situation which is as same as the situation predicted by the system, it will be warned by pushing alerts. The development team decides to contribute Global Position System (GPS) based mobile phone platform considering the cost and the modern trend of using smart phone all over the world. Accidents are unavoidable incidents which contain many causes. The following statistical data represent how it will influence the world community. These statistical data are taken from World Health Organization [2]. Table 1. Accident Statistical Data Country Year Australia 2010 Pakistan 2010 India 2010 United State of America 2010 Sri Lanka 2010 When consider about Sri Lanka, as a unit cell of the world, also faces a major influence of accidents by considering the following statistical data. The following static, Ministry of Internal Transport represent the criticalness of the situation [3]. Table 2. Analysis Of Data (Year 2013 -2014) Overview Total % Deaths 24001 6 Fatal Accidents 22368 6 Minor Accidents 127466 34 Critical Accidents 56387 14 Damage Only 173807 45 Considering the above factors, these statistical data represent a criticalness in an accidents and how accidents affect the society. Further studding about road accident the research team has undergone through related researches and the authors of the article has represent that road accidents can be categories into four groups by their nature [4]. Equipment Failure Roadway Design Poor Maintenance Driver Behaviors From the above categories Texas A&M Transportation University has expressed that road design play the major role in accidents [5]. The research team expect to research in the area which is highlighted by Texas A&M Transportation University.