Decision Support System for the Analysis of Traffic Accident Big Data

Abstract

Today, as traffic accidents continue to occur, road safety has become a major concern of contemporary social problems. Many factors explain traffic accidents, such as the nature of the accident site, its environment, time, driver behavior, weather, and other uncertain complex factors. As a result, the occurrence of road accidents is non-linear, so it is necessary to explore the correlation between data from many different aspects to minimize the risk. After pre-processing the data and a classification using the data mining tools, relevant information can be deduced on the causes of the high-frequency accidents. Depending on the actual results obtained, we can verify the accuracy of the extracted information, and this can help predict new situations with similar data in the future. The ultimate aim of data analysis is to choose the most accurate extraction process after validation, by analyzing the characteristics of the data and their relationship to the analysis and extraction process. In this paper, we propose a decision-making system for the analysis of traffic accident data in order to extract information relevant to the prevention of road risk. In addition, we introduce the concept of crowdsourcing in the collection phase to gather more information from the users involved in an accident. The proposed system is based on appropriate data mining and Big Data techniques for the extraction, pre-processing and exploration of accident data, entirely managed by the Hadoop framework, in order to categorize road accidents and detect problematic sites.