A visualisation technique of extracting hidden patterns for maintaining road safety. (17th February 2022)

• Record Type: Journal Article
• Authors: Selvarathi, C.; Subha, S.; Raja, G. Madasamy; Lakshmi, K. Vidhya
• Is Part Of: International journal of advanced intelligence paradigms. Volume 21:Number 1/2(2022)
• Page Start: 100
• Page End: 108
• Publication Date: 2022-02-17
• Subjects: predictive analytics -- road accident data mining -- machine learning techniques -- Maptive tool -- random-forest method -- data visualisation; Artificial intelligence -- Periodicals; Machine theory -- Periodicals; Fuzzy logic -- Periodicals; 006.305
• Languages: English
• ISSNs: 1755-0386

Abstract

Predictive analysis is a technique in the data mining which deals with extraction of hidden information from the collection of data and then it is used to predict trends and behaviour patterns. Predictive analysis extracts the relationship between explanatory variables and predicted variables from the past occurrences and exploits those variables to predict unknown outcome. This research paper is designed to enable traffic management controllers to use historical traffic and road accident data, to observe how traffic congestion and accidents are created in the roadways. The information about the accidents is collected and analysed through data mining techniques to predict the traffic congestion and accidents. The main contribution of this paper is a visualisation technique to predict the heavy traffic and highlight the history of the accidents. In this paper we introduce a risk index value that represents the history of accidents and predicts the accident prone areas.