Yan Jia

This is a part of analytical summary paper of EAP507. To complete this homework, we need to choose one peer reviewed paper in big data field, then annotate 6 rhetorical elements, micro and macro the paper for deeply understanding the information. When we write the analytical summary, we need to present the key information in own words, such as objects, new offerings, methodologies. Besides, we need to think and analyze how this paper uses what kind of language patterns, it helps me know how to use language patterns in our filed. The whole process teaches me how to understand the stance that authors forward in the paper, and what to focus when reading a paper.

 An Analytical Summary of Kumar & Toshniwal (2016)

Road and traffic accident are an important concern in the world. To promote the transportation safety, traditional statistical methods and data mining methods are used in many research on road accident data analysis. Besides, all previous studies and research were limited focused on analyzing road accident data and identifying the factors that influence severity of road accident. Authors tried to fill this gap, they proposed a method that by using trend analysis, and time series data, to analyze the road accident data. The results show the method used in the study, can be used and performance a better understanding of certain location road accidents that in a certain time.

In this research, Kumar and Toshniwal try to use hierarchical clustering and cophenetic correlation coefficient(CPCC) to do predictive analysis of hourly road accident counts. The data set used in this research “consists of road accidents count of 26 districts of Gujrat state from January 2010 to December 2014” (P.6). The result points out that 8:00PM is the most dangerous time in most districts. The relevance and new offering for the same topic is this paper gives an clustering approach that can group the different districts with similar road accident patterns into different clusters efficiently. The method that used hierarchical clustering and CPCC can help to oncoming research on road accident data analysis. For example, revealing more hidden information that help extract valuable relationships between the variables of sample data.