A Paradigmatic Approach to Exploratory Data Analysis Utilising New York’s Road Traffic to Derive Coherent Inferences

Exploratory Data Analysis is a systematic method of analyzing the data from multiple perspectives [1]. From graphical representations, one may be able to draw useful and accurate inferences. This is the most crucial part of any data analysis and computation, as the analyst gains greater insights into the data that is being handled [2] and may hence avoid errors like the misjudgment of correlation, among other misconceptions that the data may lead the analyst into, owing to certain anomalies such as hidden outliers, that may not have been detected during the data cleaning process. The paper intends to emphasis on the importance and relevance of Exploratory Data Analysis (EDA), which would give accurate results, on proper conduction. It hence proposes a simple yet powerful paradigm to perform Exploratory Data Analysis on New York's road-traffic, to derive various insights on the work culture, migration patterns, etc. of the locality. The paradigm proposed involves four major steps, which includes the process of choosing the right dataset, cleaning it, conducting an exploratory analysis on the data and drawing inferences from the exploratory representations. The paper delves into each step of this paradigm, illustrating them with New York's road-traffic case study. On analyzing the traffic in New York using this paradigm, various inferences, like the pattern of traffic on different roads, could be drawn which was further verified with actual facts and figures. Furthermore, it could also draw larger insights about the city. For instance, the exploration could demonstrate that the city of New York has a highly work-oriented population, among many other inferences drawn. Hence, the approach proposed by the paper can have significant value addition to any field, by emphasizing on the process of data processing and analysis.