Abstract: Nowadays, big data are everywhere because huge amounts of valuable data can be easily generated and collected from a wide variety of data sources at a rapid rate. Embedded into these big data are implicit, previously unknown and potentially useful information and valuable knowledge that can be discovered by data science. Due to their value, these big data are often considered as the new oil. In recent years, many governments make their collected big data freely available to their citizens, who could then gain some insights about services available in the city from these open data. In this paper, we present a cognitive and predictive analytic approach to analyze open data for discovering interesting patterns such as tipping patterns. In general, tipping is a voluntary action conceived as social norm that is valuable to service workers in many countries. With the introduction of ride hailing services, traditional taxi services have began facing increased competition. As such, there are increasing interests in factors that are associated with generous tips. Hence, to evaluate the practicality of our approach, we conduct a case study on applying our approach to transportation data (e.g., taxi trip records) from New York City (NYC) to examine and predict tip generosity. Although we conducted the case study on NYC data, our presented approach is expected to be applicable to perform cognitive and predictive analytics on big open data from other cities.
Authors: Carson K. Leung (University of Manitoba, Canada)