Page 179 - Big Data Analytics for Connected Vehicles and Smart Cities
P. 179

160	       Big	Data	Analytics	for	Connected	Vehicles	and	Smart	Cities	                	                       Transportation Use Cases	                 161


               • Analytics to be used: A list of the proposed analytics that will form the
                basis of the output of the analysis work. Here again, this is an initial list
                of analytics to be used and will be supplemented by additional analytics
                that will be discovered during the work. The analytics can be supported
                by several different analytics techniques, including the following:

                 • Graph analytics: Initial relationships between data elements and peo-
                  ple; can also show the strength of the relationship based on data at-
                  tributes.
                 • Text analytics: These can uncover underlying sentiment within so-
                  cial media, and compliance are infractions and communications and
                  documents of all kinds. They can also be word cloud visualizations as
                  used in this book.
                 • Path pattern and time series analytics: These provide insight on inter-
                  action patterns between people, products, or data elements.
                 • Structured query language (SQL): This is a standardized query lan-
                  guage for requesting information from a database. It provides flexible
                  ways to manipulate data and to make queries from a big data set using
                  the language of business tools.
                 • Statistical  modeling:  This  includes  statistical  modeling  techniques
                  such as linear least squares regression, nonlinear least squares regres-
                  sion, weighted least squares regression, and locally weighted scatter-
                  plot smoother (curve fitting).
                 • Machine learning: Techniques to sift through data with minimal hu-
                  man input to gain new insights previously undetected. This can form
                  the basis for decision support and automation.


               This format is an approach adopted by a major big data and analytics
          practitioners and solution providers [2] with many years of experience in de-
          veloping and implementing use case descriptions associated with big data and
          analytics projects.
               The purpose of Appendix A is twofold: to explain a few transportation use
          cases and to illustrate how the use cases are put together in practice. The inten-
          tion is to provide practical examples of use cases that can be applied to smart
          city transportation initiatives. These can be used as a starting point for a more
          complete set and as a model on how to create a practical use case template.


          8.6  Summary

          The use case is a very important tool to gauge the implementation of big data
          and analytics techniques regarding smart city transportation. This chapter ex-
   174   175   176   177   178   179   180   181   182   183   184