Many of us here at Agrible have degrees from the University of Illinois at Urbana-Champaign and realize the value that existing near a world class university creates. Agrible makes an active effort to establish and maintain relationships with both graduate and undergraduate students on campus. One such student shared his experiences of working alongside Agrible. Christopher Urbanec is a senior undergraduate student majoring in Technical Systems Management in the college of ACES.
“My interests in understanding Hadoop, Big Data, and data analysis led me to Agrible. I presented past work that I had done, and as a result, I was then given the opportunity to analyze some real world rainfall data of a collection of Tanzanian municipalities. My goal was to aggregate flat files into a single object and analyze them in a distributed way. My tools included IPython Notebooks and Python data analysis libraries such as Pandas and Numpy. I started analysis by using data smoothing techniques, as well as finding consecutive days with rainfall above a certain threshold. This analysis is useful to users in the Agricultural Industry because it provides a means to realize important cyclical patterns in rainfall ,as well as pick out time periods when high/low rainfall quantities may be an advantage. An important takeaway from working with the data provided to me is that the majority of time spent working with real world data revolves around converting data into a more convenient format to work with. And I was not even given truly “raw” data!
The opportunity to work alongside professionals that do programming and scientific work for a living has been an invaluable and pleasurable experience. I was taken aback by how willing these guys were to share their knowledge and provide advice along the way. It is really helpful to learn what tools are used to accomplish real world data analysis tasks, and then be able to get advice on how to best use them. I found that the most important dynamic in the student to professional relationship is the willingness to learn new things and to be curious about the relationship between the science and technology that drives Agrible.
Goals for the future are to be able to be able to better understand how to utilize multiple cores on machines in order to efficiently crunch data on large datasets using intensive models. A possible solution is to use the Hadoop software framework in order to pool together multiple computing nodes.
Many thanks to the members of Agrible for the outreach to students such as myself. You guys are the best! ”