International Scholar Visits UO, Teaches Class, and Publishes with Students - All During Covid

Juan Flores and Students

The photo shows, clockwise from top left, Dr. Juan Flores, and students Stephanie Schofield, Joseph McLaughlin, Alec Springel, Kyra Novitzky, and Seth Tal. (Student Ronny Fuentes is not shown because he is already at his summer internship.)

In early 2021, a team of UO CIS students came together to develop a Python library to support machine learning and data science. The students were presented with a unique challenge in the CIS 422 Software Methodologies class, which is taken by all CIS majors, and in Winter, 2021, was taught by visiting scholar Dr. Juan Flores, from the University of Michoacan, Mexico.

The students not only wrote the software library, but also wrote a paper describing the library and its scientific value. The paper was accepted for publication, and will be presented in July, 2021, at the International Conference on Time Series and Forecasting. The article is entitled “Pipeline Trees - An Auxiliary Tool in the Creation of Time Series Pipelines.”

The challenge presented to the students in the class was to address a problem that occurs when applying machine learning and data science to time-series forecasting. Such forecasting is used to predict future outcomes, such as to predict solar irradiance and ambient temperature, which in turn permits engineers to predict the energy that will be generated by solar panels, and the energy that will need to be generated by other means.

The challenge was to develop a software library that could be used to explore the “pipeline space” of time-series forecast modeling. That is to say, to explore all possibilities of the processing steps that could be taken before launching a forecasting model. Several software libraries already exist to support the execution of pipeline-processing steps once the steps are decided, but few or no libraries to support the exploration of possible steps. This library-development challenge was built into the project that Dr. Flores crafted for his class. The project captures the research-informed teaching provided to UO CIS students.

Congratulations to the student team for this great accomplishment as you start your careers in computer science!