Authors:
Kaan Koseler
;
Kelsea McGraw
and
Matthew Stephan
Affiliation:
Dept. of Computer Science and Software Engineering, Miami University, 510 East High Street, Oxford Ohio and U.S.A.
Keyword(s):
Machine Learning, Domain Specific Modeling Language, Baseball Analytics, Binary Classification, Supervised Learning, Model Driven Engineering.
Related
Ontology
Subjects/Areas/Topics:
Domain-Specific Modeling and Domain-Specific Languages
;
Languages, Tools and Architectures
;
Model-Driven Software Development
;
Software Engineering
Abstract:
Accompanying the Big Data (BD) paradigm is a resurgence in machine learning (ML). Using ML techniques to work with BD is a complex task, requiring specialized knowledge of the problem space, domain specific concepts, and appropriate ML approaches. However, specialists who possess that knowledge and programming ability are difficult to find and expensive to train. Model-Driven Engineering (MDE) allows developers to implement quality software through modeling using high-level domain specific concepts. In this research, we attempt to fill the gap between MDE and the industrial need for development of ML software by demonstrating the plausibility of applying MDE to BD. Specifically, we apply MDE to the setting of the thriving industry of professional baseball analytics. Our case study involves developing an MDE solution for the binary classification problem of predicting if a baseball pitch will be a fastball. We employ and refine an existing, but untested, ML Domain-Specific Modeling La
nguage (DSML); devise model instances representing prediction features; create a code generation scheme; and evaluate our solution. We show our MDE solution is comparable to the one developed through traditional programming, distribute all our artifacts for public use and extension, and discuss the impact of our work and lessons we learned.
(More)