MakeML : automated machine learning from data to predictions
Author(s)
Tromba, Isabella M
DownloadFull printable version (1.678Mb)
Alternative title
Automated machine learning from data to predictions
Other Contributors
Massachusetts Institute of Technology. Department of Electrical Engineering and Computer Science.
Advisor
Sam Madden.
Terms of use
Metadata
Show full item recordAbstract
MakeML is a software system that enables knowledge workers with no programming experience to easily and quickly create machine learning models that have competitive performance with models hand-built by trained data scientists. MakeML consists of a web-based application similar to a spreadsheet in which users select features and choose a target column to predict. MakeML then automates the process of feature engineering, model selection, training, and hyperparameter optimization. After training, the user can evaluate the performance of the model and can make predictions on new data using the web interface. We show that a model generated automatically using MakeML is able to achieve accuracy better than 90% of submissions for the Titanic problem on the public data science platform Kaggle.
Description
Thesis: M. Eng., Massachusetts Institute of Technology, Department of Electrical Engineering and Computer Science, 2018. This electronic version was submitted by the student author. The certified thesis is available in the Institute Archives and Special Collections. Cataloged from student-submitted PDF version of thesis. Includes bibliographical references (pages 61-64).
Date issued
2018Department
Massachusetts Institute of Technology. Department of Electrical Engineering and Computer SciencePublisher
Massachusetts Institute of Technology
Keywords
Electrical Engineering and Computer Science.