M
Michael R. Smith
Researcher at Sandia National Laboratories
Publications - 42
Citations - 717
Michael R. Smith is an academic researcher from Sandia National Laboratories. The author has contributed to research in topics: Instance-based learning & Semi-supervised learning. The author has an hindex of 10, co-authored 42 publications receiving 533 citations. Previous affiliations of Michael R. Smith include Brigham Young University.
Papers
More filters
Journal ArticleDOI
An instance level analysis of data complexity
TL;DR: This paper identifies instances that are hard to classify correctly (instance hardness) by classifying over 190,000 instances from 64 data sets with 9 learning algorithms and finds that class overlap is a principal contributor to instance hardness.
Proceedings ArticleDOI
Improving classification accuracy by identifying and removing instances that should be misclassified
Michael R. Smith,Tony Martinez +1 more
TL;DR: A filtering method called PRISM is introduced that identifies and removes instances that should be misclassified and achieves a higher classification accuracy than the outlier detection methods and compares favorably with the noise reduction method.
Nehovah: A Neologism Creator Nomen Ipsum.
TL;DR: A system called Nehovah that generates neologisms from a set of base words provided by a user by evaluating various attributes of a neologism such as how well it communicates the source concepts and how “catchy” it is.
Posted Content
Recommending Learning Algorithms and Their Associated Hyperparameters
TL;DR: This paper applied collaborative filtering techniques to learning algorithm and hyperparameter selection, and find that doing so avoids determining which meta-features to use and outperforms traditional meta-learning approaches in many cases.
Proceedings ArticleDOI
Mind the Gap: On Bridging the Semantic Gap between Machine Learning and Malware Analysis
Michael R. Smith,Nicholas T. Johnson,Joe Ingram,Armida J. Carbajal,Bridget I. Haus,Eva Domschot,Ramyaa Ramyaa,Christopher C. Lamb,Stephen J. Verzi,W. Philip Kegelmeyer +9 more
TL;DR: It is hypothesized that the full potential of ML is not realized in malware analysis (MA) due to a semantic gap between the ML and MA communities, and existing open-source malware datasets used in ML are reviewed to find a lack of behavioral information that could facilitate stronger impact by ML in MA.