Open Access. Powered by Scholars. Published by Universities.®
Articles 1 - 3 of 3
Full-Text Articles in Engineering
The Effects Of Mental Training On Brain Computer Interface Performance With Distractions, John Larocco
The Effects Of Mental Training On Brain Computer Interface Performance With Distractions, John Larocco
Theses and Dissertations
The overall success of a brain computer interface (BCI) is largely dependent on the features used to make decisions. Noise in the electroencephalography (EEG) increases the difficulty of acquiring meaningful features. Previous literature suggests teaching subjects meditation and relaxation techniques may improve features relevant to BCI operation. The purpose of this study was to investigate performance on several cognitive protocols for both individuals who use meditation techniques and those who do not use these techniques. Both groups were given a motor imagery based BCI protocol, a P300 speller BCI, a verbal learning task, and an N-back test. No significant difference …
Achieving Grid Parity For Large Scale Photovoltaic Systems In New Jersey, Ulrich Schwabe
Achieving Grid Parity For Large Scale Photovoltaic Systems In New Jersey, Ulrich Schwabe
Theses and Dissertations
With photovoltaic (PV) power systems becoming ever more prevalent in today's world, it is an inevitability that this renewable energy technology becomes more competitive from a price standpoint. Explored in this thesis are several engineering optimization and cost reduction methods that will enable large scale PV system costs to achieve grid parity in the next few years without requiring government subsidies or market support techniques. This research included actual data from numerous real world, large scale photovoltaic projects the author engineered in the northeastern United States and is informed by those design optimizations, best practice guidelines for proper photovoltaic system …
Incremental Learning Of Concept Drift From Imbalanced Data, Gregory Ditzler
Incremental Learning Of Concept Drift From Imbalanced Data, Gregory Ditzler
Theses and Dissertations
Learning data sampled from a nonstationary distribution has been shown to be a very challenging problem in machine learning, because the joint probability distribution between the data and classes evolve over time. Thus learners must adapt their knowledge base, including their structure or parameters, to remain as strong predictors. This phenomenon of learning from an evolving data source is akin to learning how to play a game while the rules of the game are changed, and it is traditionally referred to as learning concept drift. Climate data, financial data, epidemiological data, spam detection are examples of applications that give rise …