Computer Sciences Commons^™

Simplification Of Eeg Signal Extraction, Processing, And Classification Using A Consumer-Grade Headset To Facilitate Student Engagement In Bci Research, Jesus D. Rodriguez

Theses and Dissertations

Brain-computer interfaces (BCIs) are an emerging technology that leverage neurophysiological signals as input to computing systems. By circumventing the reliance on traditional input methods (e.g., mouse and keyboard), BCIs show a promising alternative interaction modality for people with disabilities. Advances in BCI research have further inspired a range of novel applications, such as the use of neurophysiological signals as passive input (e.g., to detect and reduce operator workload when managing multiple machines). BCIs have also emerged as a tool for student engagement due to the intrinsic interdisciplinarity of the technology, which spans the fields of computer science, electrical engineering, neuroscience, …

Approximate Set Union Via Approximate Randomization, Pengfei Gu

Theses and Dissertations

We develop an randomized approximation algorithm for the size of set union problem |A1 U A2 U...UAm|, which given a list of sets A1,...,Am with approximate set size m i for Ai with mi ∈ ((1–βL)|A i|,(1+βR)|Ai|), and biased random generators with Prob(x = RandomElement(Ai)) ∈ [1–a L/Ai, 1 +aR/Ai] for each input set Ai and element x ∈ Ai, where i = 1,2,...,m. The approximation |Ai | |Ai | ratio for |A1 U A2 U...UAm| is in the range [(1–ϵ)(1–aL)(1–βL),(1+ϵ)(1+β R)(1+βR)] for any ϵ ∈ (0,1), where α L,αR,βL,βR ∈ (0,1). The complexity of the algorithm …

Mapreduce And Heterogeneity: Power-Aware Bag-Of-Tasks, Framework Parameter Sensitivity, And Dynamic Cluster Aware Framework Configuration, Jessica L. Hartog

Graduate Dissertations and Theses

This dissertation presents the techniques for adaptation of MapReduce frameworks to incorporate heterogeneity-aware scheduling algorithms, an inspection of cluster configurations and how they impact these scheduling algorithms, an analysis regarding how the cluster configuration and the heterogeneity-aware scheduling can work together to minimize turnaround time and/or power consumption of the cluster when executing MapReduce applications, and how these lessons can be applied more broadly to Big Data infrastructure outside of MapReduce that supports multiple Big Data frameworks simultaneously.

Heterogeneity exists in various capacities in any given cluster, from static (Physical and Platform) heterogeneity to dynamic heterogeneity (Transient Data, Transient Applications, …