5063 research outputs found
Sort by
How negative sampling provides class balance to rare event case data using a vehicular accident prediction project as a use case scenario
Get PDFRare event case data occur at such an infrequent rate that even having high amounts of it can leave researchers starving for more information. There has always existed a tug and pull relationship among rare event case data, where a higher count of entries often leads to a lack of explanatory variables, and vice versa. In the research spectrum of rare event case probability prediction, several methods of data sampling exist to remedy the main issue of rare event case data: a lack of data to collect and learn from. The most effective methods often involve altering the distribution of the training samples in a data set. The least utilized of these methods is negative sampling, where positive entries in a data set are used to generate negative entries. To outline the utility of negative sampling, this work discusses the application of five types of negative sampling on a vehicular accident prediction project, where non-accident records are generated through manipulating the temporal and spatial attributes of existing accident records. Moreover, different methods of data manipulation, including feature selection and different negative to positive data ratios, are used to explore what types of explanatory variables are most important when predicting vehicular accidents. Additionally, two types of predictive models, a Multilayer Perceptron and a Logistic Regression model, are created and directly compared in terms of predictive capability. Ultimately, the best model for predictive performance is heavily dependent on the specific implementation and desired results
Detecting and Identifying Single Event Transients using IRES and Machine Learning
Get PDFSETs are analyzed and characterized through ionizing radiation effects spectroscopy (IRES) and machine learning. Potentially catastrophic radiation-induced errors can be exposed with IRES as it simplifies the identification of transients through statistical analysis of waveform behavior, allowing for the capture of subtle changes in circuit dynamics. Leveraging a k-Nearest Neighbors (KNN) machine learning algorithm with IRES data, the identification of transients is facilitated and makes an on-chip implementation feasible
Analysis of the soil lead contamination issue of south Chattanooga
No full textThis project examines the South Chattanooga soil lead contamination and the factors affecting awareness of this situation. The fallacies of conducting Community-Based Participatory Research (CBPR) in target neighborhoods are outlined
Buterin\u27s scalability trilemma viewed through a state-change-based classification for common consensus algorithms
No full textWe classify common consensus algorithms based on how they decide the order of system state changes. We then determine the extent to which each category prioritizes scalability, decentralization, and security
Global warming: modeling surface structures for the capture of carbon dioxide
No full textBurning fossil fuels releases carbon dioxide to the atmosphere and contributes to global warming. Using molecular modeling, our goal is to create simulated surface structures that can preferentially trap carbon dioxide from combustion exhaust gases
Modeling abatement of malodorous molecules by adsorption on carbon surfaces
No full textTo reduce pathogenic diseases, toilet systems disinfecting human waste are being developed by the Gates Foundation. Our goal is to model the interaction of malodorous fecal compounds with porous carbon structures that can trap these offensive odors
What\u27s in a name? an assessment of sexism in the name of climbing routes
No full textBy looking at sexist attitudes in the names of climbing routes, this research explores how gender bias affects women in the realm of outdoor leadership. Few research has looked at how these constructs create a barrier for women in the outdoor field
The Chattanooga drag herstory project
No full textResearch through oral interviews on the history of drag in the Chattanooga community in hopes of establishing a starting point for the LGBTQ+ history, archival materials, and ephemera
Analyzing the effects of big data on medical jobs: a systematic literature review
No full textA PRISMA style systematic review of modern literature pertaining to Big Data technologies in medicine specifically aimed at analyzing the likely effects of the technology on the medical job market in Chattanooga
