Matching Items (149)
Description
This thesis describes a synthetic task environment, CyberCog, created for the purposes of 1) understanding and measuring individual and team situation awareness in the context of a cyber security defense task and 2) providing a context for evaluating algorithms, visualizations, and other interventions that are intended to improve cyber situation awareness. CyberCog provides an interactive environment for conducting human-in-loop experiments in which the participants of the experiment perform the tasks of a cyber security defense analyst in response to a cyber-attack scenario. CyberCog generates the necessary performance measures and interaction logs needed for measuring individual and team cyber situation awareness. Moreover, the CyberCog environment provides good experimental control for conducting effective situation awareness studies while retaining realism in the scenario and in the tasks performed.
ContributorsRajivan, Prashanth (Author) / Femiani, John (Thesis advisor) / Cooke, Nancy J. (Thesis advisor) / Lindquist, Timothy (Committee member) / Gary, Kevin (Committee member) / Arizona State University (Publisher)
Created2011
Description
Intuitive decision making refers to decision making based on situational pattern recognition, which happens without deliberation. It is a fast and effortless process that occurs without complete awareness. Moreover, it is believed that implicit learning is one means by which a foundation for intuitive decision making is developed. Accordingly, the present study investigated several factors that affect implicit learning and the development of intuitive decision making in a simulated real-world environment: (1) simple versus complex situational patterns; (2) the diversity of the patterns to which an individual is exposed; (3) the underlying mechanisms. The results showed that simple patterns led to higher levels of implicit learning and intuitive decision-making accuracy than complex patterns; increased diversity enhanced implicit learning and intuitive decision-making accuracy; and an embodied mechanism, labeling, contributes to the development of intuitive decision making in a simulated real-world environment. The results suggest that simulated real-world environments can provide the basis for training intuitive decision making, that diversity is influential in the process of training intuitive decision making, and that labeling contributes to the development of intuitive decision making. These results are interpreted in the context of applied situations such as military applications involving remotely piloted aircraft.
ContributorsCovas-Smith, Christine Marie (Author) / Cooke, Nancy J. (Thesis advisor) / Patterson, Robert (Committee member) / Glenberg, Arthur (Committee member) / Homa, Donald (Committee member) / Arizona State University (Publisher)
Created2011
Description
Advancements in computer vision and machine learning have added a new dimension to remote sensing applications with the aid of imagery analysis techniques. Applications such as autonomous navigation and terrain classification which make use of image classification techniques are challenging problems and research is still being carried out to find better solutions. In this thesis, a novel method is proposed which uses image registration techniques to provide better image classification. This method reduces the error rate of classification by performing image registration of the images with the previously obtained images before performing classification. The motivation behind this is the fact that images that are obtained in the same region which need to be classified will not differ significantly in characteristics. Hence, registration will provide an image that matches closer to the previously obtained image, thus providing better classification. To illustrate that the proposed method works, naïve Bayes and iterative closest point (ICP) algorithms are used for the image classification and registration stages respectively. This implementation was tested extensively in simulation using synthetic images and using a real life data set called the Defense Advanced Research Project Agency (DARPA) Learning Applied to Ground Robots (LAGR) dataset. The results show that the ICP algorithm does help in better classification with Naïve Bayes by reducing the error rate by an average of about 10% in the synthetic data and by about 7% on the actual datasets used.
ContributorsMuralidhar, Ashwini (Author) / Saripalli, Srikanth (Thesis advisor) / Papandreou-Suppappola, Antonia (Committee member) / Turaga, Pavan (Committee member) / Arizona State University (Publisher)
Created2011
Description
The objective of this project was to evaluate human factors based cognitive aids on endoscope reprocessing. The project stems from recent failures in reprocessing (cleaning) endoscopes, contributing to the spread of harmful bacterial and viral agents between patients. Three themes were found to represent a majority of problems: 1) lack of visibility (parts and tools were difficult to identify), 2) high memory demands, and 3) insufficient user feedback. In an effort to improve completion rate and eliminate error, cognitive aids were designed utilizing human factors principles that would replace existing manufacturer visual aids. Then, a usability test was conducted, which compared the endoscope reprocessing performance of novices using the standard manufacturer-provided visual aids and the new cognitive aids. Participants successfully completed 87.1% of the reprocessing procedure in the experimental condition with the use of the cognitive aids, compared to 46.3% in the control condition using only existing support materials. Twenty-five of sixty subtasks showed significant improvement in completion rates. When given a cognitive aid designed with human factors principles, participants were able to more successfully complete the reprocessing task. This resulted in an endoscope that was more likely to be safe for patient use.
ContributorsJolly, Jonathan D (Author) / Branaghan, Russell J (Thesis advisor) / Cooke, Nancy J. (Committee member) / Sanchez, Christopher (Committee member) / Arizona State University (Publisher)
Created2011
Description
Sparse learning is a technique in machine learning for feature selection and dimensionality reduction, to find a sparse set of the most relevant features. In any machine learning problem, there is a considerable amount of irrelevant information, and separating relevant information from the irrelevant information has been a topic of focus. In supervised learning like regression, the data consists of many features and only a subset of the features may be responsible for the result. Also, the features might require special structural requirements, which introduces additional complexity for feature selection. The sparse learning package, provides a set of algorithms for learning a sparse set of the most relevant features for both regression and classification problems. Structural dependencies among features which introduce additional requirements are also provided as part of the package. The features may be grouped together, and there may exist hierarchies and over- lapping groups among these, and there may be requirements for selecting the most relevant groups among them. In spite of getting sparse solutions, the solutions are not guaranteed to be robust. For the selection to be robust, there are certain techniques which provide theoretical justification of why certain features are selected. The stability selection, is a method for feature selection which allows the use of existing sparse learning methods to select the stable set of features for a given training sample. This is done by assigning probabilities for the features: by sub-sampling the training data and using a specific sparse learning technique to learn the relevant features, and repeating this a large number of times, and counting the probability as the number of times a feature is selected. Cross-validation which is used to determine the best parameter value over a range of values, further allows to select the best parameter value. This is done by selecting the parameter value which gives the maximum accuracy score. With such a combination of algorithms, with good convergence guarantees, stable feature selection properties and the inclusion of various structural dependencies among features, the sparse learning package will be a powerful tool for machine learning research. Modular structure, C implementation, ATLAS integration for fast linear algebraic subroutines, make it one of the best tool for a large sparse setting. The varied collection of algorithms, support for group sparsity, batch algorithms, are a few of the notable functionality of the SLEP package, and these features can be used in a variety of fields to infer relevant elements. The Alzheimer Disease(AD) is a neurodegenerative disease, which gradually leads to dementia. The SLEP package is used for feature selection for getting the most relevant biomarkers from the available AD dataset, and the results show that, indeed, only a subset of the features are required to gain valuable insights.
ContributorsThulasiram, Ramesh (Author) / Ye, Jieping (Thesis advisor) / Xue, Guoliang (Committee member) / Sen, Arunabha (Committee member) / Arizona State University (Publisher)
Created2011
Description
The wood-framing trade has not sufficiently been investigated to understand the work task sequencing and coordination among crew members. A new mental framework for a performing crew was developed and tested through four case studies. This framework ensured similar team performance as the one provided by task micro-scheduling in planning software. It also allowed evaluation of the effect of individual coordination within the crew on the crew's productivity. Using design information, a list of micro-activities/tasks and their predecessors was automatically generated for each piece of lumber in the four wood frames. The task precedence was generated by applying elementary geometrical and technological reasoning to each frame. Then, the duration of each task was determined based on observations from videotaped activities. Primavera's (P6) resource leveling rules were used to calculate the sequencing of tasks and the minimum duration of the whole activity for various crew sizes. The results showed quick convergence towards the minimum production time and allowed to use information from Building Information Models (BIM) to automatically establish the optimal crew sizes for frames. Late Start (LS) leveling priority rule gave the shortest duration in every case. However, the logic of LS tasks rule is too complex to be conveyed to the framing crew. Therefore, the new mental framework of a well performing framer was developed and tested to ensure high coordination. This mental framework, based on five simple rules, can be easily taught to the crew and ensures a crew productivity congruent with the one provided by the LS logic. The case studies indicate that once the worst framer in the crew surpasses the limit of 11% deviation from applying the said five rules, every additional percent of deviation reduces the productivity of the whole crew by about 4%.
ContributorsMaghiar, Marcel M (Author) / Wiezel, Avi (Thesis advisor) / Mitropoulos, Panagiotis (Committee member) / Cooke, Nancy J. (Committee member) / Arizona State University (Publisher)
Created2011
Description
Multi-task learning (MTL) aims to improve the generalization performance (of the resulting classifiers) by learning multiple related tasks simultaneously. Specifically, MTL exploits the intrinsic task relatedness, based on which the informative domain knowledge from each task can be shared across multiple tasks and thus facilitate the individual task learning. It is particularly desirable to share the domain knowledge (among the tasks) when there are a number of related tasks but only limited training data is available for each task. Modeling the relationship of multiple tasks is critical to the generalization performance of the MTL algorithms. In this dissertation, I propose a series of MTL approaches which assume that multiple tasks are intrinsically related via a shared low-dimensional feature space. The proposed MTL approaches are developed to deal with different scenarios and settings; they are respectively formulated as mathematical optimization problems of minimizing the empirical loss regularized by different structures. For all proposed MTL formulations, I develop the associated optimization algorithms to find their globally optimal solution efficiently. I also conduct theoretical analysis for certain MTL approaches by deriving the globally optimal solution recovery condition and the performance bound. To demonstrate the practical performance, I apply the proposed MTL approaches on different real-world applications: (1) Automated annotation of the Drosophila gene expression pattern images; (2) Categorization of the Yahoo web pages. Our experimental results demonstrate the efficiency and effectiveness of the proposed algorithms.
ContributorsChen, Jianhui (Author) / Ye, Jieping (Thesis advisor) / Kumar, Sudhir (Committee member) / Liu, Huan (Committee member) / Xue, Guoliang (Committee member) / Arizona State University (Publisher)
Created2011
Description
Although there are many forms of organization on the Web, one of the most prominent ways to organize web content and websites are tags. Tags are keywords or terms that are assigned to a specific piece of content in order to help users understand the common relationships between pieces of content. Tags can either be assigned by an algorithm, the author, or the community. These tags can also be organized into tag clouds, which are visual representations of the structure and organization contained implicitly within these tags. Importantly, little is known on how we use these different tagging structures to understand the content and structure of a given site. This project examines 2 different characteristics of tagging structures: font size and spatial orientation. In order to examine how these different characteristics might interact with individual differences in attentional control, a measure of working memory capacity (WMC) was included. The results showed that spatial relationships affect how well users understand the structure of a website. WMC was not shown to have any significant effect; neither was varying the font size. These results should better inform how tags and tag clouds are used on the Web, and also provide an estimation of what properties to include when designing and implementing a tag cloud on a website.
ContributorsBanas, Steven (Author) / Sanchez, Christopher A (Thesis advisor) / Branaghan, Russell (Committee member) / Cooke, Nancy J. (Committee member) / Arizona State University (Publisher)
Created2011
Description
A thorough understanding of Europa's geology through the synergy of science and technology, by combining geologic mapping with autonomous onboard processing methods, enhances the science potential of future outer solar system missions. Mapping outlines the current state of knowledge of Europa's surface and near sub-surface, indicates the prevalence of distinctive geologic features, and enables a uniform perspective of formation mechanisms responsible for generating those features. I have produced a global geologic map of Europa at 1:15 million scale and appraised formation scenarios with respect to conditions necessary to produce observed morphologies and variability of those conditions over Europa's visible geologic history. Mapping identifies areas of interest relevant for autonomous study; it serves as an index for change detection and classification and aids pre-encounter targeting. Therefore, determining the detectability of geophysical activity is essential. Activity is evident by the presence of volcanic plumes or outgassing, disrupted surface morphologies, or changes in morphology, color, temperature, or composition; these characteristics reflect important constraints on the interior dynamics and evolutions of planetary bodies. By adapting machine learning and data mining techniques to signatures of plumes, morphology, and spectra, I have successfully demonstrated autonomous rule-based response and detection, identification, and classification of known events and features on outer planetary bodies using the following methods: 1. Edge-detection, which identifies the planetary horizon and highlights features extending beyond the limb; 2. Spectral matching using a superpixel endmember detection algorithm that identifies mean spectral signatures; and 3. Scale invariant feature transforms combined with supervised classification, which examines brightness gradients throughout an image, highlights extreme gradient regions, and classifies those regions based on a manually selected library of features. I have demonstrated autonomous: detection of volcanic plumes or jets at Io, Enceladus, and several comets, correlation between spectral signatures and morphological appearances of Europa's individual tectonic features, detection of ≤94% of known transient events on multiple planetary bodies, and classification of similar geologic features. Applying these results to conditions expected for Europa enables a prediction of the potential for detection and recommendations for mission concepts to increase the science return and efficiency of future missions to observe Europa.
ContributorsBunte, Melissa K (Author) / Bell, Iii, James F. (Thesis advisor) / Williams, David A. (Committee member) / Saripalli, Srikanth (Committee member) / Clarke, Amanda B. (Committee member) / Reynolds, Stephen J. (Committee member) / Christensen, Phillip R. (Committee member) / Arizona State University (Publisher)
Created2013
Description
With robots being used extensively in various areas, a certain degree of robot autonomy has always been found desirable. In applications like planetary exploration, autonomous path planning and navigation are considered essential. But every now and then, a need to modify the robot's operation arises, a need for a human to provide it some supervisory parameters that modify the degree of autonomy or allocate extra tasks to the robot. In this regard, this thesis presents an approach to include a provision to accept and incorporate such human inputs and modify the navigation functions of the robot accordingly. Concepts such as applying kinematical constraints while planning paths, traversing of unknown areas with an intent of maximizing field of view, performing complex tasks on command etc. have been examined and implemented. The approaches have been tested in Robot Operating System (ROS), using robots such as the iRobot Create, Personal Robotics (PR2) etc. Simulations and experimental demonstrations have proved that this approach is feasible for solving some of the existing problems and that it certainly can pave way to further research for enhancing functionality.
ContributorsVemprala, Sai Hemachandra (Author) / Saripalli, Srikanth (Thesis advisor) / Fainekos, Georgios (Committee member) / Turaga, Pavan (Committee member) / Arizona State University (Publisher)
Created2013